From c6ba77a3275d13f66bee109601147ba8d7ed4aa7 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Przemys=C5=82aw=20Biecek?= <przemyslaw.biecek@gmail.com>
Date: Fri, 13 Oct 2023 00:41:03 +0200
Subject: [PATCH 1/7] open calls

---
 docs/images/positions_mi2.png | Bin 0 -> 51306 bytes
 images/positions_mi2.png      | Bin 0 -> 51306 bytes
 index.Rmd                     |  83 ++++++++++++++++------------------
 3 files changed, 39 insertions(+), 44 deletions(-)
 create mode 100644 docs/images/positions_mi2.png
 create mode 100644 images/positions_mi2.png

diff --git a/docs/images/positions_mi2.png b/docs/images/positions_mi2.png
new file mode 100644
index 0000000000000000000000000000000000000000..42169647399fc68c203671a7b69d113d855c05c1
GIT binary patch
literal 51306
zcmY(r1wfW*(>82^q)3;9(v36{f^-ThDJ?184GKt$G)PHzgMf64lz>Qsv~+j<bFaJK
z^Zx9vuB!`I%sF$^5co_%5(AYO_1d*-7}8Q=%Ga*lDuI7bzk>uneqvf)fuE4{WhKR~
zT_OJZT%R2Y{{_WbO5N_-HGFEsKi8#|skY%iBHK&Liz6?iqT}Bn*ENdazIKh`nzWe6
z3+M6mDHmtWWtZiPt!5c-D()M<8@4j2g2q)fV+z!i*Lh2m+N8(3<|-VLOcNYGNoUos
zRwwmk46RgHaT~9;nNB(@7u!B~9Ahq9RvA3*w0WlR@zx!d(YHr;WkO749#G5TQctBY
zes%Zby@p9fO1gdSEPRySre8`ZdWX^%>+cUeN(@9d{@<_eu*XV!Q~&pScPN)QQA+;z
ztF}a%>oNcSe|IR`@TAtZYeXJCrTFhp1dvb?hd$!u6r;m!(J%cxv|_n6f)mL0#}Ow~
ztYp;uPlbj{q}9FLChY&dejCQX)KqV==K-VV#hT9ij<e+zK?WTuzw<|O;WMUe#Uv_H
z%<csDV5$FIQ$QFqcHlEDt<e50gT=gP-lRTtm%aqPcEwEPQqNYTva&K>*L_0ED$VIX
znqoJY{=2g|ngEZ5MyIdI$z;SNBwgQo9`Mi}WeMw0VEz7?`<#=LlZ~5Of(pBaz8{C2
zyuZw_n~<1zuEh_{SCo>k+Uoks+yD1aCU1S0ymZTc^X84g>({>2!32>XKUxo^SlQc0
z&3A@z{FqTR(;(hk>U~H{>+kD(Yu{Bc$aY?$p`ak-dA5Sh;kx?w?(Wiw4O{R2NB`&9
z?vx(83%R<wvUQM1%gZL%x^WUylVU2K2TP>mTwTKz7g56fX~snyid3i)tdd`EcE_-M
z?07e}QO-l_zHwi&;nAknp97nP#l`2ZUWwV-KHAyYG2tW^K7E!y92FC@+xK|u$Mp1v
zVKebUm2MK@V5<M~4m#%!Fy|fb&e`*pj1@#$qTRYtQCT@KJ>B>FH$`h}>+!kJdBe3?
zYeMnB)X$>%L;7rlZ1HlV8<>NG%B5rGR8&;DoWx4GS&kxuYX?0E!YM9$%R@6W{VPM+
zxrK#`Eq;5vq>eJO-CbQ>&F0*CY=lB8FIDtJD3LKqKiAa6!)HM$F^b2oji2iKua556
zbmkWqFDE%o8*01ot!--ZtbAEoqM3G{A3?u+*PwDNAS|rYge`4k?S~0x&I^VI;dh#K
zy-StITOYQ<V?6g1M)G-EP{1T3EBmddr^TFGkwnx_lyY@-HK(X3tFREGQbS%v1>eul
zFC#NEGA_;=o@mj?$@p6Fz|?RrLF2J_L`1|wXo|S<|6Y4QSo_)mHl{zDX3ukYM(?nK
zOC2{fR0=i3U%i@qNzO=$$-&F}#7z>JxtH0#!s(f+s;sLk5~FGn6%P+Vy|!2T*_%oY
zc=hzGtj4;!-sQ&q%Bq@@22UTVBt*Bgh}zoOm1-2Wb$04Jf2m?5L4$FCQ9Wg8&OL##
zQBGZ?@V`P9eO0B#)G<0rNF4gf*Y}vd>g>W3PBcyAN?!92H55oT2T?Xhe0wlH2OC?{
z$sPf(-4bVU{UI@2%IlNe#plHB$UY((8X7-V?0HEs=jP|7Wn}cHzB&|Wm%s1mK+0o%
zVYm8|ULjdPT0#Qp<m3b%kzS>S@r&r_hzL<8d;_@n$;nso61@MvB60{8dvNQ<^`8tF
zt`|R78B+TFWeRIHqpwFsL{u*)S4KtQ!X1-gk_mkK_U#S{k7b()r-Ey;LFb2g+k^1P
zNO@0BA$ofHxL*7U4QA?g>L}b;HeTLuerN<a`S}J%MN@G-G`qWd-}=6HSXLQ4iZ7~8
zp&4>!Wj^|xV8Y38{TEZjHQE1qT6G>F#q;(eUj9--sM=^55x01}EB6L=T=5#2gW;Zz
zkrn&h?bf@ISSd{d2Ayn#80Kn?dJn{c!orpt&VCL2_#x%talf~hIj)xmx=B}WZ~LtE
z$GEtFUL%9vPh3no4T&)^_-1d`avGks)|4+>^Cm4UFNTMQzn!v_%pQ^-M@}CIXx44w
zScq2>QR>0`-ysjkkJvV&(f{s@%e7)$O6b{09xdaikbN;apFgxRTB3Ke!8^=AGAum#
zlf+4#OVjbs)}KFLC_jI$-r!CY8XEfZ=TBv==kf=BXc-w<5wWq7uCBx?`4~wGl6rcS
zP|N6;Uvu;GQyUuI`XJxEM{Y&M$QYcMNDBABMu^_z?8Tg<;NQO-dF@Ti|7r2==uU1P
zF*2E=`2$#?Vkmy1k(R*{2DfgA{4~SUDBOG>QmjyHFjc?yXJu1wYr6jRYa>1G1ez(h
zQ6&ux8B<ew4;BK9g9A=3?oz86A*iOVZ{H}O^HuEI%LN9GnyaW3sC7k9%bLD<lb)LT
zHY<yU|M6qmEQ}j;LGM)Z+m}fmVh4U}77pIbyM{PIY-~^HaNEr8>wMiSBmbZlQdGp^
zE<~=buKrvG$*(%1p+Or7c}ncTy#f_RC>J>TM1+4k{H8AAJ~RqDbUCq6Bne4K&b}Qu
zL|OBONTiFT4@{QqRB9A5X?=}ya;$5uu`^b$cRk3MxDX=0xOp?K*C-JaS;5~YQ#N*R
za#Bu9>waNjAtyJdt`qOKt}e=l4{!RnzE#M6$jTa4S0{jkjJ&et1WzHmZW?M2Ppy2t
zD8Gr{{trdtNiX-{&=9Q(DHJ({py2&TQ-RW#xPF_PWZEu^)G^GOec3_>^eClm6)dl+
zo9y~87&kUHprGC@cE=iUClnJ#q5gX}Y)&@xD*E~lO4V?~IyqLB!%ckQUKFy=jxqX$
zIU9n~p1pe26Np2bQ&?yShdS~`+2z~oj+>LxiQ0<Vo<d_|V^p-X&tJR{Q&PHj^X5%>
zJ%gh^+Xm7l`5abpS5{VV^kUlen~lo#%Ez+SB{+%2m0pYtf4p>sM{A1TuZ6E0O531;
zNiHNuhfBrIey?6z*aj;$CMF!J4o+%K4PR7j|9mLVm$YHC(K?r4U$wd2<sWJx>5^gF
ztXI2NHVaaC3EUP-$!?pJw6w_LN-#}7MU0~Tdv{OZ?qYh49`@;eE-nt!W#hHEAVra6
zQPDM3%Kgb>?;g4%vAeg&XES#TUQY%Ut>R?N$B$2-%6ofzNrmhKhO!kSqoc(w6<CPk
ziKwWl!=s|wTE5GSkB^(0n0ziQjPB@=fC{OpsaaT9_?Va|K#vyCt`DR9r`dr#TJ*W+
zRr_kOXWcaA;NW0acX#cYu(p%d?O*37d!yAmo}#I2-Lb68J@G?7hlfioCb%!>g)e>J
zD8dj^tG1?WZx`oHQt-UoOGzIv8LKdj%t+P`nz=lzI$xmex1DXd-LMmCwT<kx?H_p~
ze|6Oi3Z9;kF|zqz)H4|lmv%_?SW8Q*`V|Y*%iEo6jM#x#1O)yX9tIQnIk<SHrnxWf
zqs-gZ(A-wT5B3|*SADLoE-fpI2KQf46^8uMu{NovfXTM3Tv}fKp|Eh+`MJR9*_o-O
zWmaWnG>37YvZiWxM`#MgW5zdgHm$R4&AM1V&DKx%t{0EUT3N9Q3JUHY9?q@Wv9s}d
zYZNZwq#XN_>Q4GySCNpQLBoFm!>jY=U4bTpFhLnj&Bq4^2S0!P8XO&cVqoyFGmQLc
zp~|c^uc^8D!qs_-SBqI?nb|M{pIMu4)wFBy)rs&GvSn3*@WqhuVjyjU5IY`T$B;|<
zz&Ay&Q&O)3Rj<$ilPD3KzYuWJCPd<)`j|Q=7tg@p;BM_TMt(9~)5>v@X>v6sTx@LY
z0W&UV|MxVWWPxIYq@-f$1G<$O@HEp43eaP_#h*XNL6pQ}*C>u`h4AorX>DyXsQu9m
z$92cY(b3TrRaHY{V}X<`mV<Vu{{RC(A>?kTsWV&I<>irIzNm~p=Kr(iptU|`LEV)s
z#`|dcG1JM@=E<q`4;UpZ(FdlSpQOpLD>b0;zsV^sp5LB*>b}GgCP&%>H_^}_1SMx;
zZvHtpH~#&5*C(k^dW~0SSA9c@Ua`<kWT<~>^{pM0IjpI8p3jr(J&Hd$9ck>F5A)KK
zdBWu>D*rhzFG#D_arej5l!)g49!dD_?ru(Q?ktq!!qQSiObnb)YV5$tI5M@^)B%%c
z`9E*@GzX@fxC?zC<r7tO;^BijUtA22j`nMvZJM<%(QUplJ~3fxVUd|$?*p@{2On)r
zT^;ec!ox)@Em`1P#Ma}JlH7ktPrvf@CLcV=r%!J}&$D-O!ol>nrlBpj*1HW)DkCEU
z3)OG5#*S8_5Ho+62AROg+M5et#aeNr1oVNUnMRvIi}v<*1n$(*YF+B}oMWNMV~7rd
zd9SRjtoFs6=o-{HCl}Yi=;*gVoX*^$h9^+n`<!2cZ{P6rJRMqgzC2lGkC$^lo0u_b
zeUG6kKqUF#9ug7~ul=8ox77YtNx)<17U^ke@0?~lVq0InG#@Lw9W5gcZL3g4hif~q
zUH{hk_4ktRcU<dvNk2%JYZf^-1|MBmv;WMDzV$HtPGx0fqOeyoTokmGuKxZ`{Zctj
zyREIQ?83s(r6nVU6rtXD4r3{K`OMFsz0c3xq3~z~b#!#d2h{ez-Y94gA~5bxPAx6H
z7n`3#l@RT+w-g>954Z^xKiE)z;`A&3q$SVZazCw*w$X0WePUe~0kUL|L-k~@OJiJI
z+(&?948lAd=YA{W`rSsuFEQwJ>`Ub1xTm6}prJu*YHC`bR{BoSb2rktUVGoQo|Mli
zUW!UoQ<HJPBxFR}EqXcSib!33?7Us){Px;NapS5EcFQaqS(v{yFVju>zo@x79ghGH
z!KS(*Y58;?DLs4Uoko}QVm|-y_B)X}ZPtgat>_pS>;eKkqoZ`%TWsm+=>V18g=WOQ
zz0lSMW>7!3*Pf>2EsZhfz<)n8GZQZS$JCU#n;Wu)MIotxeLy6wLPU7Hjm^%N$v{&d
z_!IUH4g+Ij(sp(nv<it$M&<G-CD7{zI#)`^3pg-KZoIwL`QiTjWQ`p<I{L@R$kw^`
z;4Ue}&RKf}mqfTfMXS4|YEcyxoP0SfU*~Tyg>^oCIjl#D`8Z7Il<w?kYkBMHax1r}
zXlX!_++e!S<=TrEFRXtzU5Bb~0Ys4{xXnr2+xC`uQ^>6H6Yx51_Z{&sRx^EYRF6+b
zuDrS4{E|CU^@UR^nu?{6{Txa8FG94D99lU*Jn_;}_75q%3T|$X;Zed{=QP8hRZLnq
z-kA>x3BkgYymRXU%jb1API<c=ylVcW4=`4yy{>NN<>v$V(7UX2QDeb*EuO9f_?x)*
zr}NYN;fUyHU>4j^3+d(g!7)qt0JPyOH_uu>d-m)J3}cmihcwary1GQlek|^Zg<LnG
zlrR7}q>i+KD1jWrVR~C^wN6`0>ow&WSy|8*?uI8Nc?_?Xg@@nc=jUfnV1_gKO~hf~
zK7nrTuvuBTom#ysSF>(FUS6xf)%hNbpsQHmnL7)SUOmvly$4J(s;lE}B4a+&(NVOq
zDJ8|2%lrQQI~^TeMqOQ!mFHf+dWFeA#Sxq#V122>W^nAIMR5N)cXMBGZ{iZYCwdew
zSD@oXZu6d3A(^*pdwcu10Pi+74$imk?x)2gfW~=YQfn8-B6C}`SXOa7;(ym_X?%Hp
z0-sPxQ?q-v)jv1CnD#Maa5SSTHaWTQ%BMH(sHu+@HO#H8KVDvX+S=N&$M?j^8KLf?
z{aIV9B{brC{_<rzz#C~r#ca5I7;iwxq-111?S6&MMr2=0&&>Q3u=Zojhw16-&c?_N
z4h{vcY9fOC@J1uESc7b8Ucqb)LJiLF81SD|y8bIu`zbXwHIDE^oeP?&=?Xlm{?P`H
z$6jYu{XZrqx{Za;)MHq6`ruOe+>hDB#l^|pZeO=C=15?6Uu4vYms&C2=OPFmm=+?_
z#ly!}IPckToIk*n&iShfv{{H0;T1=By*2`*EF~xPdZxjn$+_}{r{{H-niG=>VqEc5
z7|c&7p74?yU0q&OR95Ze`jb7NrpEJcZ3aT*;7|#~HZmcSBXe+daP_@sqnC`(k6*t!
zHm7P!$IWr^aGxn_1Wj4e3JNCe@7v1+1+T2FrDtaPK}8Z1ua>GYF{zjNMhs%xk9SeN
zeClyCDD7i>yv<lpao5N@B}_8jCSX#CN;#UeyQ@*?IBtU`*<ycWkAsRNEvp!rIQR5t
zzrbD(hf@n$pxA>Ks;WTX6^)FRGA0mR915fz?_QqET49xmnORmz$ww2;@6a#5_4lVJ
zlhL(F5}_kKdh|$2N-A!^efb$(+pj{+m5q(X-0X(8&ALyNaH0KJ)gRFC@$nIc?mQ~_
zx2cace!!G0a-*(~|I6Y(efnh0pY+<$5UNlqe;B7(&ccHE{{8zuc{qOzOe?_Pg7<QK
z;pw<B+1@pj4ZO?TRTpQEpr=!pO-`U8jn21K&n)lFleaKGrK^b0ZAQ(xp$AnuI3x->
z#T*_w>JQN22j2#E2GEm(BeI~NU<b7%{9ca{2YnZPr0ab*0kXr*37^&#-JP`dp~DVK
zQ`6b{q{s9u40D&EB*v4nZYJNlcjh|>%qp04YDk|ZieUbf?Ho$l6$MV<0&`5g#&*%r
zN>M>!wzN}7*3yzp_@t}6Zl^;9V1ine<^9%HZI)NO!t;5=Hcqv;_P>6<h)hhh9e)1~
z4i7Xi&;V#ZUETJ%{>9=9gSBq^#>0mX&lhz(WszR~+qs@Yvk@S>>*3+Ce{e9cWy}7@
z(SVH*-a~dy$sNtXP}bD|u?N?1<R^t0NHOR4PNZaI4ORv-UmF|agi5$<9E_M+n5Nne
zJcwb|a&6asCXpWAQy>Sj$e)#!_)!HVTqb-*Gy)BJCWefeVZ}FZ&<+kT%x6k{ZX(}R
zR#TI5ap6<Qj{G~?p@BR}ACSu}`D#6T<LOgvm7-UE1I`C8uLIhY{6H#_KVaQof21D}
zoUr3u^U4F@CY#l?djWnvX*Id6<Sn(-X8}MK0BHJmFU~cbZMM!@%Z(`%s+60LJ&K5o
zjOVcuy4t#`|HkUMW0f;(_K2fgt<keJ3J=M*8oPX~2aWaoT}X(Wt*tGh2(gVYFfaxt
zCOU)gx(8P5odw7|w|&UB2W46Bp(Fj-++5n8ZH0y}iuB^&=hzAp>9=jopVig0#zyk~
zA6V%RA0D53zIfs;dc!C<zPVXM<5L2k{V^MK!v(DZ@2^T6Q5`o62pJ9*8S5W$aL_X`
z0lzX%D|Mm5rUY6j9v!`~ysWtDa-bBeLMq_2NqTm;F+Y@YxgVA5y3^r2S3ZwuWp(ak
z9`zoaoZQx%lY9zE{NiS242Fh=QBhGJ<9I9iYK=_-IE?!v5)<VtEjM4PrXoW(Ob3nZ
z`*#_5CdL9PFf87`N8%)AhLd26<<)lo9%2eGFcR!cZF$C$Uv4A2Z@JCPlV8q~7mdDP
ze*8EgJNeWAet&n;uD{gn$kK6Ryz4y%Ns(GJb(>^nR#vI|3F_qz`BgY{l@|<tQdBac
zq5=R&5W|0H=&7P2PiCg!`c!SKamocdOnKdtT|D=Le4R|_%GGbA=KcPv!E@>F{r%y{
zXJ%*nPbX_5P2NOz^WQnUXwzp$N4n{l9a+#L7x6rzRNaCQ9SIcCX5DLmq~KPFSJzC<
zbC!Rkwt~zAH6~_bL%FjvqVtEnL53m+@4%3;1fNL_I7-Lb!Qj-8&yCRtb0(cH=t&AZ
z7I$-Yl3&C#X?_WT@%DLSt>)_Tj8&65bHId#xMF!>2?Qi5%{Pq0TwI7622G?)ul0T4
zWqWKQze|K?Y%nsJ^Y60jBa&QIoH(d{Spt3br>CbTCMFaQANuU?qw|#rpI%O(u!?3X
zhkyDc1Fy<!Us7?`hy2{zIK?9dMrVR*U0Ad30*FSD@$uq@iz?c#E7TxyHKmCM8wv2d
z3^44D*;)3wT!u#UARG{v0Z`EQ?*$O@>iGEksZCwLN1WW;zYj3?em+;+`I8<N5g$)|
zlkQ(cM+0pE(974_T43yY);|Qb)$z7*f+g5HIt>(M9yp1xVEP-Rjez1y%AbT;ukD{E
zSM=VNHxZD2nQc4j=_?juVq%~t8AU}O>fLaFD<WQ7OG`_>%2O=`I@|~N25F{el7Qc5
zZE9T(qE)T$Vx%hLHf;N`evIbYTZ2E%3-XN!C-LHs=BMbG=w*k=l~Ct8)HrM(lam!d
zX3jJ;grhw;H6=)Wyt=-=u)i|&+Iabyni{Xgo}>5M%eR4-8XS0cZUF|OBt%!%&<K)9
zkMX&9$mukWrUC3A<!o}<d8eH~z0wS?JBGP-?h8m8YHHtj8g?)uB9jobb8=ElS~{1M
zNM8a3@Hf8-;(s2q5r<-<BY{{2k{DuqyKU9WIy!RCob)+4FN8~Gtjhd7=j_ZW{bAjU
zX|=V%o12!eSO}m|oqLj(s)g{@CudMSrp~11jf}Jz-2mM+ePj(<2M*(YuWz$1VewB}
zUVt|Ug@1Zl+sw=i24C8jFZktSav;tE_FW)Y(6RY7<bT^oM8EXU%Oh$bp=9|49voEv
z>ns!4Cj2NUC<W@}NWo(*$@Ok7C(){OjLe2(Wrk4ZT70+zCM9Z2YD_WoLpOciQXt<Y
zqr$pV;(qd|$7rwO>|SvDo%f<94&y|ATCw(yj<%cUe9-7JgpZ=9^B-^AUa{w2>Q7Om
z_`RHRxsu|%6mLw&$e3{W<m!Bx+}PRKS--Cki{oG*vP35H0Sn8A`;Q&B)5$MBe)xb%
z=5=n9?7k}v-IPUhmLO1U`|N@+R6?ER?|ccUea1$Z0Y?}BDFAsQLc$cJKedD^M04nH
z?4n|0!y_Wz!R(|IB1_T^Q^`+Ko7$Qu)~UA==#i0;c>sNIU_j|>ts96^sinE$r>FcO
zC_7OxQDsKGpYku?3Sr&B3U1HZ&pc%-RFG6q=<A$z0$CufJy@z=Y8g;>;$x=|pqBvT
zfTm6hvUzrEppb%ByTEK~-iH@Ts;d3pqZz;T^(7BfeoCYJ<G_C|B%J!pMB;2}W=sEQ
zbLvl}E0n9vaOu~CDVyuRKqIGBOnyDvA`>EUvtQ`UAlYTf3Y-_!`kBnTIani%Sa$$o
zy&o|7?ML9*EX!K|t{JP7r}0Ey=<1}i_^Sq`clTn;#^E#dnrBGQ|0$?|TDivV_QQ<{
zz?V2LMf}A50Hpu6rvVFrSLr1*9!n`TuKOVB4w!(x9T61;@*AbY!9iV(1<7?U@BNAj
zQ{&hWMaqy~4LZ<{q!blt>FEQZyZ5CCE6U4b0P>f}9x4Sj1rQX^#(c<!4}(8`JaLsE
z#>2xK`DsQc$eaW$Y;a(JeZxT)gd9Ls_!k!!=A$KVL(o?|9NLc#52p?Y*|;tW^7BEg
zXgwcGxgeUg4%qU#Iy>Lb7G7+;sJf!r`cl0@`yD~1ou<e7YV4NjSy*O^uTG2A%U*X?
z&^mxUVrPfrk9j{WC+D{Tm-JscAUf%3?<(k8kKoZ{?%N`AG&p9JuD@FHRzQbELPi>G
zbRq}aNnpF_CQLLrx5g)IY&-l$i1rj89}n}g9BvGVGF-44(=Vc_#`5x;N53h$V~d;2
zAL$I)5#ZwMJB^<F_BpcQPm0wFWFy72TOVaV+?u)D)DcReUi~bHsK~X~<MF0F74~er
zak8HCXg4@@2(gu!`92t1hZ|OQR~IDTdV6pGexu{|jzH&}8bsM*m*ZK*?#C>a#31At
z3+)jhrr=kHwd3id%j>mMLL0I2ii&p|&lel_TF_WQh?BIjK>-q)gP!wuygA+TY4+|l
z%FM4NNm6hcd%16!32^O8{RvIFfr0t>{<X|Zj<x&uQT+hOmXqmfr46lsyAmfCNRXdk
z^s6khvJyjFNlIQ;JiY#lhoYRE?tDjR-R}4Ny5qUv7c8%&^px(s=e;?t#|o|k=!^*P
z4ZRgy67Y_kIxfC%s#Ixt;{~^aCW6J<c>X9{DvoEyJrde82o*n8_HW&om)|_Uyat^d
zl*wNO>a8SJb#Gf|Uk~TK+${=DpkZbXZ9GZH&dtqjF#mq3_w?!0^Ch0f*_q3Y851k3
zPRpuN+r_-#d_v6L9Ffoej`ThlsihhgfWCj*9%v1rI`U{bIq_&#SyE%&+5YX5l4#?6
z8!NcuSw5GY?n^<#j%afO0|R?|``_05#cLkVaQ%Q$H|%G!mc1EaIozD;Zn!v8!u3;Q
zdY?vz6YU5s6rjM0pTJGHwZ)^e+`PQGt;@}=A7)(pM@K`WqhfE~FoNj;lGNV#o8p}{
zovTl9zW{^S6M8{-*zMN2jF_9#M;LWU1U$WUE_+gPa&@cR{&&fjfX>lxHC`cff!+k3
zorB`5>mb_tit0&E{{<{@&sk<#mL!z(hvVdE^p&h@1?`rij}8tHYOjQZ#8`#P5SO1S
zo6rF`Cv5!ut`l_=T1i^f$Wu%Oubii%twG^hUpGhG-K-;GIKyny=Ol)*g`WOKgIakq
z3r{Tt`Z0{$q~j})3Qag=sIc{5n)LN4d`d_Fgp~^Vqre+tj{mX%&~ty9F(bIYr0^;G
z%*E!=ZeMVM1V{?8u})(XpcnPt^*VkEr5h@dzLe@O)IQR1_!9c=fLSG7km#$)V;knb
zxFL?w($gU%jZTdTe{OESdTNtzCr^uoz^ZNs30}PUbX_25P_On3KuNGZ-8Ti$Vn~XK
zOcRL${?I=p3UrX8qFMs~R(9j*jSqd8`Z6JKpy?TLwAb8-;dGdpt!5DMSMgV%`nQ5a
zJ^)QrCNr&|VEu~;nb*W^HHq|{yhR6S1yGFmKzQz%f%yu86&PO7SvZNKqn=N`V2DO-
zoDpvD&C^zNaNz0}JR;p{sP)lhYg--8KW-I1e+aGotwegft$^&wf5$Mz*Vh?z2f|(F
zJGX9(RP0WL{sN;JQR@>MMnhR^XvMnj?}OU)Q;WV>ru`68>d4-sCn%o1`L?O;=hv~{
zj<ukPQu6T`rtvOeH_j-k8WX4l<7E%=Z5VLz!07gaq5KyKuug2fDSC~?5*utW;rSFK
zaL_-a2!Yq=0h@wuI~#MrkPtXg7I<clT$x4(UKTiLU^5Ngt*>?q{qSM;bgdZs&MlXd
z#n_cUf1o=h3180Bo-ELkCqB0u{}(h@B{%D~8Of%Vm*W(VxOL}6bTAfs_Wo|)UY#%a
zkgc2x)-}Sy`}IrZi^W6-P@CKwHQs?pnM~#0Rq~y+)l~>O(5?L{Ou*JXI5-@DrlGFB
zQc<Y)u9*VLC@Z}_Ns2Q_C7+s|GF<HWgFqA%wE`76BaW;#G3eYz<u2o$b@7lmaNejK
zAtWQ4y*%4829*sUJ0>Y_Q+OT&90cd%S^s780EdZNY5<zh2>PCD3n})Q&3yqCURp|O
zf&6O66x<B2lh~QLolbISB>jIbJsVR12f)|RC{+2EaRoe1)GqEcDhCxfU}B?z`ueZm
zP1lO8rh?q8w+;MNV_%_WgAc$^`RN%iJjqWs&ksm)8yi!;e&zoQ22yO@-e@w;O2t*|
z@0$keArSG}nt`4nH5?>hdoWade0(atR4UE<c)~_Vn!?~>LoHFSttcvbLnR*?C!m&A
zvr7F+jleLTwBc3D?*}O}Tgw|&UftW<Y+-_jgtu?s#@Xs#>P-k#l+vA<t_M`aZ8a@G
zD@<M`@4dL=e0<XH)i-l7<wYwb<T<{PK0pT?9Q0I>S*>M-2<JAllygTbOgWbOlIDYT
z&awBF<jEV}i&BDGHP&!pq3yQ8wOQ|U-)oP@>!HtI#kNiBNui1XyCyhO=;-@5#mi66
z!Ks@cLSlWd3f2YB{?7dT{J@(1aYqM``f}|RoZxnooZ?RK6YRT`(UDp+CT_mHmURDq
zT<;PQ9vMVf09~n7m%M|6DE;y^G@aSm*+4Jg5UHt@AR55G00;)ch2qzWbCxx<A|*#!
zJT1c+{|^+!#}<CkY#73mrd;H~KiBNz|7U$2jubM2URM+ohrZ{^SLF~`Upy`~fYd+?
zllr4&A;hz(^tuxIV&#<p`sC%Guw)Pw&-3n|y=1_9`@|chYZwBe2-L|(5~OZisc`}j
zwNsy6#|B*c?|=T#8TTd5RyBGAgQ53aQ&SdR7pQdD4S!uw8v^!tc|r{E0+6g+ua;zk
zB>`+Bl$O0-i~ZWulLeo<%1TN$ttpE6!w>}_A|ZK5irMd4n*xYE?~NG(7BuRb9lW$s
zybA~UT}z8;_Nia1o`$BT4KFG0`Um>w6mXqJ<wWIU?(Jv(qLdPnk~Vk0+SE=t3(OB|
zT81UW{K9{CAR^lQ!+A9O*XMz}ft7uj7a)?T1mDi@4eB*(HLuhS2?ixH^_LlK%zYSo
z@U;~Zc@^~ZuA|{IRj;#W4y{xlCA+Px|Ir2E58=T?MR8^ekyMF$<6-)bfd(mIAn@36
z{WpYN^ftyT0X@->x&EPqHb;A^?XizRNJNwgK2_hlxtM3ThI4IRx%7HW_z&@e?_v4_
zS^d3HHDk0j?E$(Nna92okmV*c*19KVX2eHRRoX~Oe{riOLx*{RsIxRU%7HXEmQ^Gv
z{X9Z9R|X#}em~ea%>2AMSQ*ecpIQYn)Mj#^sWr@rKlA(M<7Fphv%@_Svrdm)Yc?u1
zm{9^zLQX&k86%yMW!wr42`O>y{nV{6^k(Z11r|EYV$pxXFvb0gX<nbngS&c8&<H@&
z1C*>#z#^O^`y3SmsnsyQo`#{iPW!#21WjJm`}gk&4_#N**FQHje6Z#f*lAdu->0IX
z`UsA%ei};fuxJExASeZ(<jKae)UW0R8I{LlkZdkjj%d&O+!gvAE=d;u94%xD|GEk|
zq+ZAVUTxq#Ud;lbs5@O{T%WE_5Og~P-tL~#Np{9W!fmeG{O)!nW(#Pi5|W+HtGU?=
ziHw%H&*;R!`G-aT^t{w*i)`9$OAw*k)!0&d9U6Oe0APfpicpT>Xe==5oS=4fDG&+#
zwX`jB$LhVqf`<%B<DexOew!z{TT-M)z~V;)eG|M63aro2&tR7FKd$|=xvKB%9H+|`
z0GR?x%In!fT_6~IW2K{bD0pEI^6t^bA%1y$RtF83GwrJxYveNknnu>Z>p~>WhAqzi
zsuZ-36VB`E=&O)#Tn7^~Kv9Z<kJ@njH}w7c(P?SLGwm=np5?v(6cEFt-7`lm3s{Dn
z@jSlh9V5`^JGY?cC;xQo-f+}8ACZ7NZ`K7_R|YH!z%!~-wBky@f7A6vZfEx=W_|to
z31TFCd(w5xH<$o1vrF1Rtukp;1RTwHrGSH3X1B}&X8%>0$JqoRjQD={rF$R=-$C*L
z3J(^hjFC}HqFp;PUZ7YGAidbebLxWw2apbwl}bN?i+ttV9!#LvQ*UW$x%9i)=e3bh
zne&c55%0f&fk*sFWaVS;`c-`<<|#K&-(sJ$6a;moVcmqB5%Bx$Du2M^r>>{g7}eWn
z=ktsU)HHL}{PDt8Lb;~7fI`+C&k9t=Mk7fRg`M6G3=D)_sDb69s^fv@7Lg=L=CKeo
zp~XPAed5lPlT&S;5cBr!bs#h_fZO!B70)Ko@O!c|GJHYVhH)`9HH9z}wQ8O0E?aJY
zF&`6E`PJL~-~|I7lFxPE2}699+mxWb9=Ot8j+t|F5@Uo*y&iN7N|XDfoHJmeR_A<=
z%b}7~&@tp)i>STbYo*eO2Tf&aOp%vf?(PD2g^no@|3EOLZhug=?rih%`EjT4(z5Us
zABZ1@vn_s5pSy02M{c%AxOTr~r6kqWiNGjh-Ogn72=pfh$uY7ATI*Kb9wwNUvpb!Q
zPeepg(G&lrGDJX6+sz>p18|7eQ|O+Gc)D1+5@3dZH0b2u;g?D1`**QWk(5~8lQ{a{
z4;~9q%%3k$IU|J>PE}oJQR~VRC3@#c%Ln1#9JUncaOb{-KXev&>OS`tx#+6}Py}#W
z5q32AR$~?*BY+GDnt}l*akj^0HSbLun|Z%G($dn9%qX#4ltEapAm4O$vo`k3hY2fy
z=}_>&3K|?CIz%I$8mrPau#*R+{a&?G($W$%bc>`97Z1i)>@j+dd)wIdQ7{7|ACRI-
zF0P&wVJ`)DZqBMR|HmNiTY0Q$o_D33qqrSyE-uFEEPV@S02a~;m4~;~uR~${gq^m*
z7o0HXxxnJRGjQ!=rx4+Z5gjj#+NnOs>b(B(uM8&}&uj#mB_IrFvT)bVGkMe6u5NA!
zQBL=BN_hAV15NsXR3haEDXJDwGjelsD7d)r>b3hz4LT9#d1>$nH92?fR6R>QH=0EH
zJykE^W?k#=GTs~@_5Ats=XXy$2lzK}iHyF=*WP-kjE`%y%x>fP#97mcNgC63IM2VF
zsP1`P$H2f)qETq>ap5rSy2gx<>E!6Y*Tlm!9=o`-)cB;?k)Mxyn}%V%Lq`Oig824r
z08a?an0r4`qtFPN9@OmzoK8i^4!I9ctCx%0WxF@c&&?gT(F!$>SH5w%x|llOjb$wZ
zlEF7@EjKv@piDM~i4qr8-jt9K)1S|2lWBN(*nQJ31+?sB^f)ZL$xNf1xzle)<xpNW
zhA2BbY6Xyudjr~0(+1gVAUtZE?^zVWx?a=J(#D2_AYLlGe7$dLi>g;!ssOVff<toy
zIeSQ&3cG!HL{i0w21JE9w__D|5|X*Ol8KG|{r%s2Cq;}+#JjFzHE#%{52;j&wLOL{
z+`hU&1|_2;Cjji)Zp-;q@}}VlSu<NO<qy6$HuhLwLC+ekaM=^%*jMyeea_>+)g`;S
zwsx{%HPf<uR1Om9606!1QPCEE&n@Dsi<gDJ-2V_h`>~S!{yC&>f`a_kXR5+ZPhBCk
zX?3zJYM1Tt1D+OJd=JbsYhWzj;uU*{02;w9bWi^p9&S&tY7m!^VE~Ih#a53XxE*G<
zNTEu9PU6cuHhW%tf3~(_vDWwjVZ(F>3!$z^M-8|@K<!(0q2#EA*Pn<JAh-tBR4|AD
zz;qEZh{d{v3@3!mpxBzXjuiJFL&%me6bU#agmB^UnRAl>(JIHzd!+!`olyDE{hCuK
zl97G?LS43RmmX)<cQJSUSlzE9BB&M~Lgd3-!=Ia4{P0vKi-3U+l(G*F4s|9h&0GUZ
z0nSiL`Enn~v@&4x%b!g#{5~4`HQnd7)zJCdJK+Qkl0qM!Jh$U0H3^w1#N^ZV<MFyY
zH%*v+!$}+z6vXR!ehUCFWOLSIuCCiEAOmSw|Ad?alkX<-VY$h`1HQGu2~m)fArn9<
z>>1@x<{F1^>Bl6La6tm|SILiD+7#`%ju3K3zdGcjw{;%-Spvnen`3nW*RbtX#;_S^
zRP<;BDt*qU@e0}PuVpgo>ysc>2p)WWeLdJ>_wL*RT$It!a9y9>Tv~b(WG-Ahywrgn
zPo|Htv8}VM{QQskyL!7ILwzvtCi~MT;=W|TZYclk4OS>_aSsncu<tups?UEm0VXgu
zMkX#iEbIMI94nKl4~g@iJU_vm78em|#@G_ZhXcI+=XL+IU<GH@XO(<FtN_HHX=-+E
z(Fk^QTxGXGTn8*oc@o|0-~#OK+5$Xi-f%Q1AG2Ip*6%}7gRwh1s|V1@$ejWVNJQoi
zt^nG5Bdf8XjV6SiUiT*p5~BaZM;6h(FYgVEjC3M=;W2XvL`+7o5>d8oA!zHc87Sh9
z2uCQ{*>_@iKH|XxTY4m}WQ5E13L7qSbJNJ#nXq`I8=MOgPBSizB+SSzF%c0;M>lLJ
zcL4C9F?2(+r_Y$o#n+|k_gxvIa=`C5z(k;9W~OFlwm@RVhCGAwR{aqu&ISj0FIA`9
z&Fw=#v#rqBc$x4Kt{+79E`?=`yC$TL$kAwweD2jj5U@c-MYa4hKQb~>ZqSJdLKPr6
zN><^N>gwv;&PW+9_6sGYD57PiYp(U4-(~!&z9-hBHh`~4t3YZzlqC<TV4yoMk$bj|
z%I%gFrkxEK4GkAFbPRI`Na%4<ZI*h7z&f>ZA*L_V;w8<>G!)2fz+<$9EF<F0B9qtk
z4`#~N98J5?9(IF`1P3+sssCub8;`tvC?d{_xIi>R&?uHrD>XcNGq;@5;c2)FrJy4w
zfDKL&Js1Gc$O%%PeJm?`1T73sCNHT1Hg=XmU;>X7S>x$&@o1G3tBZ>Zp;dFmWqWY!
zkDot#X7d<gST(tGE4|;{_`Z;w4v?d2#*+lb3&@#DStu{~Yxzwr&5Q;j#8olOfr>-F
z$SNq<|0LQP4^#qF8qjCxKadGZXlqYnpu7VJ40ibXcqLkFcl%tstdz8T`|=(;H#fKG
z1VA?xM{e%?gHI{62zYlqV8&^e7J#TQuoo7*z)bB0D*%CJg9%t^czEuDm7%Xc7lj6f
z__rA51Sls}^BzYIz-zkz-#%q0y9OcRA*>0GeJ_DOL5>WLm_{LD0YM^NJbeJFT`3p$
z1;atxTarLAmP<4cQ0^c-1rr{+8~CO*%x)Z9oSfT*Rvu{i!!1V*M|aEd-93+~&X*Fb
zuu=VvL5>bhqXR9^W_LmQaDA*5HU^}n0*o(C58RJxXCU&1h3PN+%eGuJfP|5cFFrfP
zQ{!|<YmtCQT<N%-)w8)st2V<b4)Ka^!pl=g5@<(u6QW>e4~M?;v97Mpbj!F@EvKO1
z^+=HxU>e<?p11=KMVYcv%PGP2@%n^95RN?cqiYFgpZS7RqL5ub%BQ7gZhs5%^YAH5
z#%F55*5q?LWa@B!h=*1*dB%Vp_yh8U66ruw53<fh293&<N<Y8LdnjpQZjw>*XCCh%
zh=ech=gvWggZg8_N2{#lg0V{$HJV8&ri2TRFkH%)S@Vl-Q9<3(n{R4N5)@csg#&Wx
z0qZJpk=J8p$bavzL@q469)bcID>nvD3f%kh*Imj0NhDld`5kNT=MVGWBc`UJDsw$p
zRkuKO0}&5U=EN6^`N$Oa?YzuDyWR$mGv}4eD+;VTAYXn5)Eblq?`vSC;*b4lVClix
z#qq^u;5J+k$TY;EF%c2i$<A{q0EaAVBQK`w-N3Jd8i7zkvijW;2naYYC*5CICV&|H
zxBTLU@1+KkFnS+waYfUnxJLkfb4vTVHHc!u5@m%-*)g`^*Z?P@FEG<@xvbUyZaFBx
zL<k}}C!y8v+xp@@U(4minY66DA*{(jV2@~K#)K1E)E$2-Em_})2xTTC#?W5*TQ}bZ
zrlbhE#ekZGkP&U<Kf#xOcN;D1B8d?{6gE^8KqJmqEp9M-84c{Uwib@ohbhZO`eLO*
zmFGfMWRT{X`%{{}u(YV{#5J_NeW!M2Mjk>XxBdKZ@$j_|546@@xoj}NP%B;|egu>n
z9cjSir(K%2z5<e3Z|^)K^Lg0SjxY+an4d4~_Wvls*mXG=R(0O2@ye{I@GU4HjbYJh
zd3PI6txyGCt!0(lxFM%i24qdHX=r?Mxi}bspjbRk*Mc0Rldys4!JB}n0;z!WN7xmb
z1&V-qBI`w;qNe{`sbtI?9K#qq#+b;zb~6HRPoXq|d7Jdu2{MEMQ<eyOPbFn$eg_dD
zgjPE?#%DhDH5zms1_u1yMOLrHvbUep=s=i)3=O0zcMF-@NU=gCc23-}b>$IJdkVyR
z;E|#uof{o6fa~z?>5=<lJ{@vEoD~4#2xM>qf`e<<gWkttaIk1oPn*!^(!;ue-04)_
zmoFcvG_e!7Exs=0<;=pkND>YkHcR@_l#ySX02&i2Dw?!hoy&+SroRt(T6uYxcBlI*
zYua)7Sy`WeY=daja(b|4bz0Fdb6ZUVrKAX=cIgEgmG55?B$O3uR(+sNIX7^1{R(A&
zSNL?`F32>{jkYh(YT*q4mQ;TAs?F=7_{u@$sIt7=$AmLTCUa?6RcCQ8<%%>jvr3YB
zFmrn)f;QRt-PJ+k<y)_d5ijxl;a7*h<>llSgfFtM+Nv5ak|r)i&0wUGK6XS8mGJ3b
zZk^xhSrr19`3yp5gPGk3h6XCZZ{V{@hqpW+$_{aBNb`2_RQ=1LAf4F)>9o;CFY*=B
z74zwel}{~<gjRlbm**~Xb8`UV5|o6O$5s!=#>c-mHC+QKhyp)Y>$x{xs02vRspaWD
z2Wulb4vs1?uJX%qykBQWCXDsHfcVSnF6?m)O+eyD?8KqC`1sn-ui=4Dvc66=ZIV$@
zVXxw<46o_IEHG<5g?+cA$i0t<RUP1f119!^{m+#<7(KWx##_R?&T%~V)5&)>ti1ez
z6>>^Ta|35F<`M`JdjOt5;N9qv$8B(yT_=+7gi1WrIUU^6D^)8RP0xS`tR(rxx`o%E
zm5B*LwF(IhO;yQ<ci>z<tzL))%U8>Gzz1ipE{|MJSGH`o>esRa=^^9`Du4o1&|-q-
ze|9MoJ|h1~I_h=JpCeuo#4RpXnk6z)L`R0jF{n(i`|d9N{QezAvB^)!GnN<eaYBqY
zk>7#1c8c91I03*B#4;fJ_cbdwcBI>!+wO{%Tc3@1?282naJ$URnt0&;pb=(y+=ZOy
z^~RL8wl?*(dSXm}6#rH`qpaq{{H9&kdN9&chgPyk3udk35YDIDmX`mpSv!Q+t@Ws(
zCnH1*r>$xA4UB~yXBfos_K#d(!{PY5YJ;n$Laxt_Q5(R*<)*hcBV}$GloGnue|Y!K
z{n4Yrn)-iyImr(%Qk8RIY(Js!?wqH@y3;e>MFB*#*kdD*$YSm249F8{zwi|YBuG6V
zw+UiH`S_OnKM(4T%^N_TrLZ6cqw}P#Bb2u&-D+t2+Z_mKL1q#n5Wo<P0SZ#_@twpp
zHtf8d(*z@X)cngE)Dl>JgA6?6cgi7_Y^r%yeX3P0X!*VSdubeVxgADaco!_4V7VQw
zXF}LKUGjtOqj)K4+4)OI0e$T^t0WFp33hUJ{&k~MidU45^b^PRUx@`x8g<U>8ygmp
zF%yopQ|pr_WMHdhm6h2J9zt*n+6_Ll250`+3OuR%_mh6o=2cb2s=|)TmyH8Ng!J3D
zIOPkZ&jTjXGPaY49@@}05w<Q71o#V8(&<o}e>;0aRWDCoetpFP{*&-A>(wIbXyqFM
z0BNl7m?N#M{#AY+SXfxNZ#N_7yg-FB2~Nt!S29{~MI-^E0-6T;Gi&YK_X@e@6$@-Y
zbanOCEUhVob++a{&jtbbb4JEO-b9B!``52u10>QRe36=CWaat_G8_oiU-OGOB@b&@
z!=<NaGX>Bn)%q0eVuU46iO|Y715;pA4z3-%hT3(9^8;e5ZQ+K1iB~Iog3C9UBD>lm
z6@{er^ae7pOhC^Ap9X@5S)9e25ZZ!A4g9hEYq9p$cE&ZdCz;AShZBvNxw$J}1nBdI
z&A_+KlA{c1*9Yqyu&qV?4Q(pyRQSIlX%*!PjM&L^{kC2GmtJ@~<O&??TrTPyLypK;
zzm(4FG~4U@kfLz!SBE3%gXn)7QXs^H5XoeB1EF8VGfx06cDmNd4q<{+2w9n-_q65U
zxB6N0%1KCcaA`SdqiVw3?~Y|o`63)OvX+^jPm(aqoy)p!KCpHGq~e)UDb`bjYp*`k
zpe?QhSwD(tlZCb}9I$t<!J=%(<E^?@$eCLmADGjTJ|M+RkEZukjfpe)q-+S*MLUk7
zd(wvwA9k+)Gy|-J&{G}O)*z4z>3d@bVL>vU^z?_%G4p_0YkM5Lyj(55T>bKxogwI*
zHaS?Gpf6WXoVLwC)lXP8%0tXSV`EFmcfbU(3YELMrLwxQVF)}E9CXG+LAUpnZ$@@2
zi#<YtN%jQ}lmAt1u*zG(C~=w*f-h3D)o=z3rk*lv;^!oUN;ldn4|;J!JGM_eZn<CX
zTp``K<s$;U1G)e#S*6jzC;b4MyQG+%T=L{EaBsuh2JmI{qGlMF*Ap*c*q<=rBnC7N
z9%iTn$D7{oH?IpbGxwTQjuTZ;$S5ir!%7ySXzTViAHQHJ5Gx+Rfw+=?E6=${-Tt%*
zm)%4DPtULa0_o>vq)w>BQ^>+(q?}xM0*WKy@znoy<YaAqtx=fG(9jolJ&V0AkaW)H
z2w)ejzEI)A{Pmd2GI9-Zr5`30lSLi*#h7+~eo(&1>!SQ9li4y)q;GG(AG5U#f#2g5
z#guvQ2QIJ1-1fHfiq59b%8%mhIs40fKl&m8G^Ajc_O^J-uKSq#!~IT&IhqZN@iXnJ
znj;~_$jK@T*OWsSD`g&@`lB!Mk<(Nbb8VPD*9|1trKMVku|%=4BYt=zagyGbmgL`?
z-XfPAxBhrd{*K;O{~jq>-!HOX>TfpcM^1}oo~Wr2g2AB++k*?sR?laqLB{&`Lx%#^
zbRa^;$-`CZdBG3Mv|Ow=x?x}%E2K!o7{cL(Y68W5Z;xgg5LVAqqaGh4faL6VG1$$?
zTJ(wZdW>52s^;zat@#C8hQ-YF)nnA-D<izUz14dSv*$DC9BXSl&bU45>rKQU+n+3o
zdmHvkUF*r}wZob_k;(rNY$2@E-+$N3>&oRoiunHhOwbd)1xV@Zr)6J|l$Vzem~ci#
z#aQpQNuvuPNxb*Bp`{jNBbqReD>2(^YvpDNI@41^QBy6}UW_;XH9OBy|NWkLI6~<f
z983;al+)1>j-ZnN+I3sxGc-v^A?_M`9d2ZAQsxc>+|9Lnn;i|2fWDMQf~e-y|FQs0
zKYp+w`FJWP09F)KLSqVUZ-IVxyxlsZtfkev*H5qw{_@b!Q2Ksl_L#?WP;UPHFg%rt
zNB}wx8w%2jvJZ`)_SLmNBqJhfwk(f?)jds35`;Lrw4{JK!hr)bv)XPM^RH^c!S=(y
z{ZSB=^zrdA*6*#b;uKcB*;x>n(2)ee{4P}RSI9pIP)mCv_oC;a*0tq#6N?=)Gbx6q
zrr*kXetabA+UVf2v{<m_wU))cL3YphL65?dC)eJ-eS7Qn?R$iTQcwh`l&ORs#$@B-
zF@~pS$ZYXQ>ZOM5Axkcxph1{IUtiw@QW5Zs!0Z7}*>dE(W6qI3>CU^)zB4X^Obpla
z!G(bK2>X0;Ae5t^1d2&WKp)-xF)ya%{y?bKX>-!$b71ns2Esw8yExopmXhf^S@2xT
zK%t4+$&Sv;V_?zx8laLNzgvZY5uXxpI|j1HK*inn2P6@aKB(pmR~L<00DwM!ra&mu
z0K5?cZdk?_+WEg9--jfJdTAhd5fBof`$n70^YU`faWgKCP<ymKAtQs@oAxJIn}0%X
zGvyD<xVhEt!uWW4O;q%0U{DYZBV*$4-&i+%ZlIu}xAl)~1tQ-XS*>_o`0W9#i*5%<
zgEbdB<Ttd~Hfyb0HBmaYq1M>}SJ19s%EtLxWmP(c>ip)7fPerBhDSkv;!NP7-9bhD
zq(YRaKnsq63FiY4^mw)UhyoSgQD6}h%Q!ppawpsmZpTTPSozer?pwxW;KYk{TM4gZ
zvvXc^B_lsS0dj#<(k&w+u^&G^6BQLzD|?NSn@iu;-i|{`T1O{03V@valSK)fiieUy
zOKUn_PD&`kU=RZOK+cl(h#SA4uzB;YbcaNMf<kw|-qsu70q8EUXSfxmpWW$aZ7Wya
z8OqcRyLfwGjYj|>)Z;uY@w{*qefkvAM-Q2pCii}qOn}0<gIxk=zKjWs?{Zzw@&DQc
z{2Z17`4|@MI$wo+YGL7&l9Cb^!Nz7?3a(+Cr>@r4Smfm7Ardny{q((0AzuAsaIj@#
zqAJ`lesoj~qLTqZK@T21c=!9`$4)b8E=@6|wm)y$zU?z-LvSEH9bZC%33h;ce0<`N
zkP#u~!b%LKhgkQ>q|IM31<GQpM^rDQ?&$@$i^BRNtWDt(;H3_m>d9;gpCo&U=;pbk
zJTW#V0P`oIm%G^RrT^3iirpsOF|^*a6?v8XXaI09sB{pP-{vNZhUNsM&NCp?JwCoh
zi0+Dx^e`|G1%L?-35l$`J85{hlA;&|sMKz+KUP4lS@~H>q_3~9xLqqg9^O4%T(Q@$
z7jEin309B@?@w|kghH+{q^(U1ly3a?t!OF;6EXw^2a_|nYZP)b#rRb<?H(N&oE=(Z
zWKbDA!hqPu$^MXHwD)_68lYo-0z~#2mhOE0{IH3MC5(+}!5ahbjpaLm&wu3}6!g7^
zh=^=<cd<LJ#|Ia>aR%>?J!wfvWLOJ@W+8`K{s<3}Ul0Vcxj1zIGXez_RUd+9h>eud
z(b4|GPC#CiY-|BdO%y<mqFk0fcZunE2|I1>|Cs(I0!sy|#VX++KLW%@g&K$~7^h)s
zNJ~q*gNfO(Vyjm*My?yRZE{@}>y~?nP>4nc0-NKeyRkqF;g(XOV_bG;Z(95TF**`t
zCX4Nl@fcOE>6V7T--8nqvcQWnGpS*CHa^K8#v&sVfvrVw#wuwaU}0cly55#fV2E=w
zFt|zXc}xi_ymJA5AX&9Zl6$miLbg(^k2UB4tQJA(M#jW&TM3QZfyxjJiwm$B$3^cm
zBbI>j&YMaa_%>xX_PY|ccR94061b4Q=~%R7fApva=>E-{H2=hH;!`O(;MAxwrNj5*
z(64kYbDaF@5X>{}>;oHcFPk9A@bAY7DV%)zV!CCxp90OJ9s@2^xaHK65{P%agXZQc
zkFmyM9%ri~9b;;4^0}-m;_wSMw3o34i^tNkkpUk*T#9zA;)#csY+JXqv;r7!GUpze
znvyqOwg4#?u_O$V0(?;jSf$=khdx0{+5^(ylDNsBStZ2LZ4Z0Yk>qJ6EsyGY?i#8e
z-cfs(gJjc44w?H*)&eHC9@^mGV7zG6>><PwB;>AP<My~eopoEz^3d(3r*O_Z2tXf5
zPvXE!^H5ULO9BaJZR8Evn!Gv4aXC4<p!$FuvKx4*T{R)^;&OYrKLweFaRBG8hI&jm
z6-v1aR{rq&;^M8AnGlVQ>TYv7*%&I9i{2G`{YUYHU@X0U#LjC16><aC^4s;*Iq!nL
z|KCUS7#b491|h7BpwhN>w&ZJ~4yvjIAe<tU5>PRrn{;W^*ecUB!v3F(jBIXR!7U1>
zd5IX-eyT^0WGpJ&pz>&VdE<am!A_V6XiOu;I?FvP9Fuh}kpMQ!+}mx%M8vX(1Rwo6
zl9Q5FNM-sQ0vr@f=5rJ^$H)AIjst{pOd92(&<WufRfCcZd>0lu0^y`0(wUHMmxs42
zX>PubQ@)Mx$`)<uF-Y#+Q+V+LXf#)(SOm1Y59C+A!so45+ihzHKT>^0Hx4^pF#s(>
zP<VKFB)R&ZO<X@T*q%kI6C;z3=<u2LgC!XMF>}C9fRcbADvC9^62jtQP7YmhwV`I!
z5}2r=t+UT&5KrOX4@ExM(1o6c^P9X0E{gmtk&1yK$m8s=6jpH{82Iw#OUNO+qNo8-
z0Aw7aS*0PRpa5(A{y?;)lbIQu0PjGb)i*EzXD>CkCNS?i1JsiQJUEE3CLYBZi8PCd
zNlL;dcGr&|KkBGzBB@Xkdfzwy-fO$LrC5@4s{lwoe6JQ>uge>il_NC{YxtL!m(YiH
zL#^C>K%j!0XJBBUIvYVTGU5ii7e!+XU#ru=WzJk(SZxz{?OZ()5LmA)?lqbRf?jvB
zq`DLpV_yU(&H7;VC2Lh@=Ua(~k2qgv<@~{(sIq$Q?wfK8&UGTMjedkalQcFU`iyMj
z?TML7q{uE#%eP#d#c#LBXq%gx3qTi8sy5Aoy8ZW~CmFOZ=&_|C>dty2Xz&o2zPUf(
z<t4o4KNOfV;9PUVvBJrE_op-n`lEHG9BaP>pEltn_6voIiHqMOBI=50qkf^Hq6}ga
zzzeXs7{L<Sb1j5r5t}`uhp>hXxTbH+>^rzbm1As#`zDpaPfK2L-ay1#uFkx!tXOnd
zI?O5~K7MRMq}5dlzr$voi_&co#JW)X>~=;@lwrKL_Y;tT#9$vDc1-rhRbvdnxdDR?
z2fm3;FRik+dd$goYZzoev_CevrqO}rCn_#}f`d~6%_>_lB`Z6-Q}!)zf4G+2_459E
z*w`&^Z#F)^T(Y4gbu%LifEN^|_3_1jO-o71#T^Kq<z%%8tOFx5Hb8;UTrZ{@gkX~@
zFgTQ&j_!^(V%0jCH_*^L3KBW6?|{1Rnl!Ff{sik5eB%%l3=~ROq6m&<v4PY#P-)nh
zo2t{IrKa|S@-;InDrIsYx)*-Wjz5KVZ;lS2B|Mn)4FdU$%8wthpqVv8w}piB`LNCf
zp4TqzRrvY&xj3!%b(nM3HHqJygoTA~+`Ku~=yer!|JRDWEwHBbL)!UGhxrZ+c%Rm_
zQ$fR{^P(|=mcEejcvWrvc6xf6b~iy&Q<I)a;4!^W)Wil2EMUTSzd66S0H!R%b_70A
zM@NBTZ|nBq2Il5u&E7Zjxc?3o{2Yt}yJ{fy1qKA@!nfGjixIKp=S>12Q|;YrW(X;8
z;E}Fhx2)&dP?*vm)*ZK?f!LovH!SdbZMhTnm=UHD6d|Z#K`=U99ybn{yf*yBfF$Sh
zs#ZJQSKt#fXsxBCr7kv-iLCgcw^gL1?pQUP(ENWieFr#~ZTo+uQj|?7WMq#JGBb;e
zRAw?lA(4^2(vX#mlwC4<P-c-$Ms|`UD@l@!$X@@?{r-;sa~$t+yuI&Jecj*tI<NEm
ztW(f1kBGG*+`#p$&VC8k2i}Obc7lHk8JONJ|5*q67O1bKoCeiUzHz|Ke?E@kCa!PA
zm2Z81)7LC6&dqQA;8oJU!>vrsb}^n@pPq({#AL|~sTdLvcl?F5PITqWx3MUyp08hj
zkEaQ1(pFbj6Fvkq8USMcE-s!Y%fmr|rjDu7A5zjX%wezr#BTlcs37!rhYtMlU`1~V
zGGivoq#;WuiK(H%SY6$AE8+djcX!!GbvYd5zZTcLsp3z7;o^6D)otfzNl9T652i|M
zP)(V->Aa(8Oa_fEQ_;YQP37vks{yYBHQyiD6U?!*-i(FaBv<`OXINI-E~kl|uXu02
zAV~@6^mlP|+!A`Iu8)yiNn8VtS|i53E8!gE2P2!j6XtMljR(ZpGPAR%ofru(5|WUW
z{cWJAPtUtcGgT0+dBxC-!DB=$-L5Nm8f(SC@Q8W1B1WxwdH@2>*t8m*!XQ&V#q~e?
zP+Ew~1}qM|rD>@$4;n8smsXm-QdIp(qfCwJ*<>dPSC4n6g-ldLdwj>mpW_=Qg2i@k
zdsS@JchS&{=9#{A_&h6;+Z)z4X*Q?0D#mC)DFls)uI>TARDMhSS2vbQYsNojazY7`
z%grl#>B8asnwpWe-=KvqFh}^UwmILx>!n%u@sx&!%#ClJgbWUV*e*Xle+_;8Xxidy
zgfG;_Mm~N|H18FY-S|t1u#6tJ5VwiWVk_0o;7}pF64R!~f&3Y=(JpTZ1I(<f^vmp8
zU0dU96Hk9(WoP%Ku&|r=!*iQgW^e1Ej-^8DS3l{FaF84SGhkyX(aqi&@-DPzvSGBq
zy3iOLM@E*&`4KV!0X=@`_z3kF*Qp15^$$77iBC8iuw1cl^zslcT0+2szG~?34!hTt
zaP|>Xt9JwnJT%AbtB~V9091$5ShHSH!zC-b)?$?W<I1mE=U>}(>-T$%Y2t0E{V9Jv
z&ZM2Ir&yUCzWq%??(az!w)ADWzz2&C$4`H3KQ6><L-McfT<B^YRkxbL#t@&}AIhz%
zu7ex>UDU_Fw)Dl4DDTi$s}W+T*^nn)uD)|nQKIw~rgb<{@`XA(2wVt;q}8tNRV7$k
z-tFmB+MLuwS*3y$HEFS|0rT<pn#HzYg~xQ_0vNa1z_jA43ZDmXW<7_!Lx4hie!lH#
z5>GIU+v7DGOguYYagY;+M)vKM<~#T9b+!qkizKL11rdwtyid=2CVxahX6v;+T{>||
z=)Asuf)WNMB#y1D?<sKXuaBIX_K@o}%7vfj=#eA0_OnU_3ArJW#>vyYz^zZjpwb7!
zLc<))pe{fE*^68%yVKqLUZP80Vq)R<q7UG?$yxA^|Ml1%|7!9!LszI;#;nu3BYx#A
zZNX9SLU_)JINwA7gT+Q;;8=(Bh|Tg06O#_}p55}wHM~y0d|4OYPNa?*)vS5GvZzti
zB)p`f|0d5@E`By*%F8<y#a3q{exxl`+;8%Xg-;$GhWSH9#YTdk^?+ayfF8wB%l%YT
zQO(umW46D2@)r#Z<9X%M)6zt%tD9U%&RBYR{R%Ckc6codLJ7*Rhke|J*$c*2D{Cyo
zL~SYdZEWmzP2diwtYYn{(H{;>8%o@I@k0JCRDe#yl`ZJsD&Ih^g6~kt`IwmE#+)jY
zCrT}GgN{;4#xJ6|orcQQ(P0H~=7{dLVK%#z5@BfA!N(YEe6Id|q28RyvRnN&<ccE;
zyNc~49bfk`3-hP+I6rI>dVgR-(j)Fh$?IdL=Bw{VWIe|B#gC<lujFq(e%iN8H$Njy
zlV!i;jQASZ0ZOG@D~2Fjl6S3n^F|82&qSPlpPl5#zCP&*@zRnKzW#h(8IL~)VIxk8
z4I*$ygaB2&k(QQ*ovIxK?L=m^pnw1_bQv8T9rFSFGoc=b>)6{53T3B}>W-KUo`s_h
zN{*?Y5NU;+98aGaF-iG5=J{6vovy5|q7%gL+@D)iY*OLzQovl{3MUum7wD1G0@nDM
zM9sOy#BP|Ywco!_zVydC8Y7SEpIE(LW&TORNzz^K8P3MrqP7vkL!POn0&S$mJNPz`
zh>{K!FZ7H1vvpIx4|xV|N9TG|C|g+M_XTlCN@~BY!~DEjHL|Wyy*cgd`!ZFp!C2O7
zjw8S7O`!6i7}WA70=MK3gd8^Uh>ebJS?E$yntjE;Ov}o83Z0CtZ6@?RF6$n9)YTsw
z7hmZJwy|=&etjGQL_5jv8FlV3P3_txhr1_F4V;npTj~c97zB+|Pp4^VnYQ8>jTl#W
z2;-u{&kuE~ij~z7E345;REX0tthnOTqm04BWaiQ4=H{Y8u?KT#ugbqd5+MNh6eg0k
zE5(p1v}hL2yUPx74YkY)C+TKiG}~#SuU2dAcZZU8tm!mWPw*j;j8t<oB-=SSbNzPS
z+P%%R+)>bd^1~$can2&{Ta?Aod$@Butm}?PQtV@ioG8_$6lhk8z46v%X>siLRI{o7
z{%Zz2-P<E5O7dTS`&L_9f1O^kZ@l`+;q@=~@)LWFks)H<&<mIktVZP<%nO9~Lrbgm
zQ2Fm4>yFUesfHZrND8`%7(X~AvJdIy(w${^)EEwv%+nh7+?#(2d8%0s9U?#obYz6?
zgpExjzYos-J83oSoBo?CD+dwS1@qtiZx5+a71*cz^v6Yj4LA<GiNvemGWn6G{@uHY
z#_#;-*dU{QgiAx;KwlGf>wKJSwL2kt&dHpbcoiddj|FqzN)icAH~Q)*g^vRb0(tQ6
zNR6}RZ_-~J;o*7YIlQXSnJWClcVt~ew6Y)xWf5s4qYt@3Y6OT<<P{W%<?n4Fv@6wq
zCjf}X<`obJ895Bz<ByEkd`+m><l&^GDiY$-6i3jZfWCgz$fh+^V?^5oz>p)16X|&J
z+kUs2B^Oao#p|=v*s`;;_f(nB9v72liFEw(aO}}!9YA0uB@5e!)2?sDk~%1yy1BzR
z(N99X<;^1uP8a!gB!SyO{ISdpO-<-fdYK}^S|=!>qk*L0K*+l(6DigMhh?E6ADWQu
zzkQVN;heMI;jtV?n?9QpnyxYh1{Ss}b5@5G{K<nBO@((VZ5=Cb;5le`rE8{M+|IgQ
zbEP|~8%C2~XOD=HU)TQAKEkwBO*S(S-u3(V`x%#R4f-2NyO~b}9J;8Ja_SpPq2<<i
z4cD&{we8Jqm4A!=zr2CMAiZQRZc~k39Zwq9TEI!n3_DPRD?zM5Bc?|i#jVLHDe2xs
zn#UKmv6}WjQ8C28+7~bMOf5oxaHVte7a&oPw$0ZFzaQwgJb3=hn|I;XN>As^$;tU?
zwnWTOhK5UOm%T6;weLrO6utmtZZ>#pn|<!-c^AnlZCBN+Wf7Rh`A<~Gv=!eTFt~bL
zh$!2{^-&uXin^zrMPlx=Y8!m}c5jmj2bnUV-GO1xW$lwaq$ofC&?+h0*l<64_6#zB
zLfK={I|8?v7dtRhkdp44TQ=I2r+ND}S_V0o1O;=u(`AmVk+<f|0>;G<^{KDVZA=4Q
z(bjHAw9Mz`K}cD>e!y=i_s!)<rO}J9`}cp`Wip5BVE9_Cgw?p)#)`|*w6A^4qYr*W
zgss)6$^WoOI!r#DDRPevQbxt%?)&`e(is;Q8X#|gsqrviM-IrVw+6z1ikDpSb<(3k
zQ(M~xsx|XiJlFf(zj8}D)HpnoI^I0(XM~Ix(zlcgPI_sy2wSD{`_^&OH8(xINCr7a
z;8*8w9m~lAXw^?wEOd$5hWekZpY-*k-_EQ@XBYj5=CP?rd8o8PMY*xLPPsLZFF(7`
z_P8)vLE;+8ncZp@TwA;3Xl|=`1zq~|Y0tj;vehr`ud?^f@mkI*k}lUgaCgqrDOLMb
zy2bUbX7jke(#DZj!z4DP?H`;K8<a%Tg8FKL=N>NC)YK?}R!`w((Wzuj0&n{`aFxl{
z7jp$T9Ap~$YcahJ9lIiX-vW6i0J>mM)V)}n81+LEZwkI90c*ME-7mO}%X}6E;H6kl
zd^R;XIr^g|){U_Mh`Nryf7RND3;N#skai0BKP@ApEN?EWW(?$sS%rnss7II{Tx9-`
z?cP=B=Y90A%id%;qtl=gn3<Vd{lvt@X_0sNe+i8c5X3~MK9g!!2^TjN26PB^63WXn
zsDzbev_AD0E)+YO>_g}bKuXbPK%k$5wB!DWNsX_55y8{Yc=mp0=fJ;zgGc!I?t|_F
zei8cM!4X;6dq8Q3UxbEQ!LTon$ecf43U@|ov>~MtfT`Ks{L<2r*Gs9^BKZ~zljq%Y
zO{~te2M@~F@A%0~X@G>^$!60|pxix!gAv@CWka;<OI2!SW*fo1=W&T8B(wqJf`E@V
z?a&wuXHnBWc$ts@3zb!sQ&DK$uhvJoU#f@Xax?kDa8JQY9Or>*-EgH?*e<PHaL{KY
z25ArA9r=$00i2znOVQ1yB$6IvK!79@LtZl+-Emh>-?%Y1@j2!62Oq|8vZ{dj=5MzT
z9=y|)_?oO{(||Ie?IB6a<wnCT=2x!<+z#o4js?fbRCP6M6`$NqcAG@fRL!EMA}wf#
zduZsLs^#mKN49<4_qvb=`Fh1|ZLcn$-t_NTVdi9sY?_jYzT_ov%kJ{PX7ix$V0-Sl
z3lGwMS5lRKHnVqjM#>&t5X4ikNu-k-g3Viy9Zr4$X>>gV4L`ZX<Wh9Oq{M0WC@Lya
z6rKGqJzZQ0jun6qG&Ir0=5g`dq40(+!0gNVScGBJUDl*ZHVL3<Y<}SM6u}JIylm&>
zmW)$g;Lx2IS_!{(OPT0xE?x|8yle#_;qUJ&betR<L>3t0sQ=5$OI+B<FD;Ei%u0G6
zE83mr2u3R#lb=h|@Q5@goBd+n6Rx7g3Oi)Y_VP9oF6J{H&)nXloqC!{ub#aamqzKp
zOUjct$I8mevAZ3P9XnRJ^z|A6Fh6mBfGjJdf{Z|ym)IN=UGU%JQT9N;gMe*?+U|se
z{Kb2seeq@r6#)UhxxOkXIcVwV7?GHG4O-}Afc)Nijr^LlVCZ1ZX4n2L5yMTx#MJE8
zr4TYpFG5d3Lb9^9Ch%8!_Z}VNm!BJ5ZaciTd(EPu@yDDo5LiQ+lA68^qE-Rs^Us`Q
z5;6ORrzQ>|X0hx9j0j*__S9O66+ohch>~B&y*3BMbjEnc%J-917O?tmZj;rz<T70K
zWcvDLr8!{Qdy@AS$$cs}k`I&Q8?6^3;^W0viUW7$?!Ts^@L4c3)yqQfR@<sy^H|~m
zP@izPGt8Bp`Z+#MsPX~>mC#1vbQif#4oX~x2;tn>v$dn|LO{r%YVM<;=r461$?NpQ
zWJ1g}5oZ<_7XHi6H=6OqOx+j`K>fgh%cTPo^xLgA;x4(bAFZs&5Jnpt8#39HfOCHc
zK_iX}N`n4RHAThUpsj!la~wI6`pNb`hj;SI;Khuq<zt~jL)bTw%9Ph>4NzU)_nvZW
z=woxA&?irxFn4e@=UHE{c*)5r@AujIL$lNY(?k0ibdY+CP6toX+aG`KQUBoJU~p_l
z`S|S24Tr91p#!)qBHAX7uj71TZ)p1aSC>So85w^bKS)OgD;FH!`ymdE%8WpdiqMdH
z1ciB25I#S&5)=YK_tW1aKf(mjC%_YOdz)Z;iZiHR#p!mLF*^dpYu>dV2t=gD8B<f)
zYkjZ$zgd2?6yqa}?w2q5#%N5XO0AZxJ3+_%A@{`#p+6_3Fy#$ZcnS{Saqx+XroAS=
zffj>@&g{YkdR*I^_08>csw&;t860-<v#Lm2h-vxvDbC`9gYx=)W9{J6b6ZB;O={k2
z!5ww)6Jk#v=uPw#0Qri^K+CWbnzxvwq};LvLK}DfyoBiI54huTK@0XjU0+{6;qSA=
zxC*QT|Mw1Rdin<ZU`z((Zuk!OAiES`q<VG_4Gqn)sRj4p85>9UI6%1&0S2d~DM9G{
zu~Jl?leFSp=Z6mu#q&QnyiUE=_VBIWiQQGo*PrkrLD>@37X-AE)Nu06;1~)fkq@5+
z+ksHy;q^ycmT8S?ziOm7)|q1dIp3o67wui)eI$ZNR0ENpon^iF&)s*g@v~=E=RNi*
zQxhR6Ul2f@(d(p4UFMb@f(}_RK4i4eIt?j7_l4Tj)YJ&D*-hD6R;R#YS^|x3N^R{)
zHT?|mEA#VRKNT2!@mjO{&E0xzJJ@hXjFU4N$gtvv4fNg?h1v@Ze5|R4zi02(_mtk@
z-vy7H#`Wo&wjW-mr_Z~OO-;2y_}J3gO8WHqb97PZCMMeW-I36)3e4sL3pkbXMc_HR
zzUCya+@JS)Nh<s=X#T$zAZbbRbGl!V;j_pG`Wik<UrlYoS$_BmAPlg~W$fO(f|OLe
zL9QVUdtjeHesy&!0S5l^;e_n8a$~+N=qkx8ME+n-U{1ScXLldDwuo5XaS~Z-XbLG3
z?`QyUThiv<O#uW;i(KmbSGljq698DR^r?(XQG`vO8Y)ur1)kD@$f)iRTmH6B8YX0b
z5FW;-YMh#w)=Y58Rt19x_9f4xD)Hvj=Z-Ht|Fl*aL>|W>A&7`#WV*B|6lVJco*1O0
zoQ6I0{3vJd$jB^@<=v1Fg&PGJ*xwmd&oi<IZXAQfDwd5AZ`2{Bt-TeVH8RfQa$a-P
zIy!WP`GbTV4k02+(dFQ_f`duH2A@<{-&G{E0Oxsez~=6A=O>!gdL3T7&V3^R4OwM&
zR^hV|3xdg7QWbpJy!_BaXgt?1tYu+lwz8~^wRJrrya9(flBuo7nskDARin8zA$2?^
zY4a0r+5L3H#JxkDA@%}*_r;{k#&!$$ZBLX?8|YdXziR>PD(nmuG=ifs(S@-Oi6N&G
zzU&LG4HvCGed<(*X-ll&(86l+kvsRs*4rKa80B+D_7|Avs<SX_EF7$7l%Iw&%4Pix
z?*UWL^eARzd-fnmDij?U{1|xO3F-Vbb2ZDrz(67a0IpUgk{eT>h2aR09m|V^qzg%X
zf+MJ<;Eg^_$d0|h$0KT^u&}VFYB0H&1G9lRwJu|!!l-N9@+bNQ8D(l9aKQPv$Vt&6
zQ&K*8Xghh!?Wi&}QlbEfa)ogsxfgX~bZkuZ;wjckLs7KaGyISHwkOt~G;hrZ+HC5I
zyre7=>-bKgEnQtZlt>d-`1IXOt*n#=nAR)B{ZnrPO0~tsN$b55$Vh<t^5wrcNv{qY
z83wl~r=3=Q2f5n_-m|>i=bz7Egr=M2R^Pi55fmZhE3|a^vNbss#r9<i;{zcnc$yEZ
ze3zq>lQ+}`NwdR;oSGfk&dEj#wT0d6jyrsgkfB}$#Cb@;=TqBFRcbQGSwJ9dRPHr~
zTLB&iF-gfE_LeW9(hTOk?0>utP#w6?r{(1=nlUg1q-SQ{MXH52O)v2q{n!3wxV02%
zpFK-G)#uE?&He9IGi&6q$Q=;)qoXw7LW7qYx3)--(Hc045a81>+4*nHbm1jO90}MQ
zjDv;!f8g-Z(6|S^$HdFh>iTfe8N*Y%aWU`w-^>*UH16BPMDwp-FTQKY(2-n2kiMSM
zW?Z*^RAUF;(D$va>AAUQZk`JGM!mQY@@vZ1^X!brkpV}l!!H8$?HwG>8NP-6?0#hA
zE)EDFKYTa~cL5$NhM%r8Vi!|%n-T1WafC4If|d{v6ujz}in$!Q%A_al?D&W{85tR4
zhJp(h7zbE2;_)%Zf!||jF5(tkyC=W3b;4`zTXHAWjo+XC^ZgO(yDQQ6HA^Ulr;ndL
z{kT~v)OJY0)%`0=kG3tJZQ!b6@0FwZcB9>yb6>@iuIa3oYG10M%FmtbOOziL>2rAH
zQkC_yn~R}t>g0}2i?X@<xq}S-aoa7aZ1%+$Pq_VT&)|ZR#ml4lfI;m(O3Lr=sFZ#X
zv2`GQ6joQg!7+5n+VzjwplH=8)5UNfA!jRn1=&YP;x$}73mwh@5MQn~RJ29p{(LNI
zK@fXTbEdehg2~|%-KMM<y2q{g=j)=piHTt93qC{$BA}5JQ>C4g(-ph-?YDNa6FEyf
zyBi!~<{^xcU}`n@?f|XG!((u)C-BDl3S=sLj2Gz`jdio#lqFWdZvoTPMJzA09_;K~
zXjuU}LZ)r}QYu1&os^XH!_*=O9Y9a`FKJyt?E(T8ACzAi=Gub$L2eo`ViHV<*I!Pg
z_!Nm32GAq@52N;})oNt(`T6@B{B_6~O^TvA=&!Wxe<tQ(UJk|RCeo~JpOIUln?=+)
z335F(Z`=@r+6R3yd`SdLmQ44ZJou!$yCmlFY`oy510A@f(QqUzeJRCz4v~nq=1F1!
z1F{R*&m`;PieB}q@vRSb94?BE`w{2z%tRx#AoZ=@%jj_zO)I}T<-)RPg%L94{)L&5
z=B{z!CtgCnGo@A03cnMsDuypFj{kMrTR$Wm+`Rj0nDxxetatSg!+2c%AXkCc#REgR
zBrRe0Ikoxcz4Fp#jt3=D$NX@iOXm-4+c{<A&7o8haMw4m%5Qe5zi(D;UAH1&d|hyk
zty^>H(__ED&DJs7^87S0+s)ygi974po8@g|Y!e_-0>?1lpzxFNZV*oxCxNv8{d?@3
znI9I2v>j6PL!;1E+@*L13<<iXOY|;9fG3Z}YMKnul8r`_#`A@N>?nf!3`6KNPZ%DT
zrzaCukXUvkzIH2$gYIwKvqtVD#HPU<#FezQnNN4{;J=2c>nON+=2eM+E1(v_5OV(9
zxq2MeE5fuogQu0YSG%&?<quL**Fqv%dD&R|Zhk(n>HsZYd_AZvjAr!v$jON`Dk5R{
zvh^6XItg{$g&)fSN8gydB!WTb{+gK85vtoOU20;>xI$g<n$*$-W)~K^n35v+dJJ(!
z37nQ-cILpu=9p;DHjW!M$SY^7CSPlv%jOZQ_M4saaY^sCW0dTM6p0rTMC&rhC!Ss}
zBJd$J)U3i|e-OYMsBL?_TYfo8vh-JYiXzEiJEGSi6Jxi<_FCV@G~%F29*7DEocr#v
zHy9#fHUx-ho%AGn^1D|lMQIsXnqZ{s-No}LJ7>0>M|$b6nwZ|#i%$9xWkA)?!QthH
z<=AVnSF{^J)M*<n>;zhk#{Yh3=xG1dG;2W5K%u&14)*)OY*4vzRzc)WROdbXzW%Cp
zb-*lveai`Z1%;5^ZU6~KEIP%hA_EALAzQ$9jTml%=0#J%sNTe{q!jqZX-G>`b1&H1
zC)zvd8B;-_3m12mkGj6aEPP1XSxVPr+r`Djain@PdLp5~y?lU8s{QA)T}%aloRHiv
zVPeu#Dld7Ea%Fv8)S>hue$s>^nzV>G(@B!TQUM}^3f+iV_5T=Dh4~dsFOX(EvSB9j
zilQ9`pI5LvXbjHHW*ubXKSjq9Cev1?LHg}arD|K#{-+5|6^8LNks9^eb5wgbzRu=G
zy08<$_$#-SYP-6eO@nhao<D=ukbU#lsp9G@XfmXohskjF+`+(1xm%C%6*^GZfWi1-
zQ-eN2y89hd7d9o2xu8AgfjbMu{g|w*maQ!h8mFj;Vq2*sRMv>^OHY!La1b8!y~U$r
zWTFf2YW*Hs{vu<Z{s;%h8S&sChg;(`)P6H`avO*|=i#^KC#)QBnwh;lr=h9IPiV{$
z9Ave=beM4I2W+nG5}ulyqrOsXLi0>bpPP^GSM+}5j`kN?%b~ntr%oz8MegfGMo@*R
zu|YR6fdI*;w;IDO2nFIw`s9k7hx{3!?L<yEwy(7Crp-OduRMN@<Qi0Q7F+X7WoE;L
zIW{Iu&CMabxqaHXfjH0k|DFbOz^#Cmg~C(`gG6x-#dJ^6d_F(|E|@S)fIy*(A<D9O
zl@<hP$XujSm;vui;DNKjh@YAMul3)0S)rkUnBSi#mqRixmeXIl(C&E3{k_MYrK!c9
z3nC}~EjJ$*&y#A{`9!mc=l^Jo3HK^k@bM#%7?->A!vOzxZ9R-Z5K~}1B8<Ab>rh-F
zbxC-v9f)5+X*z)9!UJ>M$7PX&4f_43YE*^yLKbH^34-eHBDL)G0*19KB$k8WDv|ox
z^OB%Xi?}00x-aa?3!&6jgOLlJ5Iy}LZV9&&5MTV<-bmYiw?4c5FfZ@X#-hdc@&5i?
zrabMcDu1m?lDQX?b#*;%-4s+&->R0k;s`r0FmZ6L(9!9ib2VvEk){J4fUdND^afHM
z`1$xOfrjg~pSv!=$;tVp)R`UQIeJeVK_Ws5$Qi*k{xZYMsi&l%Yi&IiN2=>Oe=cBq
zc_iiYThuLvzK&22LZ{^LTEuP>us)u6SXvr8p17LY9ng}{fvjFev`qt$ULvIqe-y48
z*R2ij;ehq&J#1q)WXfE}_aV@@^^QDHMDwEq;eK{ErGGZv7CK8ta&<s>_x5bf<=pBA
z<m9^#-)kSY<9z;8RwChNKoFnM6{>3WV=YM{bd39p8=XqsAZN6#6OfQ_i%FWf6LfQ;
z^DstMbi-&GXsD^hCPs4lriegMd^$jaV%RC8uive@OGq<j<wvX@o(m}lq(jg|be!=}
z!<lamd9lwpDg(O!Q1C)bW11Qo_$anwxPfJ0IQ<Ye8;pa9!8P5udOsy444|vE%lCxN
zU}mIcU|8IBP`cxgz5L#Y-zqEQ_#i1MDTid{<On7J_&EU(Ee{8BDh0Bl5L1|@!l}_H
z{EjX{4+NXt*Q}Xb!<Uyg*Y=xLo5}MteuirGg#Q{n^ho6O?Xpb($pDO=bNYa3vE#(a
zwJ{FN_?c(6Q{XU!S%Ap*fd36QgN?@R4-j<#*^#mR3EcU#Zl{J}-uuMQZ)a{ENwy1z
zh-6o3Eo9;2cxC9?;xh>aM<EPD{Kbb`it^+K5z1&lA;j>g>lj2DxGDjphPT+HqnF%V
zUse*^-39OytE-x)J$BCZ<_A0ifYRm3vzOz<p@>WT3pQQMfu}|m&YpVHUocV`I{t~d
zNvkbxM`lxR|H}0gRZ<4#I~;+e7Cy{YvJ@Ic5td?4^)6+3%>0*lj+&qxQQ>-6=m3on
zRP@vo7%l$4FLw~h5GE=JUZG}SP)BwARv1%c`f`EO@)qyU&x8v#D&QX%3<Trj;{f|H
zIjz>Z*uguGToXn@mI$QXKrSk*mGjgmJ8J41ku_=A0qTCoE(PsUh9c7K-xp++mOr|F
zzE{s(TXJ?|SdNU0?88uRBe9N7qq&KCcEBQr8#EW;_(bTyUL<~D*obRtYHIoWxOr-5
z(iNw+ju`f!$z>{6$Wn1Vb3b^lVIno$^{1dFEp*JTZ&ydD0#BksW915_#=i&^`a3RK
z^S8k;5jgjwP?GoL!S|j%t+fvq#k)rk;KYkyPb~#@QeXcVI+>(s0~3%qk3}wBwzDG-
z!q@OWb;gwsNg{xtfK1=UXpIbhLOfqECxpY?<Kcn~D`?-&ViSSp;m5WK_wS$fm<@3G
zYAb$_(p=<DI-f{ckZtLJz5QFnRqR80Tj1&kA+;hQWN@`5NM!Hbzklou41pyQvB)x!
zzeHRDu<-J{cSi}wsE>dj@77JF&41_7>h9=jwN_Rpsjg1jru6+-aCf+7weH;>8@jZ=
z5%OypWc6DsQOABdCNAEzJ;~0d^+#z-;ycsRRQ7c;J^IyRrF98xZD`!})*PxaOszq@
z*6!!d#n5g$1cP`62v4I<gJ~@%Cp|ykdH8$luHCz19zSj-CShzY0=)JeDNj`N1?Uaa
zjV<7UMskU|t839$Z(MxG=2tca9MleZG{>!jYX8vPJ!<>xc(2Olg#r15F|yg~%ZIlM
zt}n9%=G#dOjZKv6P?O<pA|-_%A)=k~bDsIwxF$C_DO&WB;c|6w4tU<siIb6>i;2Y&
z5u;A$;(p{HJjV(nG>16(ctOzgA)G{qxO^8+ojUavEC|%7*yM&uadLKcxvZl@byN52
z(ZRGmbJZ6jmmTAGd52x2SY{4ri@mh|<L}@-TwAHuvmIM}ZDalWW_|88HaDXYmH992
z#3fxaeXs8<@6u@sj<R@R-5|0EpqGil@EUjd^*<zlM?qF6?-Ki3&eMmUA}_Sg@W$M~
zmmvVFF&FS6)f(-K=&tk6y9s^#t(|+I2TGQ_??YX)J~V6RQ2w9%j$ywA!SIi34oFV*
ziw1Ry9beeeH4aVXpD%xf&{(G!Dl9E7{*6uBXp8ed{lBbQE$9qoo{M8soZ#gzcXlCm
z_e*eZ3w8&B>LKpNDD7v$VVtlH*ISCsl?JX$<FX;F$|lEe8tAVcLT)Bl$NeDr2;l<s
zgE~4LaioUe6Cpydad)Q=LZL=4_OFpfi7ms_+n<TLgoTp7Ma^$Qj?8~CWi=sD=5UaJ
zWe1iHu|95F{9@$AqnVzSbq{{L+Ts;h*k5g2zbp?kA_na54WV=p6`?!TL2X8*TM$3m
zYbQyBKmbH@8E<6(^MI;44%bvFcMMF0xJd1;Ei}kQhKE~2(%8c$3dU3Z*>cZHF<91Q
z-R3yAC!Mc!2_4Cy10l@{KHp7l=97zy_??c4Xnv@2cKBuy<!SL?nMc3IH&v3kTgf@j
zd@K_aHwmIq*I%uX7kMz~d%fu{n=<vRxyDkQ+vNb+BOm3B?<|mWnZAr~x~D<=eGD7u
z7WZYlKCfoW?dXKxK0O)R*RaS%`^Jp|+V!_!SYli5^g4_~ns++pN2vFIeb7#f@~jO^
zdjKOljUHAuCBAc@6wc#Ck?b8scRW*+(gSws&wb3S{B1-{ZNsdYdWil~yu3=RjQL~n
zqEqi<W?!Ds8`AwV(f@;{FmAI+RETSY;bW7;Acgqp`Jv^(J3%~}wD^2|%d=P24L@g$
zOuD|MrvC6BqcIz#CpsWQiTRA5|7}FiB)6RJaVYMkAU>vJWR%MM_WFZz1D-NAbh`BR
z*hxwpc)-fSQjfrb6_m<fH{EazAh~&bJWc<+Tjb3tw!=Y{SR9A#g-~)lv`qjxhZpXL
zt$xWDKp|Yx(ji~JvfK+%$lX8D$$<ED>^vZWq>AYOoj&QDxIsxlVGZk<gvj~-;eVo1
zNes%=(1RKaaZwoZV!!2==>wl%rBn;5A>kdXL4qc3rSCHaYNLk$NNv~!5QB^rSFVse
zAU{fu-);fM$44zPwTLi)+->dhg%}Qw6&4!z@`~JMC+YbwFT{iBckcX<Ez0$+|LR`v
zW!7KkdOme7?rOrFgU%)2<s_aDAd982o4+Ae>Y2Pic0A;rGN4SHkwodKCUB8NUL<N7
z=}GLSk#-t<!flC;%>_#XGW{^i3!!H?$9i6UoF6L~ILL$h7RF};+qrJ6?pcX_BNUnU
z$OHwaNyz%b1p=2s8gku87khS3<JIvrte_%R+HrD{lDv44m+|^_qz<bz4a4D3zA(wb
z|CtE}Bi|BXGTnB_7fL?ib}uiF0n%tb<}Y2TCsgf>4<<=Dk=8BbVf~&JN)VFp8QIx)
z9tepoU-L1&=oDQe90g=5Swvczl?B@ft*0ffp;cCnv4sryt;frI<82aHN^y!>tz4Fn
z5NNn+?ZYwr<t_`M=n2oQLo}l_n6_Dv-RZr0l|EH2M5qndj{s*CciX{8MfyKN9AX?h
zOyVbJS-;gcQhYMc=Lx*1h^-VCg@v(ZG<B%(_Lnjl8x!0YX<dqf&Ew@$_4bfl2?%@#
z{zt^hA#=>kNxJc|DA!xFJd2vtZ{IArHPcH<;?O=xIe*o@aT<=<dl1(k#i}13bKLRd
z!L?9o^AS5Aq(d?_*_LO3uR*fHEh2R<dC<(p1n-BwsbLuOHS2G#+r2i;Gi2@Udu7FP
zkVDj2aEpWl+Ot=EX4zF$xA(Jtj@o8RDF13o5L2jom@bs*)nPKmO=}6)j8+j11SO;W
zD(4<&;%&aAk=r<V!xafPkS?PTsB?vJki#ONfAjUep`9w67=Fl<)pCu53wC&IZ?d~0
zqk>G?22c@NiN{C)d7MS5BGOhw>n<TA<k&<~J0T!>IwphMI3OTEf-&6Bf75I9F4GD!
z0?q)iemuuPPAcsr2>@`?ey=eR<y2j*Ya;B27+Xy()1qJO-ysbc!$#K&2A9-hxY6P^
z04LFF-Evccu8VNI8ozv_Ng5Btt1*m0e*KS|govJ{B?taC<CoT21Bo4~6h-?e;UK;@
zFwJ?)W$k))-xM+4T8hcLwEeFYKLu+a?inDHN*XcTz~pfG&(juwqQx>^Bt!2kK<J73
z<M(4OocDIqY#IfIjx0??U(&_h-Ip`DR;<)`^J?uvyS6Up68jcY(|*=o-AolUk7yEG
zm%*~bnI`wrnu3C2^=n_Xlrw2O;2I1^T$sA+>Xbp0;6}%y$F6HNAP<RI-x^AE#qDD=
z7cMBCB8_h*Z96eOvP)S_&|xnX6|eVy%_{FN7k=<6rb+71K=c=6@uy?d@b}?)zyPJ@
zy&I{t#FDJb=Qt6Lwu5jAIgmliO;6E-s8C6kXrz+Z4xr%upXBbLt$pgjNJwoEONG}I
zB+aiyAIyjFUJTCs;^4Rck5=M?yv+S9WmQM*yvsv2hgtC*rM`G?hdrSRvkv-Ci*mx|
z=5C;OMizO5u|^rT3ZVIisFV_C3I?;9jncs9Fcay2z)G&$PG*wbcE`CzbqwqG@^F87
zk(>X#`dRA!6RZec`!rIftgU_MEvvM^j1#`Xv13#Uxu(d%0E^^?awsk#PB_EG#Hf*I
z^!m30CeSmnk7-%HcQfO`VJ8u{u+F)FWwnzlsA#qau8AF_CcE|I*TIh;<tc1|SeVHK
z-l~1dktu?Xn%I#w?E#b31HLez=zzrzvL5C+apEo}FC5Q`+@#bQCzw6y+t#+W6k%n6
z8g)T78^$tOSyTG~oL%^O(L-cpddI9mTTWbjL4i9a?d|W6ncnGlWJ^1LE$fp7o@DwC
z+38q@0>+L!qugG^o(o`+qs1I-jYL6bmzjg-;Q}H><VIExAcJH?)R<Z4D3;<atMVQ_
z>S6d@`P|2I)mw}CD3FzYk?Q(~jBbnSo0zPAaC7$<{yF$o1(+YCYjB}~X%_(|%(SVB
zfj1>Z&f+ZTH|RG30`o>i$PwZ^7*ctA--3LE$6{%~SxKzy?>oZv1~s2Og{G^7j9sGZ
zqaQ9}N8&=i{}bL95)x8itRcqY&tzLEK_Iw3N)9L?k>YlQ(4wr&*oA?NPe33&BLiDh
zMxX&<;uTDwOn<8UxPZ+6rz@pk@iCj1^?}RgD$q}ao~2qU_qnIIa@;Rk{N&#A{zm6H
zf_Vde-4Psakhxrb!NA}c2Iu^7Y@K}m^y$fjKk&)~gBjuvWQTGPe_5!@<CGK~505ur
zA3`b_9hBKS$tF`eXI%G<=Y~$h-PDzuqW!bicN9rhcL)!8R$jFVxI7YiP)o}%`>(al
zYN})Pvf0IWZgkBDLU!i%`sj&E^!qB=Gz#miRs{7)*^c<ONd#QijCYB0?wQ)Z!s6d;
zSAy$^h#_^EVvnL2tg5+~-xnPp-*j?&2k<aJK16Ij=$F2YU*>%`a2$aQf{9PfFeV`h
zo4HZiX`87#yQyP-Q_=i!?11KyPqy}_#yjDaAeJ0n3p9Jbe#sqGVKF@P(}jsX%h7qY
zdS2V_=w*+Qftlr0H@}sPPbY$gMt7K;l)HP9r%hZWn@*Wa@o!_dI+rC6kMMZc_J$l4
zrMB<3Wk6SI6=FLpD5*q~lOFfWOU{(~^@T`7bZ6)k0nPxJ(4#+p{%C4d@`aU@4C3qh
z@oa)mpj;ax5U^*61wi(le-ujCkYRiow{k^17GjUt%WBZ%!3cJqQrs3jJ*(6YoM5@!
zB7m1;W7d@uqg)d5^7FL@G0oz8lB}|@@`P>oo(>5QD+9=a*NE+3@a4=Up7eBg=Nsut
zKA{Sz_*-H7mogVFoNkLet#Z95w6|~X^;$gcu0Q%d$i~y^#Kret8W${{2)y;26Sh-R
z+upHwrCvh0^5#@L=WfI%0Y=?RLE-z2S4`}ziTo56Dnd>tc=Wd!isMxTQnfB4nNO6W
z5JWc|Mmc`ii&jxo4gjxc1(V&TDw&Yg#_IQK(KFA)I@I@g%O8m0RasVO*7GitZ!Gjo
zivnKczWfiWui$>ipV%$qxAC``P<6(~UmRoZJ_VP-`KA(<v@c#^5|bgr7PQK%kLvwZ
z^g*Y>x63a-p=mkhdg4*Ut>t~UF&M+Gde(nY0Pr{-8N7?fCS^%2i{S+v-v^T@;U;Hb
zxXb-E`^AybXE&FAO^>BN9O?v}3O*4n{sVTU&S<VvSf@>iY7C+n*dL4?b(8KsUsay`
zW{J(`k}tc(ZaP6FP|-F?_JjZ3$M;1ML!&UB=9rN45_;tbqZquuE4}$!{&lT(;)_7V
znx1b&WzR=ENPQH<Lrx0(dJJKu`j^=T*MnFONxF>I_p$uv`~S57-o$tdd0WO9F<agk
zl}|~z1SSJo4@m969xENEnuoDd%Bel``SZg`9Zm2aLlrUso#T`FCfuaof#07e>$(lY
z6h^&BO=h~RGiT4@ey(m+HICtK|5>-IxE3zZGBYQ*8B1;#8X6jMXujHY^BM6DiNh3L
zXxqjJ{+S5xhAT1~ojIZokl~DL)7qw?_$iU~1wBT7Q{zrb!6I_+DZzi$+tzBeF3)d`
za<Pvk)%g3|uFgK>X`wW)R!J4fsO-F=vN3ZiYC;?C%JZw$u!Rr%FJFRyCyPGQUG^{t
z&i`d_0S1Gs5#(M!3R|TxCsxiO^AD1W?@(5N(RPD{$~<7z7QY2oJzT7$1@4pQpk1!<
z;(Q&}dS2LaFWlC!Gm4b5PW#W=Me;pp9=$ojCF*riUBTUnUyr9;{<P)aVZB3C+=^1N
zXcp7--KJTpHvJF&gGqbm;lqc~*x`y3c}$@hf5P>LHLh49O%L4Q{D$M8saAlROO_AM
z-mzeWma!vlXM82QnZWX@H7v-KpZo~L^rL|h<I$sSC0}PYwq}!`*G78Rv&=I-J6wPB
z!+&I4xNwZ;n@s4=y)vwn5;j-f-#EDY=Z=Zmgz%P|K~@#ttP`TF6t6r<xI^;&VUX(k
zRGDyV??_?qSYeaT)urVZt2_<-14}*`o*6QA<Uc)H<u<PM>C&ae&le4xd)IR(Ruy=$
zV=)h(_*1g6p#9kkk6^TWR^^J6K_)DV(t2QM#ML(0aO7sXz**D3f{fwVP(|$cBgo34
zbZSa6;*|JQDY_){>`m)xLwj~>Amtsn%e|B<bVs+|$sK){6`MZpb@}T<uVbAvl<IFY
zlR^Z;VxDBPXxqJ3sU?N$x8-r+8{_t0i?S-uIIb-GGuj3PjZ891Ii?HvqMW}3zPFPk
zC^{ltK8N#}ZuZ!a=g9bEcHlcOV4lx)GbU1fA7Ewtvz&c7`#gtkGT{;|b`q<Zt&W;o
z8kjw65y)&{u-X^6$&UdEb(P!p1d@Q3YqqPbMxX!py(`Bc_e+uO)c(?UPM(&pzn)8f
zy0#`imL;jn00%SP4tS|L_Ii6Amy(+Q9xjSiI5GPESavd;<~UeJ<S@DZcuxZy#~Us4
zmqP13zzK9T@0!*GY^FW%UYD^ec6<MI`$%Qi#sfQ5E(h(SD?gL&<V0D%`Z`}d?D@JT
z>iwhLpNuIP#Ny(g9!qG{B$1Ss=56E%0@d^vo}s7&Mq>L25tdId4}|#xA;DVDpTcZS
z$j4B%b`T4aplrBY+z*b;C|0x=CoS!tG!jpX)oVXO&`7?!s=>NYpEO?MxpHpr0gXCJ
zd&*d^Q4St8CQVzjQCpjjr{qPra`$2U9R1yUjh_C8+mvwWS@YC6UvZX5JMO(E`g5zB
zl9wt7T^cb%Gj&?jIv(TI`F-j?13Dl3^al<_TAW;PpSms&d17H?7uJeKp$JYbWVY8Z
zs6T_)2(VFlw*#LLpCvkf=8mKGh^z>}>P4jM5a8w2Wi}+`6&h2zef51VX-FZ#ey=Vp
zoRBL%GO|!M_}`N4O1mRSg+Vf4UCx7j(Wm2vhuL0Tx`yH<tD7vZ&P|(Dk|D`Mahg+A
z`hdpU{Y;%7*=F5+A6^N$cjvu3n{iPxc?oQ@_BGqF^aV$?usr!7CGyVgu3b|#;7Lrn
zI2<eX8-03QtfqkhgdI8tD(!7;W1VudKTsW?3(y{*qy*pCk$Hy6r2L-`e;Zufcr$d$
zE99cNLy=VS=6Z!~E2oRby;l#8e;Xo(@yp|_k0g&@QZ?WrT%*rVjliuhre5N^&>{Wb
z%a=L>eQ4+rlDM4p6-mB@#_jx@vx`eLS4rbBF?J?j4%nQ!4ep}YZ*@MR?ak}g3q5ID
z6xgDHH=}D`m&C1;Hx?i<DS{mv(s|O8SeJrfL0j<?34HH`{_7qz3CAdV<{4zz%hFzL
zuScq_)%LSg?xDJDS}Z~n{ZEl+?sJk_{-<}AEYGe_h)4zsRNIZ=YfFo$6;JBG9}zh<
zY4zF^YhY=JH~PZeoR^tc@UBnd7?%DOls85`?D0;m9s7F&bPnKJb*Pq`0fO98ll+1@
ziP@TV&p8gSF8#Oq?8->g>$cNYv&;<r_i57o*W&K<LOT{^8(M+SUf(|;KfmxG#!4<R
zYRk@KPASLJl|pAqFQsUxzG+i`)7FRs3V8;iy1C4d!B^a$Tx6kI6W^2|pA{~4p1J+b
zp?@EW`W^q&TvgiM4ByT|Bnd%W+DZ0}D=<79bLG?$VtHP^W~_Cu=j`IC{H{~L-Q_%I
zpsfVp7kMwKy`J^2ju6W(29{S+Vl}RckhbXhJb56)RbcFHER+c~5VlD%!VcDjdw$NR
z3ftuh@IS!p^EApNu{!o#k5Z=gbR0ML&5e|Z#xa;vOK1#>gy{C3D)^cg_>X*>nah$t
zEN1@bnRr{oNg-&?QY@%z;yE&Ha*bI3%dKb_Z!;o1+Ox}Uf4?^_tJOla$;Lnumz3mt
zp5Yr>14N)nLlY#4Xbt}KNN!C6fO!`(a+;<fCl>}R<Kxk%=`~{r$^;!%YSEq-qt9~x
z`~E0Qs5!etk4N)67yH|&Sl+0fPwx9`M!FCjbqyUuiX_K@yDE?<>v0R%NBp{mMOpA_
z^}oKhtn-@K-&<_E{{ZdYBZa`REyY$+lebk<kBe6Qqa9iMEbTp47h+ssN%FtFsbXIK
zcV6J95Di=vBu)gI_YYBMG}|&FMs|_b{b`(ZCw4s{Ht_@oG@?(^Q&^;jEkdi>3{-J*
z-n@#R=m(igHbVh&TdFl&?mrJXNIuwJ?y?$iVSOVv;}<yNF5gspM)a9iRj8uaR3TO!
z{o_6N$W;gtcjyCJ!crjSXFkgGJk1DVizbBl;)QD2i`6)SStSp(pg_Y}{Eod4){s^q
zd?E!|gc%ln01o;qy^HyVIjJ}F%l=+<(&XaVmFLb@VfvEswz8f)qpE(KT3jc&#N6MK
z*LTjxNSclxEKJTgnqC3a0Zk_%q-;--dyLRCw{sqnm?;K}x7i+4y4?Xv1pRV5c2C_i
zDL*~tnUJXL5q+>VoUHcBYke~_ZJ>9I#V7P1oqcq={Gw*)-$T1~CigTpJ)3mF@<sZ6
z(T^WLKKDaIN86P*W8L=8N}aD7;`AML_V2Dr?Y;IH17zpkb#={{2Gk`2!ODDW&hoE5
z9+Bqff4(*Uy3wxrCp<2vvEjM?KOg_EQq5y7_0eJ{X8!Q^92#aK{fq<?O>=Wa1opjp
z^~&b%6L>3rPvsN&wY?5U<m5ssE6rd$gqs?N?N-H1`L(YDUxtPro%619l!98?d5r_z
zpDPw_SFd(BKg(&o&&MGOJB=)k6qf4U#V$+pz|EU1tob}z?u?AC6DgiYJ3eFw7w*=&
znWI_|gz1)0Eyn73Jk7cB@0VP%qJP-*fG2=>FDw$q{bp?XObjU!A;@0IV*i02%j7l#
z9|{EEHyH1iwQE^qsjBB6<ztoWhLYb;iS65nsYUI<>c9iI6mX<uV4TIjjX&WD{$tmw
z<wZdj+)oHZj0-RC0-uCzs;+>g;vX`=t5L@qr#-?Su-vj=G)<<GKe_VfvGQFze}yfk
z)aLSOM(%a`jEmKM&Ror2GiR-j`PKN4&c^PEzI|keP_B`+S@nFC^Td99`mA1#H`tMg
zltSKZbHpTuGYThuaF%xb5`;-|Vu2)-?>!NStJRo+PSq7oU1ERPvmAl^%6ToVK*CH_
zKaGu;5}Io<1yAq^wIyjgI4qg(H(h>Tlsu=Lcwyh@m%TR64TQ?vCuxauh<^6ZNKIN^
zUJKKj*unjnA9ra6KR=aw0^B_uIJ#LM{~J@^I_;q0qQNq%mdUpdvE4`xvXs&j;i<-<
zM&?)+167ZANaV-OrYAe8yx~?dBB(xi?on|1_S=(T!a>yD;nORJFA80uSJ$i%(bS2|
zJJ;j=Ao3;~<@(EXKDFIAGBk$RW`fw&#*OAQ-_C<1XR>>c$MNoQ^*<hDi(*12VwA2(
zpqnC0%mYpXEx-x^Q3zwMdFS{BTzW0hjcF!WjI5~YIcwP=&b`s5eDyjuMLE44H{6;F
znK%)!VcZ&kUK{u;lqFa2rV($B#MMr2Et94(02A0*8^(WLr8&GP@c_xS7dbzOg*Tvc
z{+6{9S_rA@T9sBC$cK)W`ec~fTjVmP0m8tq=lW?iOB0A()SidFe*>pR{pbf@pTk1N
z1wp81+h@<6YZc6|Z2vpPi)hu;=Qy>lf$kY}X;z%6n5~Y(mX%xI^bm@)&#7wBkbvh=
z;4uca%!2OgzzcWB+qXRAOz!@HVSnQQ&-=GZqc|IFNV}A}YOj|@zutZ)ee8&4+>R(u
zkEeTeW&)1bm)IRTkiklG?cPji%N`d7o<m26hi`4)o%prc>3P4@^XN>WuG~zQKihSu
z;XD~bu*W4OCECB`<GJ@%3@JW)EDKf<k41LX)83vR?nb(=o!B9R2n+mcI|xJ4hY_XM
zz1UeO0oi9pdU`No5LMoF=H9VwkU|)HgSs$v`<wNL+*WN`E7MuN%CCAI5NLrIL>rT8
zkqAAnFy2fW0rB@HF=`;<z&!wGhPvNV^?UD}_ouZ*);X^=$P1t4Rr;L`VpURb7K+wA
zeB?$;QAAz?0)})o!v4FxdQbJTk*kY6-wAycE=>qAGL9#8ai>RW`^PoOzhS`@@j}rj
z4k70S$eKw?Op-fvz^P(diU5^SX>UXBmD_8h_@hOsH=@q*w@i9`PR?yMquuU4LoZeX
zxoMU4>GRHe)g&#F8h)>{+zSd|bwLo;>Z{%9cTugQAa}#G*kky_?XB60pCZ`yY{`B6
zhpVa-TIf~OHGhmH?#VqeD1>8t74%TDyf<ef#ToR>2gFX{YGgvKj|1n#&HmZrPg&#%
zx2(^5lno@pjdF6d&x{!2CA{3Y4_W=meq&#-p;Mj>eTSXts1Ss?JNK|XiQ(f$!eFn1
zPW5;kcj!ruTKrRzwp=8XE&qavk#UZAJAG$2#n6X8%gv=Yy^<+>sr<r|Hxy)n-wClH
zDo2)D^bWrtT<p(EN-nsc-HqjIV`VV%IkW3{SzU5^x?Pi2Tj`-t9>Y2-{dVNca$og5
z8%b?}@smau)H<`bu6(SI`Lj40xU{S`^@>}7@@VUEn%O@bbHA^VMIBZ>ocAK**l!nw
z`3Ym^5npSSeqW?sZZrpO>_L%gy)(^x!2q4w+y_-E1b419nt`5x{#a{xT~JVvcSD5_
z32c}T+>m!-bN4?uIswx=wd!&XNmwCE<l}(TeFFY;8j^K6MnP6gpLG3f?LJ#Z1l7J@
zWLg%8Ez9Q)c=~{rPs1oJmOC^t(NvNpRBmY|g_hkGzObA}f_Q^|e!NRTPtV9{FG=SZ
z$a33-|44xSaZ_NcWuM(6KO^s|jyiDV9ed0h(Ra(@Z%f~8T|NCZOk+%~{zwU?lTdAC
zORjaD){NRCZTp&|*iYs5xU>nEYHMD2g^nBXQyW{$U9Mnz@rE#4=*Mb=JgB>=TIe!U
z6MdQtb}d~w(^~8{d|xEbRTOyTQ;#+*4lvNNQz`uWaA(2!rtIOc^c&NrV<r2H|9T}4
z4_9Y=d72}DyzcWJ)Hrg)Ixnmz&C1GR+!Px>pWEvSu$Ks9hBwdzv7rxN+%I+-GIcu2
z?)zy~Sd%$Yo6c&pXDki*#E9w$AZ(Hr7VOv-K&+4?Bi-5ZagJ-xY9y<)Yn$*umd!et
z*E}IPq*xjl?8#Mq0s9U7_sz#WtMt>m`5>M_q`tnvYnsPXYx{BZT@4N6>B-4@&KMXl
zfay3xQ-G9Z!V?vzWUPJa({zbgw8!s7W6wVgK2wIf57gWxBcrO({5F>>!?sqe>FjJ5
zuC=|xe|w8P)~U!|pQk@<Wuay0`7N(Aa+gN7t-g|ud@Km@^4FDmrL=x<Ms&kjQsP_W
zq*~`O0|O%e-1!$<wS7;P%dR}EBf)DO)5QR77=%YU4srzA1y!idnk5kthM!MY>jrC7
zWu-iHovn~}C*;YKhdP9UBBQze^rRR?G&jGACC?=?Ot*D*;<QoCuPl~7{+Oxz{qT2e
zI$7ic?ilXS8vwC7lAb8tWzY!^3r|oAR7B^<mjgLtLQp{F75b<P2AL3gozK>VN$?DO
zFd#Ed#n5hDis9zTVLzbZ%O4h;&exYl!OhQq<!-PtmW7<vAhUjv(C~DxZvJr=VFiUD
zsEvvpCqME>K1{*#2ZF77WJKFbc|Q2cr%=HibSJ{U|CK>%==EFWM@X*m1)U-{eg|Ft
z6+C=<u9xBpWT=JMT8$AZ`%8h<Sjp>-u}SP~$Z(X4R<TPg`d0;A$I^5~EZV<pH7lU3
zzynsHxzhA2kPXs!KKl7$>3LIAsh3o-TuN!L$c2I+8ar=d(t<>CY@L-~{8SgwcE#3~
z02TBmm##})3%e|$Ny`>>)pVB-rh;dxHDkYfH9&90l72qRb2@q-H+5x=zQOUvs3R<q
z59I1PT3avUvmTOlCy#25J){Kbk?hRUy;cs@`Yy}7oMKff$yoaV-!LIw_Kru!^ajz*
zo1rc(=l-lO)5&Zw?uhPW^9wm(HB?P^$)!b&ip4?bo`7Jif2WR#`P2;#`=jDbxg5>s
zvyLQW&}!~1uau7vH2hY!{7!~_P39xFze<y=QAjxd`J0NAO1#eW;f>-)M)Dx#0Q#D`
zQStPtrONJ^cSx+b-YQXt2Ded_N<s1`>+)z{|5=tMWA2YGpNTQrX0Z<8(75&`X95mX
zsj~;{?_xP(%HfrUHB`9gS}gBC65e8`63?KbNfFyJbSXnWxYv&Ern){924<y6NZ1p5
z<S~do94#k^coih47Ue}Gsd)K&l@EJpYGsUjSD7C;>=S=OrUBO>MEJtc>`?+aFfzg#
zn?^gy%i>ALelN;gdKEEDB&W%5{&Rjf*80o*4BhJtU9mMCE_=wAV4&4M5x5n{l03EO
z!zgO5viWFtrQ+<1s|aN})i9W8|L?=8k6-wsvCPp}KHl)@IfF-t7uU#5p)q)MHI7hW
z5Z&k8EhGB7puo^>YZ)0$l)QOaTFPr1^DO7L<-TX$t#2^h_rDQ7#{b`cCz9rakfI#x
zT@jWJo0}}~BPC(dX5Z`I6+x2yO!ZYHcHo)kORqkapZ5cg70H?l*f{ay^u?Ekl)J54
z6@BFW`KkV#k3?!TY_v$3w-z$@+kEK&E&*)<Ob$rG$+5X`cVw2AEegTB+rJn%$Vlrt
zpiujQsIzY6v34Yx;8#tXLoZ8*Te+g->j4V^S;3z*Mi-_c0=&9z4?8zYB%IHD0<Fox
zuX-l&d=C?VAC%dkW9{h{oPT)NmD%vWT33v)5l??@%6s(Wq`PtHZ0&08sZg`BCtmU7
zA))3~;#MSQ#1jjK1LW4=vKXP`*&Epyb2dgiR6i)A$J(@nyn~GL2hYm`Jw>)M%l?*P
z`^8if+^b0ZyANp#8pXVx?c`ss1mb`;@UW0s8MS5%X5l&=?<~qfpG5~YuWuwO&py0Z
zPlPV}tbhi_5`c55AC{0BnCiD<q`>N3#;!c*ivA+w{^RiQgPa_Q@*eP?#zcpau?%SD
z?oFxdbmfLj*xG2Zy=bujeSbQ1)kU@xFta3_wHWkY9x|))VL+@g3U*22cg#1MgQXD~
z?TivoY(B5<b9fEe6za{h+WkIrIn3wY-|L3$T&TIf%*DFMVdO$gN=hU&{+XE!#r=}6
z$UXhP)Ho@NYSOYqsu>yaq7);8U?=oX@~cTOS`$gqz_@*G-?HQgQqI5H*!qb0+3ibX
zy}Y05a(>l1-2EtlRGuF0!vf=W`{$3H;*o|HFKp2Sk-z=yFQiers#RqE_27Ql^&~MX
z@Oz<`DGAV!N^?<oF*2yH_&iDD!}W!!BPI<yD69W2+DFDs{yB7g<4U1%)}Ty4cY32A
zF81_i&r(lw$53iM=H_P>u`eAUV%thCJY)8Gm#F)KMf^b;QkZIzVC9X58=TA9=43zs
zTN)ObiLvOMFP}bge(Rb4WA4z@)S09JiR8WNnv`CMLBx)U%;bwm%ReN)!VR^}6`(LJ
zb(et_U>iqlZ0tKU@>v&7Vmh0JiZn$paIMF0ttU<QfgGI`)~ETw+pW!<f0Wo_fz+ps
zuDCtFi5*;uONCLycYJh~L_$NR&#u6KPS=sW(rx9bY~e%%LoVOSXV3dLOStThzxXRt
zp8h>lPo?P4kFD8liE4@I3(xwjYnGpXnJuF`J|<-#d6Yui`dCs!@n`9W-1Z+QhKvu%
z+z7%rmyzemIWZ9_<~d_l{WsEiFzzW5otu$qQ}x5mM4pX(!8V;+6VuV(Pv`Y`mR&g4
zi6C}frB%(X@1k3m&rJ)8i0Cr2-gBhEX=G<dGDW{l)70Vv77FNyv1Y(aUYFzG!I-m0
z|BK1&7W*_cwY<8SEbjvIoC<arDGIO4T(~1lzc0PVzB#H^Yr(`>Pw(T157uZ19g6<>
ztJ8(BR|^S&>vCD0x6L!0+Mivj$y|J}c{Wg8A1fJul(}Wunjh5!!^U7%?jB0`^J<F3
z6SAJ1W;Wn0aT^n6+;{iB#&dhFJbLxIi3KZhma7eeZl6cG(bB+VQ$uWD0WmaWKcMUB
zN!4xV*dUQ<)ylpP9=L%wL-hc(diL>RIt5IRWX0o|wn-tengyB{IPU9C{cUe8Dh}LM
z!0wxfZu8QiH?eST>b*N;<=%fbs##U$P-?vY{*ZT)(iV{T**?G1hgu5V--#RFp7Z}K
z!gPfWYaV*(qa8y8&UM9((~m25k#%**=1q3D9lT!9XduMDWXLn?XQ+n6=(xLE*s=Ue
zQTDy2?Nd_eqKOhMd+U7#e`soO#q`{58f~A;zh>H)8y);HGBtlJ<=wlkj!Eh>H)V%(
zPI2*Y?v&)0qN1~(n)9#hp1LVsbJ*1Uz7c6%yK;69G_P2E&7Yv4+Lz~YLy>ylz1_~z
zu?IyzDZf?MnRxi*o7r{YAq54*xa<zU>BDJXVtW0f<t6>N?%>1*FZLTH<%zdycR&lu
z&SOpjaTsP^m21~nAy|j|sOai)>R$E#9t`{3GpmdEY5bfiqR|?<tdbu3=|mijkG5z2
zF8M;zRiW6V_u?*{TnK61NsXvn;>Dqcp1SM)=?#;^=ksy|hz<4d=E1C%Nk-aZbnI1b
zZn&|KLFF;so_F6jRu?c{Nh6dJsN0WsAvrmyUxc^cc6`IW|BX-^+3mU#dtYqJOnM@b
zJm6FjiJ2SAf0VQuTxBo${d>MAYy3M_o!s{K-|TP=tdV{hx^dwfwrgPb8~|5%6|~N@
z@$vK5*-2uXPlB|wh?D2cixHETjDs_S%+a@nsPP1Eoy20KqvCxZzTNnLU43^vmVN(z
zg9aK#5>jT0>@pJBduCTw5h^1J5!sQ<tgP%E*+Q}_QWvR=5)u;SMhN*HXFd1xdtSf4
zdUfA*bve(^c)ySLF~l$yK-Oe-=S7%c+Xz_*Fa)}r-(@`&S}yS-^R#lXa<RF9GW>l&
zZ^c7WACNlY*Q75i|4uzWBpvb{{AgIp$SNHvAK7yYnspez0J{Kej<S$l^xetdTQcWA
z<XRiKrT>Er6@^GEI^&c|ucQ1thxqx!(O{NbUjC<(|4=5=O6T}o#yh6&F8=0W0bq7X
z-@fgm4wfz;FN79_aLdv$m5M`5WkYI&&-C4c9TW&0kk269^UHoWl|6Ql5kmQzF?v(C
zVRsA(qUc`WPJjy}zRybrs-8J|Nlf$r?p<$y$Yy+w=5AZ|m6Z)>ns?nqfxb4)bC_fv
z#X%&Hz??$(<X%6`6Gk_G{4$5Dvx2LuVMv6SK4HBDCc^#u4+$d+Z4;9=AP!D7NoY`T
z>~I#C?xYU(pM<w{bu}5};>08V^Q7aKm81MS#*h*L>G`dmN%iHAmXFK@7!>2{3TSDL
zw1u>xRz^wPgIqQ!z;ap|1ZX22;~m9BY>ZMB?X-ozL&*tPV|RGFapeV<z>suq;1qzF
zkg8*@Nn-A#pVNTIb#d@nUdjD5*JF%3fPeXzM<7f1%Hiukw%nb^o&IDu-4`Oeoj0kq
z_u6Vyz(+5B#ZkCQ>+1u^Nz>joKt?&OAzEhe)DD@-?b5$l*VatI^KmFa^I|(-0I{Iq
z!?XK1&F`*y9D$9<h4HR)%+g+xd6otFFMy-vRQO*-fCpTAr%xf(22@*4ItohxvFwGV
znSrWs@0BX=oT8#*?z4z7z;zLnb0sl9Ka=5+oe1OO+FFpbC<yiHJ?R0rmQg2sqm{+)
z?PFJU3=Ng8Op6mQ6)s&RCZyfP*8g$=j-G`o4H_GGZ0I8OS=q{pJdUq-c-GCR!OPE$
z;=jw>dV3TH3>mfQ$Z^y;IpwcIC)jd9^@3$4tqxapjE;Bvh#+L7fPp(xM^xR}XWb;J
zsK}mmT~Eu8=GlM~+q8IZy-awc4F*7;7H-h4FD(C9S$pP`u>V9Q_zNZdTnTs6HP@OL
zRpg);T-dS0O)|xnryQjz`j@f6WoMD&rmc2mY5VjAxXMEF!e)K)^WEy9A<EX7JEyO|
zos9Ny@lICr4`Rrl1$>tXN&6h&5nGd1o?GLo>zmPi$i|LkmVj^tdAB4cBRw?BRuBwg
z2tv4hDTUSTP`(!vLx@?Jj^`$BBa&+J@@((jiV=tiCwOW*4|pX5-u5tA@x-w0`uC47
zX|*IZC0QUR1g6zqYAeJ{2L2iLGS9x*djc2lI-v@xbQqEv%U*76ZI%4znVIDyBg})!
zD=SIw-C{1iV+8kN6NG{AX=xZn*Z>_Joq+dmPpBbs3zM{rjT8RA=WkJFEXY7Lzwk5c
zmnjA=f@ih``1%=hN6$W0<*&wZfj~;|?}<4^iB`ro)(#=a+PFhxWn`#7oJ&^MJtm6)
zk-PHlz><SoE!`w<l5J*nkvKZ6Os3L}0o9komlup!jGDvF1)A~r_1Y94Hyy)$2>qm1
z;;Hz!R3#9h02e|=%%Q>9ctq}grikLT7YJZ~dJCX(p^j7}h#`y1*o7A^)>^k*6PY3-
z`b46(4`X&ulwK**azq}L*7F-dtH~FPH4O&lZq&nLIdS#dMd=Jz$B2C)f!#JDxYi>v
zahmd+kUL<B5*D;w>~gNNKiwsX<S*Ragm05nDnttQsonE6xH!a^xmdzUZ{KzeH#_`*
zTa9IChy+bh(G?=`n2UKQgf1|Wphd0HD6kOt6SjgIppJvB;E)x3Zix31W;}UuWVBY5
zgN7ApffNBYa05hF@fqj~k|abY7s&zG6C{kGwHa5;vbGYlEbB7zWIoc)43dkS*I87B
zb#+smUHf-##gLX$jmGM!hld9csC2mFsl|PTxPp!V;n9Ji%|^+^ld|bQ7YLcJu`h>8
z{89SQE$l4ehl<?Jca#p>{-iRP?HPJ{{0+59U`y7WYtODGL3N(UcR`RI7wy@as{!z?
z0AMyTVRLbCLm_qKDHIrhR}oEt7R(yF+0PfoKRHUcVZdI4>D>-W`Gmp;nUOnGs$2s1
zNABSqwmPfZ1SEwRan#iG2y4N(4eP%LFr>4ybCWf97Qdc$uS#k(u&ni`8)Ej5XME|G
z+p`U*)y^Pi#j{Gn21Fhfrt21a<+tW&HFQL>I}i>_AxbRE9I3zF<YZs=fbeK^GADHe
zK!Kh<D+3kI6=L^6(xDu+rQQ@RF4{Zx4ySaT80ppFMIwd=sKr(15AQD<|5*U-VFUJ$
zHBjoGarn?RReaz}e}y7;n`qYFmP^I;FW~I>3d`@y7X<rN-+ng`Zd$CwB9g{LlNCs|
zAF$TV#QQ`Lg}8WMXKF))$4sB}Sld6iX@?<2!JR~48-m`y-xR{GEET@}Rm5`iZ)WWu
zIFVc2;5+nPDMyW=X3FIT$9Bs{Vj*l`Drg~`O+4fd9<A8cW*&W%9Bm+c<_x};#taHB
zqzIouYT^;xsvwx%2g^jzbw;wh-}s8zp?}ZP)#oS=O5PW`)j!cOtpfg5`#}>cZ}+wl
z7Y$i~i0%&Iko4CdNsUS30XPxV&$wudFlJO%(j#*fSHx?M&70c#Qe4jQLeCCZLCSGu
ztJ`qvxZu8*mxmthFFQQUb8X>kuSyn2nEZaFCe<}J-vJ$cjKyF`@d<|v^GTDJAL|b!
zMLoFVLbr#L><L6i<&4v}W7W=o^Z2rccs74u*0T>u7)d8>`wIH4aQ3bDq4Xuh#GYe2
z5@t?axv~R?7A{Y)Z6+rs;P1Tbz4defJ@TN{q3qfK9d$JQ4UD1VMT=?t8smAx3rRNB
zD>FE(6+t*8VqgsJ@OHF}3bI9VV?!VvSA<bPpU}*}z4i3lU0BJ4iu@?J3#h|2Z4Zop
z`2yXtrU;`|qi+DwNg~G<46PgWe8xiO^l3>ZBpW0g9iJGY<6^uE2o=X|kM)*nhoNLd
z)U})vQ=m5GiVg;pBgwKWKOYWWXz$|H+b)PC7!)w6C*BKgA4Qrn=tLx&begPmjwTH?
zHMK(`r=FK~0#w)@B!_gt8@kiT#XzbgjK&B&@OWab6!*WcwDUNX#O9r~gr3s3(fkFo
zYC&&QOw6$fmCu|fPIyFZ+QGglm}YK|aafsoqbx6dNBOtt{CwEl!463d>;g)9zRy?B
zoeO-H{pj6!0fzV!4t&;SoZ`C}EvpS@cT|eJlvmqNSCC14CBba~?3ncQC?MmGUpi|6
zlsvL%o?qGnkq!`z+=2o*{TQc3{@Lg6*(_l+6q25~rzD-k0Liada+j@eZYnRfErO^=
zT^m;RS5#b_UD+Bhd~qkt>2?7|%>ki1^_7wsb9zor9B_r5sjm{bE%LZQrT&dRb3C&y
z9VIE8H8CWE8H<MMF@VAP`q3=Lr}L^+qeQM`PZKm2w8MUzvtiD8!Xi>R){0I+Mu)~>
zr#(l@H9}}0wbWRD`mrV)&9RaD0Omy87D@~iJML2GqY^NDquvxTWnR9`!x!Mna(U25
z!Ps~*v$Ed2ag)>d-hYd3YgQ!4!Y)!yZZq(U!vs;O(XNb5WxI|=_ne}l(!M;qrZY93
z_U>)sUvb6}>@x!4Dz7Tb&nL(bq8~l|m|b^*xzT1L@PPY;e@pZ*i=LdN{@F0I@bbsO
zw+rS^y7+ad=?0n|q9~elF##r~2rwOC9$ctzA!S#o`k}hEr>i$5qp&a(;7pddcR5Hf
z7F7emifp$1iCkPfcTb6)y)^N4Mo~dyDjd<nNU#TR(;0usrj5b~>#?*{w`q7cMz7YR
zZRit^C<g1^yvYJ)7<!4(k=8q4|Jr{2%10bH$1lGZt2wO~)g8Bz^z?fmTctIV=>1>W
zy>?r-%A72VVS2eS4qZNoy%oThBH&*ztF=C_1xvYt&P37hg(*>hNQyX`R1gtXUSa%i
zO}%^7iDXkj=$*scC*ho!$)V4HVKT#AE+8$lMxIfJy)#-_klgm{)uHy7=<Dau83W=t
zZD`ou8@NS6qQ9^Dk)5p1LChG=_x2`<d(a!l_C2<uM3*yKY1`6yWYj*B_OpGM2({{w
zTwAKHtYCXXMw9T6o=HMmBDwsznDAZ&C(|0~G1#h3h6p$lhx#1?{|6xJKatusHfAN3
zoRN_Zw>PL!%PY%62a*%A3Z|xK7Tp9G)r}N0w;}EhFd^gJh`2;i=Py5mmp2VPzQ$BF
zc$Ej#;SPq<*yL%p69cVCo*`BnfcLYH^^^<^;}}@Ug(+aZ?YB!x{>+&Z=ru2G*|Lei
zbk^5~8XjQU3A?{Xwx_Yz3Y7)j4iEu#`Km&h2LFA0oh<0~h8-XD+JS<|v{2u7xy`O@
zWMTQXJUi`P-3@xdbdieeY1~3tV^;B<+2ztiDiGL4fUVd%oIrn7b#|tNE)*SMM5@^>
zl3TsGz6fI$Vm&(yzrx6_eRTFg?h&r&!&YHeKRDYa{S41@P*JZ8VI3{vn;xIk{r*NL
zQF9=Qn%eSs;tu35Hro@v`sL-7LJl)$VcY{GMSVL9qVrs?Un5~?j(3JG;d#f)th&O`
zwoC7z_0`}p7iZpylX3Ff1jC~>64%W0bJ5*#`u9eKOdX_+jgRxP&(kaDM73_=qOqAZ
z70e(n1R=xn(R=^7qeTWhur&g*gLX^luQtXqldh8i*Md8uu!7sqyGgEj{wzz)f{fDj
zqj*!JU&EKGhMV`hyLDlw)r!lEFh+oh52`K*#pN(Yv9cOTqW|wMXNuwgR#9TrH!+7B
zNL`WkC+`W&u;I}ZArig2BvbEIt}?f;VuppLj!qkI7)FS63KkfnBmE#FpEz4DsZnAn
z9qfn=Xte|Ifn|>SjSNA3a!^a9)zu|oN(7=->WI!Uo#CU&HA~m}RFji#iO#eMWs<V)
zJdKL4j$9F-Q`or2pvd%PZ(9K&Hks+t#BA*@Y`yF%(t9T(MupO}Z~6FaX#BMBq;BeA
zEvQ);d7P0Js0~VW2H<Cc#A8PpZ|a62>GW}->tCLQDao?Y+(j4UR6D0=%ZJ;(kAwu8
zgtIa}w$_T2@Cg#0JSMU)w9S9=7g6OLlA0YSo-Fs5uH+7^)|Q6V4z1gY?4HAvW}ik{
zDVfFpIbQtamjvPYknH+#_;s2bpjhHmLOwRgw_W%2Eit<Li&p@Xv@Y8hC(Pu~5NE_2
zXzKkTO2OeZS}?+8uxChmmQMB<>y{mR{UKs1G2*TMP-9zUPp1H$X8}royjzq6IL595
z;<`T5uYmPhpRu{&m68AhqHJK9LeyITJu#0RnSF15?I%^}p-n2$zm~XE4#leMi)EfX
zlx1ym@z>B9Uixyz9X0l%CtSa0I|y{Y{Juy_75I9x9Es?t=KLGYKeYspR+pZ`m7%B4
zIWU%9R&d8$ApWhVG(Z0?M9v$!iViz6AZNw8&?LK=uxBto_$-Puqs$xRAFL7t-Ke<V
z(GnH@sC}_Ca02@t!;>ByI?=e!(7vIIeZ+8>TzCuMco48*;`2#*eNOs$N5@5sc2Nq?
zHSYQ-gNaRhnEu?asi3@vQf;7!JnOHxwA3B<#J_o%ZvX!)0CZ8=xVzW9c~jo=eIe>Z
zC%X!AjbdUHBw5<07Oa07u_S!t;%vFcDvpMp)cYY|V7klbsMPLluit*`0{t2GU%C0J
z2C+Nzbs*2n!pbz~mghxP)GO0qgBO;7Hpc_5_zOGm8iPG#!rypw%S-2dU)ZRUBa_E|
zzFBD7xqbK2Y(xUvM_VVaK5}_Z1LOus7G7~W_wy$rjCac`s@M$;&kG)s%!OQre0doK
zFbR2W22Zcn&{%@Yv)8NP03^WoNP@i0qEYfFA0lo1S64}!-adW00oQD2L2i6E+cnXS
zl{V_fSy=%WQ{3-ExEUK8Ykk#u&Zyi*5YdQqn_|t|P~T!o!*udwJ)%RT;Uj@a<bYPo
zG&q6743eODVY=`qmGgS#9UQMAVYL4)Piq)q1!5Qgu3r6z4}Wj=6f33TccXL54-H3|
zic~^>o!)Ql^9HG0_=Zz&A2!=4o0S)2WJGHVDBG$4rr1GYbiMqszqBE=u*T=*KSG%c
zC<Lnd46-o!6PYBtdlEmrGCespys2els0QNAKcC5d#?>$sH1r*pIi2#h=*jeWEIB%=
z-NEv(SAOnpX>ZC>0SmO=-WY?95|bk6V$pA$!oQDc9fyDZ{K`?Q_pM=Jnn*bWYbm3D
zY;k2<pzMWL2iJMr4s0{Rdmz%J$5U2LB14o99b)*+Z=>Hn)W)r@Zo*GG>vmSpfwof&
z6|l$8Z)te8?FYP*z>UDRyd`hr^jCs}J?;4;Fp5(`<KGJ?9s@!~h|wgHQ^Fo<(jPTV
z`sZfshrs6-nnowJ<ln06sqd>a!m<fTk4Q}oH;9wEkzJ9s@1U``b`%FZ3UISn?3UUB
z{$h@B<i76S-T=)UHQS3L-n~pO$m{sdW^qlZh$b3}1=|!yZud_xOz_G>Yz1aj)jlF{
zrC7KDkH^oStE1Z+$F5q4Ft+ik$3;iW;Y28|+Vy2urOAu(@JSxM>n`Pp>_CcC9;U-!
z%0t+wBhwiA*vq;oAD;m~wY^KN{3#Ezv$x<f9+z4a|2b3d*VNQ>f$w2{zU4i-P9rc4
zYSw?RAg*IQnaK8#{Q3TTMl5v`rkH;#@E`4pR!0UC0z(JON=qlcF(1`7;x)ho3if8m
z7rgi4+Gz2K$QSQOBp;wwG11AS_5AuW+S(5(bm<$^x!7@D&7`ngFz$+%08)iF2(xzh
z-3cA>y{OC|`R37z+KY*!$Eq1sl1E14tWR{e#A-CC9=kGHe6BX|`Q_}A635O|&?fQ2
zAr5D7X6mL5LKg@21MlZJ`A~7>suha*AM+kjw!xG4kLKC+*CWJr-u>6JR~-kgg~2Wm
zQ%~zV34?WY`>yS=v+i&ty??KaNbkJ7;MCvqIfEYGgWSJ8D()WoKAdr2Z5vJlg0|=W
z583|x(}(xQaj1X>7z4|uAH%zb)Wnz>MXra3XoAiE_rSNNJ1v@1Zq$ihp=Y3{`P9LP
zf0FwybD>?`*l@tXd!1}8>rG<~KLCc58B;v7w<5n)o|=Ew<hQRnk??qnxxfU>f76Pq
zT;&65Nw@+)c0s?CafV%^u%mlY`Us7(o$MeZ*DVdLvmy7Tc4)`gFWrk!c8l2aDj}L}
zXEr;{PNc@&<vEd9SU9Dl&4s(C17lsJrDv{P(bSZM339zWTjZ3EL&kNh^4>bds2}e2
zOLbBx_!O-Q6f$)B2l}5s?@DlLq$^W=O0mUn_2rP{i;Qro7gbLdv;gS-@Y&^g>%i)#
zt&m~+oHD<KIPI@fOiTa7HFQGnQAN;h!7*i!3<iNM?bHv`5*rd50oY;GNF;wu?DjFI
zPb1m1J$%Z(Azo+8gFjDgRgg3}U+wG+v!Vwc^7rFSi-Du2<sM;iSSPzCZ2KcYzG-}J
zc+MR)DFantx!U65;^^Y?;Uo8>L;E*Ha2y4@*PlI7u;<;a3%#;seWB@@Rgw*?BwK--
z0<3lF$#$GptLj#nKZI2qUX-&>l8Nn!;J~FP^6+jXh~C9j;l8V@_pI)2*$ikMg98T#
zhqm9Y63HN-ZSrHY>A$H?>jPx;Ceh7CrJ4z(uN^_J?ByVh&RLj`!RBq|Zd5e~;0~_F
zU*B?a(erl0uY!imI!`{RpfMp9LnP1o+O86}SxW%IxkW|8U^|cDf8>DaTIas!Pxci?
zY?du_>O45S5Owi+fCj00V=al>6?E|WxkH4Os?f>GMR6c{5+<XcU#AT{LUW6uHJ8Nc
zgYo1LQ_UexS>v`*0t>@d_B5>|F;)q=Kl_!2|JtCqI1p+8mhmTC?O)vNK~8Cf$@T^R
z2rla7ERxN{%vUbvVCWK~bxuKX$4VcrV`L=oHRYV1p1)>J1`bFN{zBco`FVLWGpplg
zQc}ug+uiz=Za7`gw%6%BEbxT!iNWF1?26u*npc$+zlAs3UxeGtc|M*Fp>rQT$fr}I
zi&soOpqWEkWDxXfTy4c8SrdV9BgEyt_S^p&vUY9OBRDYSxFCe0raR>TbSH%0Wc8(s
zZ?_qNKmc6xm7`da=kK{xt=Q<x$fpU|o09XlY1lxF#NcKFl5R6d`Y__dp*8eR6D9wT
ziHz_N>eB*6(0PnLYG)Zm!H%)t22|7wjn4kQ(j3+s<=LolU*%Jnt@OW>C)dF8e2pxG
zs)n^Ikq@PIJ~C2Y(xo8w_9<B#Mw&c{9(jqrvTrDPd-_P?!4>(rR0?mo-|zn!IvJGg
z$70N0Uw|ehW|~b}I{a7Q4K%W`Nv|@SCtY;Ky`>OJyL2<}a1agckBqKQtj83tpC4t<
zdUZ`trzY|ff9k6sHXhtM%<u{MQVOI70zLU7){JS-@VZ18VMOz)It)q{(EO$rvTyF(
zzyC~Z2-IIQj-mGxkBf?)Kb{!9WnDPf*lOg7<FG77LeX@^Aj88w1v}Ed_R7T<rGUR6
zC^s;mcO7)via}Q9liw7eT8-wXf$bsSU`h_xyY;7pMN`VnpNqF%9IYA*d1=kbymQx`
zC%TbiV)*;Xn&Bjyr#2$k#E9qSqzS&=MYqj*V}3r(&L9P)zsF#_0M7WNq@)Yw9ggf*
z?(`x7GbQsjJ8Op%AR(^k0NyiP8`ZMPYAKPG_0pPYp-9HT9HaWW-@h!-ki)U+RkZ@X
zr<DTvUcRL9rq;PmGl5imUfP#XrjxNGJ3EP%0+l-XXbxN1D-8Y8Dp9xP*_cmUhaWBd
zC4*v?#a8Qvb~!aQO6QL8sAK6A2RU@fu2k}RN*%z=sH{)YE<yfYEC9Ycy>?6$FXfLZ
z92=dxgawpMy=7+mdcpCtn&Z_hrruQ6F}JJ)7^-216YU+;Y=IEI9nd0zN#9u)EwLZ1
zq_D0GUEMnPTfjSSIe3nyzOVNXUH|E@t_YgO;?TPB^>6PtW-6Y`qPpjY+F^Jv`|eJ2
zwn$~w3t8q@srJs16Q|$O=@xSVZ9Tn~$(|gdWrcirIr+H_$G{llJ2~6WXJx}W<`+F4
znH;@f)z{6pmm*+x-W~QJ8dH(jy4-w+6*5SU;9Col^WO4pk|hvPP@<dgVFV*KiX$!G
zFH-${g?O^>F}=p!@!i$aE{VFr6DD8%y;9ho*3WgFz^Pt>8DeRWd>s{-?P`~p8qFib
zIVXGOF{g*AFJLxsIJ+r~Dhqi#=_ySey!)N4xHs`cM;IBHN~z8Rmv&inDRY@D-6wva
z_hlvJ%_X1N<+{4rHVarXi$1K10pR}9+P4wN#nzmj!`&yzdN3WtP;I7qbSJh*8?Dx=
zFENFX{Jk_vWK3O{)E#!<+}&vn=ibyG_1|x5xwPy6(zwNswP@V;PZ=USKhvrgL=K`I
zn9;X8oM1knOU8?~yeB4CNHlV<)Yng|>~s_uCp$_cP3)LFy^f#Z$Fe7{X(`7Qf6$WA
z)@}=s6%i33Izy`LgsdrY7zw5;P}<gA2h&gsj-f)A>qq6y?~;o1IaqfX#tG^5B`%R2
zy-C@vz@3-N?q5Wuo0JdcPtDu+&k$^b`86>J82w5$?0B1UZx?NNJMJ^(+E9zv&(?6#
zxz(FO{Hk>He1*{#9_n9@`bxVm8)^*P(AV$D2CAY8hQuQy&o-033G1od>D&~C#mBp0
z$ocr%6xZdHfp_neU{ViOWm;jOdT6V!EnfSG%i|?f{LhJFwC|Ny8*!y9JyV()_1VgD
z@KF0v36+EK#FT>d>7Q-+<52UPtx|t$SnKG7x0MEOAMuTO<n;LTOeRI#;@lXO30NE@
z9%R9C#cKTK*IZrMJA~|Ae{r+w>ti~qk<I=V>MFRM3VJjSJC=G)gJ>xs28HDV>Q(s+
zS0Vi+CJWVvA3r-OGwE|?oy+0yxi|7T#j6%$MpW+(2i%$T1K@VWF&);{Er?L%>Khds
zBPOO*%&fEvdXXJ~t8U&y(l!oV$pu(mS5#zGq((+`E7e3QOS@#wHrP2hG(Qjx6^cz2
zoSW*C@aI@94*Sbd8}Ro=kcRk^g(8R97FsQgom-7~-L&8eqo;pmWMybwDqQj(6|dzE
zGhD@sb3xLQha{B#E$o+QIY0r&gh!(P1fJC4Aw&uS--V=r_MiTH9D9nZQqkCbm^oFT
zaiE}kAaS>PQrP@HmC~f#P#ae<6eWa!P@7w1=I3}<Vh7edF419loRMwXC~|+6EN4-O
z-EL$+wZ%&1aB8NR<`^VF=>=tXl54N)>WN$&oG4J{=tNY!Amd9+AYLPH=aDxG>(^5|
z+=TVmm`#APLkX#v@TGvGn>nb8n}K3@ZUtWPDzoR*%8wF0DelWP)z%Nz_jFnkvvI-?
z1@4igbcIpAA()1p4?o#<QEE<l;@h`R5G65-UzHEN@4HBcAcN*z-l0&Mlw$ioSsg3>
z5gG33Z)oa{sdF>F5dH$)erRd`mbP`T@LK5h(2vY5BbGvIx$E+ABA))>?gu<sx)wKl
z<fBC97u+K_(8KNH*?j}kVvX$3NZB-PF;+*qxDu;5YdH5oVD`aj5jHfaT-w0M0&~-d
zSL$K#%6pSn%6khF_tbU=nCAt1D_t>uY2zhTcdV2RT?>IE@)#C&z8{yvg!S}k;T<~`
zh@@(TJM_bgD?oZD$HV9;t;@DI45gAhj^Sk*5d;^8Jw=B&J!WqV9TPo9y*)@N^9I+%
z)ZA-qC0pBl_xF8mm%K#yOA;!0qFl*OD&<9LjGz?kfoJDZ*O@TbSRIxA&2F~#)eufU
zr(&%HuTp%RbFL~WiD+0()S3gocMs-7L|)(>ey`a(n5J&`xPOH1Z!3Vu_Gt}G9_+4m
z8S4Q4Q%RNzuAbhoVQaC9yo_G%B@x6ULUlXI;MsWeuEmvyo1{{AsdQUbnDi)0=jGoQ
zaIdn0<}LCPWq|C`nis_#ig-hv=Sd@$nkkY2hvXOs9OrLrW)d~fLA%e}aW=0(*38VT
zZLB@^w%!|yNf>~eJl^$lYwp*HW8q;>Ri`^n9X#c%MmbxpjxhpirI$9<#bFZzK$y>0
z^ylXH--hiMWAMoIV)k0m^UE;4qGi|yRzwdU`pMU}BA7sp%(|HzC>-{9|K<%M>A;2I
zw=%}TaOn~!AGCI&u3t=r%a$#N>OFSB+hboGDwgq2jfa-Uy?<Z)wQ_RMYh~+FZNmyP
zC?LBtvOa~w8TZS>+8Il*Z&v4X)|*HfN<g@eTnX6zhw4MDx*xw~qc}>POOZY5a^2-Y
zzFkOe1h|pS_Bk2#$>^<;4&~aaj<=o{+%n=srFc>?T5$R&@F+aTpyq`62u+oHr}bC8
zq)9e(K_7g2?TB!hKjZ&rC%!wjf2^myKRS40s%QhYguga-3ljRGSB}+9%Uw;9NY{Ee
z&}z1B{VZ%%akS)3qqnX!)7kNo<9YcZ`P>;&YULVom{2r;;$wTct_|T^?=ktv)vLRA
zUUJDyL5Hw1M#Rck<$yhmiNc{=5;56vGi9DFwb2}Ytldae2F!y@xHm#Tll@3goMJW`
zE+I;%ppi+fN=&YCK11lt&2OlnLySk+J2bsae|OefgST^Wiu^#ZyxVN+URXjqs0Ozy
zb2LHyy42<R<YiykMbJ(;2^B)kSnM>yW(0L9teNpflEmlu5L*#9LG|zR{iNmp2kANi
z7fCoc3KhqUiS18LmS@xXLbnGP11xF|G4DLBtsOyE4lEllR0+A#itK27h(2t<uI??w
zoM|Nw1g>yJgIo-+EOF-$nn~M7+%I1i-BN`UQk!x{q&(494$LlletzsMc%~DLy1A>^
zv*%rk86?D`Lp-dRt!7tYqNk9RCr+KId-iP7w{Soon2K0gjWOH}R9V3Lhj4g5a7YUa
z3sf%?-{gyXA_<hGKq1x&cfU3rXV$I(iT>?DvYm(T%R&DH1`N>3yx%xsprOQ~AU=!g
z2C9(1_QQiSq}F=C(zUgT*cb=~>Mws~Bccifa#)#xj+>kJ>98Xeg#k6>sE$;>5?V){
zVE%>tD?0jYmUTQeq%*>yue&?p%r|=jN}~wv$ZHbkbZuc0^EodkWh4kuUHSQrgpx3T
z-_r+^OhE~X?4(Es6?>QcXdlsM%E<Uk=>7XDh@wVOV2YeqUmv$e8tkF8xu5OwUSiiw
z64Qi|1<4AcNB})~K5%|JnW!A_L&$0mJE;CWoBwk}s~;_sO4a;>Ho<vd(0hFiQ%^_O
zUjIO8U<OX=soRFWwxg&QBymgn%gT`U{r9NC7f@$^yx6P>ZY**(UqLJW{N=T|nBvvV
zTl`F!&o!10&n+)C?t1bn(cwueE;&#cvUaQeDPjL}MZ|M`6b<_V^42Q|KQ9klzq7lT
ztfvdrsH~P2|K`n`F`e*kM=ImpSL3w%+5f#)@ZB3l1g1`w1ftl$yP>0}Pk%>AttN1i
zJH+?hRO{GPf*m`X{M|DVLGXlp8S@UU$mr20w&z{{_wkXvb2r;=1fr}T*u{P@=neuc
z$(gV`fPtkk1SW^%dH?s%XK$NE<Y{m56Ejyn5bXvB{o8xOn}@4yWmxI&1lu|L>JSX2
z%Y9b~`v|zH)m+U=0bc@@0bIJ${r>mKSF3z{M{Pt*1SrjxQUO^0h5<BEUI#Raip$EP
zF>}|^q2A&rlXTLT4&7$>$)Dkppf?!tzFlvqUr`F?3CY;tN!9GJv*~s~fXro~tSvGn
zf<|ZJy#{Yp_Oy_P_Ds-QQjgMDp0_}pra9;P+fTZh+M0YwZ)7&<oR<8&Lhm@t;?+C+
zTLly#jF;fe<-d8uh7Ars_`-t*hla-I=S5&d1^bnR)KqV0_ZSXLlWJ>+oM8qC`s0ly
zIoQ_~SH--1>2SD%?SsAO%8K-9ef_A<KYt!a@VquY>WA$288V~<p3cvIF+1zT!ot#R
z@BLzMa6B|Tyup3CtgP(v<-#1^j`@`}UOql+mvY+>w=bVQ8NzGI+4(ejCC+5cLn@`^
z^R379@NsTwZ5{)!pFLAXW_ej<<xjJB^|iGfA3uiEDZt|=At^OA=Y#!4yRRATkKL`W
zEnEVCR#smRgd!S$z|=v_og5!h{Aq8KH*`}maND?N71`HUqpq(083_Y-73IWMdlwe2
zL6=a5%i5rTb2ssil9SWpmkkv^WypwM-z03J#4oo@PMdGUuiBR=c!~dPr{P0<`l+JB
dCB$!YD~1%rWXN^~M6naURZ-BCe<*A2`+tAh7sCJm

literal 0
HcmV?d00001

diff --git a/images/positions_mi2.png b/images/positions_mi2.png
new file mode 100644
index 0000000000000000000000000000000000000000..42169647399fc68c203671a7b69d113d855c05c1
GIT binary patch
literal 51306
zcmY(r1wfW*(>82^q)3;9(v36{f^-ThDJ?184GKt$G)PHzgMf64lz>Qsv~+j<bFaJK
z^Zx9vuB!`I%sF$^5co_%5(AYO_1d*-7}8Q=%Ga*lDuI7bzk>uneqvf)fuE4{WhKR~
zT_OJZT%R2Y{{_WbO5N_-HGFEsKi8#|skY%iBHK&Liz6?iqT}Bn*ENdazIKh`nzWe6
z3+M6mDHmtWWtZiPt!5c-D()M<8@4j2g2q)fV+z!i*Lh2m+N8(3<|-VLOcNYGNoUos
zRwwmk46RgHaT~9;nNB(@7u!B~9Ahq9RvA3*w0WlR@zx!d(YHr;WkO749#G5TQctBY
zes%Zby@p9fO1gdSEPRySre8`ZdWX^%>+cUeN(@9d{@<_eu*XV!Q~&pScPN)QQA+;z
ztF}a%>oNcSe|IR`@TAtZYeXJCrTFhp1dvb?hd$!u6r;m!(J%cxv|_n6f)mL0#}Ow~
ztYp;uPlbj{q}9FLChY&dejCQX)KqV==K-VV#hT9ij<e+zK?WTuzw<|O;WMUe#Uv_H
z%<csDV5$FIQ$QFqcHlEDt<e50gT=gP-lRTtm%aqPcEwEPQqNYTva&K>*L_0ED$VIX
znqoJY{=2g|ngEZ5MyIdI$z;SNBwgQo9`Mi}WeMw0VEz7?`<#=LlZ~5Of(pBaz8{C2
zyuZw_n~<1zuEh_{SCo>k+Uoks+yD1aCU1S0ymZTc^X84g>({>2!32>XKUxo^SlQc0
z&3A@z{FqTR(;(hk>U~H{>+kD(Yu{Bc$aY?$p`ak-dA5Sh;kx?w?(Wiw4O{R2NB`&9
z?vx(83%R<wvUQM1%gZL%x^WUylVU2K2TP>mTwTKz7g56fX~snyid3i)tdd`EcE_-M
z?07e}QO-l_zHwi&;nAknp97nP#l`2ZUWwV-KHAyYG2tW^K7E!y92FC@+xK|u$Mp1v
zVKebUm2MK@V5<M~4m#%!Fy|fb&e`*pj1@#$qTRYtQCT@KJ>B>FH$`h}>+!kJdBe3?
zYeMnB)X$>%L;7rlZ1HlV8<>NG%B5rGR8&;DoWx4GS&kxuYX?0E!YM9$%R@6W{VPM+
zxrK#`Eq;5vq>eJO-CbQ>&F0*CY=lB8FIDtJD3LKqKiAa6!)HM$F^b2oji2iKua556
zbmkWqFDE%o8*01ot!--ZtbAEoqM3G{A3?u+*PwDNAS|rYge`4k?S~0x&I^VI;dh#K
zy-StITOYQ<V?6g1M)G-EP{1T3EBmddr^TFGkwnx_lyY@-HK(X3tFREGQbS%v1>eul
zFC#NEGA_;=o@mj?$@p6Fz|?RrLF2J_L`1|wXo|S<|6Y4QSo_)mHl{zDX3ukYM(?nK
zOC2{fR0=i3U%i@qNzO=$$-&F}#7z>JxtH0#!s(f+s;sLk5~FGn6%P+Vy|!2T*_%oY
zc=hzGtj4;!-sQ&q%Bq@@22UTVBt*Bgh}zoOm1-2Wb$04Jf2m?5L4$FCQ9Wg8&OL##
zQBGZ?@V`P9eO0B#)G<0rNF4gf*Y}vd>g>W3PBcyAN?!92H55oT2T?Xhe0wlH2OC?{
z$sPf(-4bVU{UI@2%IlNe#plHB$UY((8X7-V?0HEs=jP|7Wn}cHzB&|Wm%s1mK+0o%
zVYm8|ULjdPT0#Qp<m3b%kzS>S@r&r_hzL<8d;_@n$;nso61@MvB60{8dvNQ<^`8tF
zt`|R78B+TFWeRIHqpwFsL{u*)S4KtQ!X1-gk_mkK_U#S{k7b()r-Ey;LFb2g+k^1P
zNO@0BA$ofHxL*7U4QA?g>L}b;HeTLuerN<a`S}J%MN@G-G`qWd-}=6HSXLQ4iZ7~8
zp&4>!Wj^|xV8Y38{TEZjHQE1qT6G>F#q;(eUj9--sM=^55x01}EB6L=T=5#2gW;Zz
zkrn&h?bf@ISSd{d2Ayn#80Kn?dJn{c!orpt&VCL2_#x%talf~hIj)xmx=B}WZ~LtE
z$GEtFUL%9vPh3no4T&)^_-1d`avGks)|4+>^Cm4UFNTMQzn!v_%pQ^-M@}CIXx44w
zScq2>QR>0`-ysjkkJvV&(f{s@%e7)$O6b{09xdaikbN;apFgxRTB3Ke!8^=AGAum#
zlf+4#OVjbs)}KFLC_jI$-r!CY8XEfZ=TBv==kf=BXc-w<5wWq7uCBx?`4~wGl6rcS
zP|N6;Uvu;GQyUuI`XJxEM{Y&M$QYcMNDBABMu^_z?8Tg<;NQO-dF@Ti|7r2==uU1P
zF*2E=`2$#?Vkmy1k(R*{2DfgA{4~SUDBOG>QmjyHFjc?yXJu1wYr6jRYa>1G1ez(h
zQ6&ux8B<ew4;BK9g9A=3?oz86A*iOVZ{H}O^HuEI%LN9GnyaW3sC7k9%bLD<lb)LT
zHY<yU|M6qmEQ}j;LGM)Z+m}fmVh4U}77pIbyM{PIY-~^HaNEr8>wMiSBmbZlQdGp^
zE<~=buKrvG$*(%1p+Or7c}ncTy#f_RC>J>TM1+4k{H8AAJ~RqDbUCq6Bne4K&b}Qu
zL|OBONTiFT4@{QqRB9A5X?=}ya;$5uu`^b$cRk3MxDX=0xOp?K*C-JaS;5~YQ#N*R
za#Bu9>waNjAtyJdt`qOKt}e=l4{!RnzE#M6$jTa4S0{jkjJ&et1WzHmZW?M2Ppy2t
zD8Gr{{trdtNiX-{&=9Q(DHJ({py2&TQ-RW#xPF_PWZEu^)G^GOec3_>^eClm6)dl+
zo9y~87&kUHprGC@cE=iUClnJ#q5gX}Y)&@xD*E~lO4V?~IyqLB!%ckQUKFy=jxqX$
zIU9n~p1pe26Np2bQ&?yShdS~`+2z~oj+>LxiQ0<Vo<d_|V^p-X&tJR{Q&PHj^X5%>
zJ%gh^+Xm7l`5abpS5{VV^kUlen~lo#%Ez+SB{+%2m0pYtf4p>sM{A1TuZ6E0O531;
zNiHNuhfBrIey?6z*aj;$CMF!J4o+%K4PR7j|9mLVm$YHC(K?r4U$wd2<sWJx>5^gF
ztXI2NHVaaC3EUP-$!?pJw6w_LN-#}7MU0~Tdv{OZ?qYh49`@;eE-nt!W#hHEAVra6
zQPDM3%Kgb>?;g4%vAeg&XES#TUQY%Ut>R?N$B$2-%6ofzNrmhKhO!kSqoc(w6<CPk
ziKwWl!=s|wTE5GSkB^(0n0ziQjPB@=fC{OpsaaT9_?Va|K#vyCt`DR9r`dr#TJ*W+
zRr_kOXWcaA;NW0acX#cYu(p%d?O*37d!yAmo}#I2-Lb68J@G?7hlfioCb%!>g)e>J
zD8dj^tG1?WZx`oHQt-UoOGzIv8LKdj%t+P`nz=lzI$xmex1DXd-LMmCwT<kx?H_p~
ze|6Oi3Z9;kF|zqz)H4|lmv%_?SW8Q*`V|Y*%iEo6jM#x#1O)yX9tIQnIk<SHrnxWf
zqs-gZ(A-wT5B3|*SADLoE-fpI2KQf46^8uMu{NovfXTM3Tv}fKp|Eh+`MJR9*_o-O
zWmaWnG>37YvZiWxM`#MgW5zdgHm$R4&AM1V&DKx%t{0EUT3N9Q3JUHY9?q@Wv9s}d
zYZNZwq#XN_>Q4GySCNpQLBoFm!>jY=U4bTpFhLnj&Bq4^2S0!P8XO&cVqoyFGmQLc
zp~|c^uc^8D!qs_-SBqI?nb|M{pIMu4)wFBy)rs&GvSn3*@WqhuVjyjU5IY`T$B;|<
zz&Ay&Q&O)3Rj<$ilPD3KzYuWJCPd<)`j|Q=7tg@p;BM_TMt(9~)5>v@X>v6sTx@LY
z0W&UV|MxVWWPxIYq@-f$1G<$O@HEp43eaP_#h*XNL6pQ}*C>u`h4AorX>DyXsQu9m
z$92cY(b3TrRaHY{V}X<`mV<Vu{{RC(A>?kTsWV&I<>irIzNm~p=Kr(iptU|`LEV)s
z#`|dcG1JM@=E<q`4;UpZ(FdlSpQOpLD>b0;zsV^sp5LB*>b}GgCP&%>H_^}_1SMx;
zZvHtpH~#&5*C(k^dW~0SSA9c@Ua`<kWT<~>^{pM0IjpI8p3jr(J&Hd$9ck>F5A)KK
zdBWu>D*rhzFG#D_arej5l!)g49!dD_?ru(Q?ktq!!qQSiObnb)YV5$tI5M@^)B%%c
z`9E*@GzX@fxC?zC<r7tO;^BijUtA22j`nMvZJM<%(QUplJ~3fxVUd|$?*p@{2On)r
zT^;ec!ox)@Em`1P#Ma}JlH7ktPrvf@CLcV=r%!J}&$D-O!ol>nrlBpj*1HW)DkCEU
z3)OG5#*S8_5Ho+62AROg+M5et#aeNr1oVNUnMRvIi}v<*1n$(*YF+B}oMWNMV~7rd
zd9SRjtoFs6=o-{HCl}Yi=;*gVoX*^$h9^+n`<!2cZ{P6rJRMqgzC2lGkC$^lo0u_b
zeUG6kKqUF#9ug7~ul=8ox77YtNx)<17U^ke@0?~lVq0InG#@Lw9W5gcZL3g4hif~q
zUH{hk_4ktRcU<dvNk2%JYZf^-1|MBmv;WMDzV$HtPGx0fqOeyoTokmGuKxZ`{Zctj
zyREIQ?83s(r6nVU6rtXD4r3{K`OMFsz0c3xq3~z~b#!#d2h{ez-Y94gA~5bxPAx6H
z7n`3#l@RT+w-g>954Z^xKiE)z;`A&3q$SVZazCw*w$X0WePUe~0kUL|L-k~@OJiJI
z+(&?948lAd=YA{W`rSsuFEQwJ>`Ub1xTm6}prJu*YHC`bR{BoSb2rktUVGoQo|Mli
zUW!UoQ<HJPBxFR}EqXcSib!33?7Us){Px;NapS5EcFQaqS(v{yFVju>zo@x79ghGH
z!KS(*Y58;?DLs4Uoko}QVm|-y_B)X}ZPtgat>_pS>;eKkqoZ`%TWsm+=>V18g=WOQ
zz0lSMW>7!3*Pf>2EsZhfz<)n8GZQZS$JCU#n;Wu)MIotxeLy6wLPU7Hjm^%N$v{&d
z_!IUH4g+Ij(sp(nv<it$M&<G-CD7{zI#)`^3pg-KZoIwL`QiTjWQ`p<I{L@R$kw^`
z;4Ue}&RKf}mqfTfMXS4|YEcyxoP0SfU*~Tyg>^oCIjl#D`8Z7Il<w?kYkBMHax1r}
zXlX!_++e!S<=TrEFRXtzU5Bb~0Ys4{xXnr2+xC`uQ^>6H6Yx51_Z{&sRx^EYRF6+b
zuDrS4{E|CU^@UR^nu?{6{Txa8FG94D99lU*Jn_;}_75q%3T|$X;Zed{=QP8hRZLnq
z-kA>x3BkgYymRXU%jb1API<c=ylVcW4=`4yy{>NN<>v$V(7UX2QDeb*EuO9f_?x)*
zr}NYN;fUyHU>4j^3+d(g!7)qt0JPyOH_uu>d-m)J3}cmihcwary1GQlek|^Zg<LnG
zlrR7}q>i+KD1jWrVR~C^wN6`0>ow&WSy|8*?uI8Nc?_?Xg@@nc=jUfnV1_gKO~hf~
zK7nrTuvuBTom#ysSF>(FUS6xf)%hNbpsQHmnL7)SUOmvly$4J(s;lE}B4a+&(NVOq
zDJ8|2%lrQQI~^TeMqOQ!mFHf+dWFeA#Sxq#V122>W^nAIMR5N)cXMBGZ{iZYCwdew
zSD@oXZu6d3A(^*pdwcu10Pi+74$imk?x)2gfW~=YQfn8-B6C}`SXOa7;(ym_X?%Hp
z0-sPxQ?q-v)jv1CnD#Maa5SSTHaWTQ%BMH(sHu+@HO#H8KVDvX+S=N&$M?j^8KLf?
z{aIV9B{brC{_<rzz#C~r#ca5I7;iwxq-111?S6&MMr2=0&&>Q3u=Zojhw16-&c?_N
z4h{vcY9fOC@J1uESc7b8Ucqb)LJiLF81SD|y8bIu`zbXwHIDE^oeP?&=?Xlm{?P`H
z$6jYu{XZrqx{Za;)MHq6`ruOe+>hDB#l^|pZeO=C=15?6Uu4vYms&C2=OPFmm=+?_
z#ly!}IPckToIk*n&iShfv{{H0;T1=By*2`*EF~xPdZxjn$+_}{r{{H-niG=>VqEc5
z7|c&7p74?yU0q&OR95Ze`jb7NrpEJcZ3aT*;7|#~HZmcSBXe+daP_@sqnC`(k6*t!
zHm7P!$IWr^aGxn_1Wj4e3JNCe@7v1+1+T2FrDtaPK}8Z1ua>GYF{zjNMhs%xk9SeN
zeClyCDD7i>yv<lpao5N@B}_8jCSX#CN;#UeyQ@*?IBtU`*<ycWkAsRNEvp!rIQR5t
zzrbD(hf@n$pxA>Ks;WTX6^)FRGA0mR915fz?_QqET49xmnORmz$ww2;@6a#5_4lVJ
zlhL(F5}_kKdh|$2N-A!^efb$(+pj{+m5q(X-0X(8&ALyNaH0KJ)gRFC@$nIc?mQ~_
zx2cace!!G0a-*(~|I6Y(efnh0pY+<$5UNlqe;B7(&ccHE{{8zuc{qOzOe?_Pg7<QK
z;pw<B+1@pj4ZO?TRTpQEpr=!pO-`U8jn21K&n)lFleaKGrK^b0ZAQ(xp$AnuI3x->
z#T*_w>JQN22j2#E2GEm(BeI~NU<b7%{9ca{2YnZPr0ab*0kXr*37^&#-JP`dp~DVK
zQ`6b{q{s9u40D&EB*v4nZYJNlcjh|>%qp04YDk|ZieUbf?Ho$l6$MV<0&`5g#&*%r
zN>M>!wzN}7*3yzp_@t}6Zl^;9V1ine<^9%HZI)NO!t;5=Hcqv;_P>6<h)hhh9e)1~
z4i7Xi&;V#ZUETJ%{>9=9gSBq^#>0mX&lhz(WszR~+qs@Yvk@S>>*3+Ce{e9cWy}7@
z(SVH*-a~dy$sNtXP}bD|u?N?1<R^t0NHOR4PNZaI4ORv-UmF|agi5$<9E_M+n5Nne
zJcwb|a&6asCXpWAQy>Sj$e)#!_)!HVTqb-*Gy)BJCWefeVZ}FZ&<+kT%x6k{ZX(}R
zR#TI5ap6<Qj{G~?p@BR}ACSu}`D#6T<LOgvm7-UE1I`C8uLIhY{6H#_KVaQof21D}
zoUr3u^U4F@CY#l?djWnvX*Id6<Sn(-X8}MK0BHJmFU~cbZMM!@%Z(`%s+60LJ&K5o
zjOVcuy4t#`|HkUMW0f;(_K2fgt<keJ3J=M*8oPX~2aWaoT}X(Wt*tGh2(gVYFfaxt
zCOU)gx(8P5odw7|w|&UB2W46Bp(Fj-++5n8ZH0y}iuB^&=hzAp>9=jopVig0#zyk~
zA6V%RA0D53zIfs;dc!C<zPVXM<5L2k{V^MK!v(DZ@2^T6Q5`o62pJ9*8S5W$aL_X`
z0lzX%D|Mm5rUY6j9v!`~ysWtDa-bBeLMq_2NqTm;F+Y@YxgVA5y3^r2S3ZwuWp(ak
z9`zoaoZQx%lY9zE{NiS242Fh=QBhGJ<9I9iYK=_-IE?!v5)<VtEjM4PrXoW(Ob3nZ
z`*#_5CdL9PFf87`N8%)AhLd26<<)lo9%2eGFcR!cZF$C$Uv4A2Z@JCPlV8q~7mdDP
ze*8EgJNeWAet&n;uD{gn$kK6Ryz4y%Ns(GJb(>^nR#vI|3F_qz`BgY{l@|<tQdBac
zq5=R&5W|0H=&7P2PiCg!`c!SKamocdOnKdtT|D=Le4R|_%GGbA=KcPv!E@>F{r%y{
zXJ%*nPbX_5P2NOz^WQnUXwzp$N4n{l9a+#L7x6rzRNaCQ9SIcCX5DLmq~KPFSJzC<
zbC!Rkwt~zAH6~_bL%FjvqVtEnL53m+@4%3;1fNL_I7-Lb!Qj-8&yCRtb0(cH=t&AZ
z7I$-Yl3&C#X?_WT@%DLSt>)_Tj8&65bHId#xMF!>2?Qi5%{Pq0TwI7622G?)ul0T4
zWqWKQze|K?Y%nsJ^Y60jBa&QIoH(d{Spt3br>CbTCMFaQANuU?qw|#rpI%O(u!?3X
zhkyDc1Fy<!Us7?`hy2{zIK?9dMrVR*U0Ad30*FSD@$uq@iz?c#E7TxyHKmCM8wv2d
z3^44D*;)3wT!u#UARG{v0Z`EQ?*$O@>iGEksZCwLN1WW;zYj3?em+;+`I8<N5g$)|
zlkQ(cM+0pE(974_T43yY);|Qb)$z7*f+g5HIt>(M9yp1xVEP-Rjez1y%AbT;ukD{E
zSM=VNHxZD2nQc4j=_?juVq%~t8AU}O>fLaFD<WQ7OG`_>%2O=`I@|~N25F{el7Qc5
zZE9T(qE)T$Vx%hLHf;N`evIbYTZ2E%3-XN!C-LHs=BMbG=w*k=l~Ct8)HrM(lam!d
zX3jJ;grhw;H6=)Wyt=-=u)i|&+Iabyni{Xgo}>5M%eR4-8XS0cZUF|OBt%!%&<K)9
zkMX&9$mukWrUC3A<!o}<d8eH~z0wS?JBGP-?h8m8YHHtj8g?)uB9jobb8=ElS~{1M
zNM8a3@Hf8-;(s2q5r<-<BY{{2k{DuqyKU9WIy!RCob)+4FN8~Gtjhd7=j_ZW{bAjU
zX|=V%o12!eSO}m|oqLj(s)g{@CudMSrp~11jf}Jz-2mM+ePj(<2M*(YuWz$1VewB}
zUVt|Ug@1Zl+sw=i24C8jFZktSav;tE_FW)Y(6RY7<bT^oM8EXU%Oh$bp=9|49voEv
z>ns!4Cj2NUC<W@}NWo(*$@Ok7C(){OjLe2(Wrk4ZT70+zCM9Z2YD_WoLpOciQXt<Y
zqr$pV;(qd|$7rwO>|SvDo%f<94&y|ATCw(yj<%cUe9-7JgpZ=9^B-^AUa{w2>Q7Om
z_`RHRxsu|%6mLw&$e3{W<m!Bx+}PRKS--Cki{oG*vP35H0Sn8A`;Q&B)5$MBe)xb%
z=5=n9?7k}v-IPUhmLO1U`|N@+R6?ER?|ccUea1$Z0Y?}BDFAsQLc$cJKedD^M04nH
z?4n|0!y_Wz!R(|IB1_T^Q^`+Ko7$Qu)~UA==#i0;c>sNIU_j|>ts96^sinE$r>FcO
zC_7OxQDsKGpYku?3Sr&B3U1HZ&pc%-RFG6q=<A$z0$CufJy@z=Y8g;>;$x=|pqBvT
zfTm6hvUzrEppb%ByTEK~-iH@Ts;d3pqZz;T^(7BfeoCYJ<G_C|B%J!pMB;2}W=sEQ
zbLvl}E0n9vaOu~CDVyuRKqIGBOnyDvA`>EUvtQ`UAlYTf3Y-_!`kBnTIani%Sa$$o
zy&o|7?ML9*EX!K|t{JP7r}0Ey=<1}i_^Sq`clTn;#^E#dnrBGQ|0$?|TDivV_QQ<{
zz?V2LMf}A50Hpu6rvVFrSLr1*9!n`TuKOVB4w!(x9T61;@*AbY!9iV(1<7?U@BNAj
zQ{&hWMaqy~4LZ<{q!blt>FEQZyZ5CCE6U4b0P>f}9x4Sj1rQX^#(c<!4}(8`JaLsE
z#>2xK`DsQc$eaW$Y;a(JeZxT)gd9Ls_!k!!=A$KVL(o?|9NLc#52p?Y*|;tW^7BEg
zXgwcGxgeUg4%qU#Iy>Lb7G7+;sJf!r`cl0@`yD~1ou<e7YV4NjSy*O^uTG2A%U*X?
z&^mxUVrPfrk9j{WC+D{Tm-JscAUf%3?<(k8kKoZ{?%N`AG&p9JuD@FHRzQbELPi>G
zbRq}aNnpF_CQLLrx5g)IY&-l$i1rj89}n}g9BvGVGF-44(=Vc_#`5x;N53h$V~d;2
zAL$I)5#ZwMJB^<F_BpcQPm0wFWFy72TOVaV+?u)D)DcReUi~bHsK~X~<MF0F74~er
zak8HCXg4@@2(gu!`92t1hZ|OQR~IDTdV6pGexu{|jzH&}8bsM*m*ZK*?#C>a#31At
z3+)jhrr=kHwd3id%j>mMLL0I2ii&p|&lel_TF_WQh?BIjK>-q)gP!wuygA+TY4+|l
z%FM4NNm6hcd%16!32^O8{RvIFfr0t>{<X|Zj<x&uQT+hOmXqmfr46lsyAmfCNRXdk
z^s6khvJyjFNlIQ;JiY#lhoYRE?tDjR-R}4Ny5qUv7c8%&^px(s=e;?t#|o|k=!^*P
z4ZRgy67Y_kIxfC%s#Ixt;{~^aCW6J<c>X9{DvoEyJrde82o*n8_HW&om)|_Uyat^d
zl*wNO>a8SJb#Gf|Uk~TK+${=DpkZbXZ9GZH&dtqjF#mq3_w?!0^Ch0f*_q3Y851k3
zPRpuN+r_-#d_v6L9Ffoej`ThlsihhgfWCj*9%v1rI`U{bIq_&#SyE%&+5YX5l4#?6
z8!NcuSw5GY?n^<#j%afO0|R?|``_05#cLkVaQ%Q$H|%G!mc1EaIozD;Zn!v8!u3;Q
zdY?vz6YU5s6rjM0pTJGHwZ)^e+`PQGt;@}=A7)(pM@K`WqhfE~FoNj;lGNV#o8p}{
zovTl9zW{^S6M8{-*zMN2jF_9#M;LWU1U$WUE_+gPa&@cR{&&fjfX>lxHC`cff!+k3
zorB`5>mb_tit0&E{{<{@&sk<#mL!z(hvVdE^p&h@1?`rij}8tHYOjQZ#8`#P5SO1S
zo6rF`Cv5!ut`l_=T1i^f$Wu%Oubii%twG^hUpGhG-K-;GIKyny=Ol)*g`WOKgIakq
z3r{Tt`Z0{$q~j})3Qag=sIc{5n)LN4d`d_Fgp~^Vqre+tj{mX%&~ty9F(bIYr0^;G
z%*E!=ZeMVM1V{?8u})(XpcnPt^*VkEr5h@dzLe@O)IQR1_!9c=fLSG7km#$)V;knb
zxFL?w($gU%jZTdTe{OESdTNtzCr^uoz^ZNs30}PUbX_25P_On3KuNGZ-8Ti$Vn~XK
zOcRL${?I=p3UrX8qFMs~R(9j*jSqd8`Z6JKpy?TLwAb8-;dGdpt!5DMSMgV%`nQ5a
zJ^)QrCNr&|VEu~;nb*W^HHq|{yhR6S1yGFmKzQz%f%yu86&PO7SvZNKqn=N`V2DO-
zoDpvD&C^zNaNz0}JR;p{sP)lhYg--8KW-I1e+aGotwegft$^&wf5$Mz*Vh?z2f|(F
zJGX9(RP0WL{sN;JQR@>MMnhR^XvMnj?}OU)Q;WV>ru`68>d4-sCn%o1`L?O;=hv~{
zj<ukPQu6T`rtvOeH_j-k8WX4l<7E%=Z5VLz!07gaq5KyKuug2fDSC~?5*utW;rSFK
zaL_-a2!Yq=0h@wuI~#MrkPtXg7I<clT$x4(UKTiLU^5Ngt*>?q{qSM;bgdZs&MlXd
z#n_cUf1o=h3180Bo-ELkCqB0u{}(h@B{%D~8Of%Vm*W(VxOL}6bTAfs_Wo|)UY#%a
zkgc2x)-}Sy`}IrZi^W6-P@CKwHQs?pnM~#0Rq~y+)l~>O(5?L{Ou*JXI5-@DrlGFB
zQc<Y)u9*VLC@Z}_Ns2Q_C7+s|GF<HWgFqA%wE`76BaW;#G3eYz<u2o$b@7lmaNejK
zAtWQ4y*%4829*sUJ0>Y_Q+OT&90cd%S^s780EdZNY5<zh2>PCD3n})Q&3yqCURp|O
zf&6O66x<B2lh~QLolbISB>jIbJsVR12f)|RC{+2EaRoe1)GqEcDhCxfU}B?z`ueZm
zP1lO8rh?q8w+;MNV_%_WgAc$^`RN%iJjqWs&ksm)8yi!;e&zoQ22yO@-e@w;O2t*|
z@0$keArSG}nt`4nH5?>hdoWade0(atR4UE<c)~_Vn!?~>LoHFSttcvbLnR*?C!m&A
zvr7F+jleLTwBc3D?*}O}Tgw|&UftW<Y+-_jgtu?s#@Xs#>P-k#l+vA<t_M`aZ8a@G
zD@<M`@4dL=e0<XH)i-l7<wYwb<T<{PK0pT?9Q0I>S*>M-2<JAllygTbOgWbOlIDYT
z&awBF<jEV}i&BDGHP&!pq3yQ8wOQ|U-)oP@>!HtI#kNiBNui1XyCyhO=;-@5#mi66
z!Ks@cLSlWd3f2YB{?7dT{J@(1aYqM``f}|RoZxnooZ?RK6YRT`(UDp+CT_mHmURDq
zT<;PQ9vMVf09~n7m%M|6DE;y^G@aSm*+4Jg5UHt@AR55G00;)ch2qzWbCxx<A|*#!
zJT1c+{|^+!#}<CkY#73mrd;H~KiBNz|7U$2jubM2URM+ohrZ{^SLF~`Upy`~fYd+?
zllr4&A;hz(^tuxIV&#<p`sC%Guw)Pw&-3n|y=1_9`@|chYZwBe2-L|(5~OZisc`}j
zwNsy6#|B*c?|=T#8TTd5RyBGAgQ53aQ&SdR7pQdD4S!uw8v^!tc|r{E0+6g+ua;zk
zB>`+Bl$O0-i~ZWulLeo<%1TN$ttpE6!w>}_A|ZK5irMd4n*xYE?~NG(7BuRb9lW$s
zybA~UT}z8;_Nia1o`$BT4KFG0`Um>w6mXqJ<wWIU?(Jv(qLdPnk~Vk0+SE=t3(OB|
zT81UW{K9{CAR^lQ!+A9O*XMz}ft7uj7a)?T1mDi@4eB*(HLuhS2?ixH^_LlK%zYSo
z@U;~Zc@^~ZuA|{IRj;#W4y{xlCA+Px|Ir2E58=T?MR8^ekyMF$<6-)bfd(mIAn@36
z{WpYN^ftyT0X@->x&EPqHb;A^?XizRNJNwgK2_hlxtM3ThI4IRx%7HW_z&@e?_v4_
zS^d3HHDk0j?E$(Nna92okmV*c*19KVX2eHRRoX~Oe{riOLx*{RsIxRU%7HXEmQ^Gv
z{X9Z9R|X#}em~ea%>2AMSQ*ecpIQYn)Mj#^sWr@rKlA(M<7Fphv%@_Svrdm)Yc?u1
zm{9^zLQX&k86%yMW!wr42`O>y{nV{6^k(Z11r|EYV$pxXFvb0gX<nbngS&c8&<H@&
z1C*>#z#^O^`y3SmsnsyQo`#{iPW!#21WjJm`}gk&4_#N**FQHje6Z#f*lAdu->0IX
z`UsA%ei};fuxJExASeZ(<jKae)UW0R8I{LlkZdkjj%d&O+!gvAE=d;u94%xD|GEk|
zq+ZAVUTxq#Ud;lbs5@O{T%WE_5Og~P-tL~#Np{9W!fmeG{O)!nW(#Pi5|W+HtGU?=
ziHw%H&*;R!`G-aT^t{w*i)`9$OAw*k)!0&d9U6Oe0APfpicpT>Xe==5oS=4fDG&+#
zwX`jB$LhVqf`<%B<DexOew!z{TT-M)z~V;)eG|M63aro2&tR7FKd$|=xvKB%9H+|`
z0GR?x%In!fT_6~IW2K{bD0pEI^6t^bA%1y$RtF83GwrJxYveNknnu>Z>p~>WhAqzi
zsuZ-36VB`E=&O)#Tn7^~Kv9Z<kJ@njH}w7c(P?SLGwm=np5?v(6cEFt-7`lm3s{Dn
z@jSlh9V5`^JGY?cC;xQo-f+}8ACZ7NZ`K7_R|YH!z%!~-wBky@f7A6vZfEx=W_|to
z31TFCd(w5xH<$o1vrF1Rtukp;1RTwHrGSH3X1B}&X8%>0$JqoRjQD={rF$R=-$C*L
z3J(^hjFC}HqFp;PUZ7YGAidbebLxWw2apbwl}bN?i+ttV9!#LvQ*UW$x%9i)=e3bh
zne&c55%0f&fk*sFWaVS;`c-`<<|#K&-(sJ$6a;moVcmqB5%Bx$Du2M^r>>{g7}eWn
z=ktsU)HHL}{PDt8Lb;~7fI`+C&k9t=Mk7fRg`M6G3=D)_sDb69s^fv@7Lg=L=CKeo
zp~XPAed5lPlT&S;5cBr!bs#h_fZO!B70)Ko@O!c|GJHYVhH)`9HH9z}wQ8O0E?aJY
zF&`6E`PJL~-~|I7lFxPE2}699+mxWb9=Ot8j+t|F5@Uo*y&iN7N|XDfoHJmeR_A<=
z%b}7~&@tp)i>STbYo*eO2Tf&aOp%vf?(PD2g^no@|3EOLZhug=?rih%`EjT4(z5Us
zABZ1@vn_s5pSy02M{c%AxOTr~r6kqWiNGjh-Ogn72=pfh$uY7ATI*Kb9wwNUvpb!Q
zPeepg(G&lrGDJX6+sz>p18|7eQ|O+Gc)D1+5@3dZH0b2u;g?D1`**QWk(5~8lQ{a{
z4;~9q%%3k$IU|J>PE}oJQR~VRC3@#c%Ln1#9JUncaOb{-KXev&>OS`tx#+6}Py}#W
z5q32AR$~?*BY+GDnt}l*akj^0HSbLun|Z%G($dn9%qX#4ltEapAm4O$vo`k3hY2fy
z=}_>&3K|?CIz%I$8mrPau#*R+{a&?G($W$%bc>`97Z1i)>@j+dd)wIdQ7{7|ACRI-
zF0P&wVJ`)DZqBMR|HmNiTY0Q$o_D33qqrSyE-uFEEPV@S02a~;m4~;~uR~${gq^m*
z7o0HXxxnJRGjQ!=rx4+Z5gjj#+NnOs>b(B(uM8&}&uj#mB_IrFvT)bVGkMe6u5NA!
zQBL=BN_hAV15NsXR3haEDXJDwGjelsD7d)r>b3hz4LT9#d1>$nH92?fR6R>QH=0EH
zJykE^W?k#=GTs~@_5Ats=XXy$2lzK}iHyF=*WP-kjE`%y%x>fP#97mcNgC63IM2VF
zsP1`P$H2f)qETq>ap5rSy2gx<>E!6Y*Tlm!9=o`-)cB;?k)Mxyn}%V%Lq`Oig824r
z08a?an0r4`qtFPN9@OmzoK8i^4!I9ctCx%0WxF@c&&?gT(F!$>SH5w%x|llOjb$wZ
zlEF7@EjKv@piDM~i4qr8-jt9K)1S|2lWBN(*nQJ31+?sB^f)ZL$xNf1xzle)<xpNW
zhA2BbY6Xyudjr~0(+1gVAUtZE?^zVWx?a=J(#D2_AYLlGe7$dLi>g;!ssOVff<toy
zIeSQ&3cG!HL{i0w21JE9w__D|5|X*Ol8KG|{r%s2Cq;}+#JjFzHE#%{52;j&wLOL{
z+`hU&1|_2;Cjji)Zp-;q@}}VlSu<NO<qy6$HuhLwLC+ekaM=^%*jMyeea_>+)g`;S
zwsx{%HPf<uR1Om9606!1QPCEE&n@Dsi<gDJ-2V_h`>~S!{yC&>f`a_kXR5+ZPhBCk
zX?3zJYM1Tt1D+OJd=JbsYhWzj;uU*{02;w9bWi^p9&S&tY7m!^VE~Ih#a53XxE*G<
zNTEu9PU6cuHhW%tf3~(_vDWwjVZ(F>3!$z^M-8|@K<!(0q2#EA*Pn<JAh-tBR4|AD
zz;qEZh{d{v3@3!mpxBzXjuiJFL&%me6bU#agmB^UnRAl>(JIHzd!+!`olyDE{hCuK
zl97G?LS43RmmX)<cQJSUSlzE9BB&M~Lgd3-!=Ia4{P0vKi-3U+l(G*F4s|9h&0GUZ
z0nSiL`Enn~v@&4x%b!g#{5~4`HQnd7)zJCdJK+Qkl0qM!Jh$U0H3^w1#N^ZV<MFyY
zH%*v+!$}+z6vXR!ehUCFWOLSIuCCiEAOmSw|Ad?alkX<-VY$h`1HQGu2~m)fArn9<
z>>1@x<{F1^>Bl6La6tm|SILiD+7#`%ju3K3zdGcjw{;%-Spvnen`3nW*RbtX#;_S^
zRP<;BDt*qU@e0}PuVpgo>ysc>2p)WWeLdJ>_wL*RT$It!a9y9>Tv~b(WG-Ahywrgn
zPo|Htv8}VM{QQskyL!7ILwzvtCi~MT;=W|TZYclk4OS>_aSsncu<tups?UEm0VXgu
zMkX#iEbIMI94nKl4~g@iJU_vm78em|#@G_ZhXcI+=XL+IU<GH@XO(<FtN_HHX=-+E
z(Fk^QTxGXGTn8*oc@o|0-~#OK+5$Xi-f%Q1AG2Ip*6%}7gRwh1s|V1@$ejWVNJQoi
zt^nG5Bdf8XjV6SiUiT*p5~BaZM;6h(FYgVEjC3M=;W2XvL`+7o5>d8oA!zHc87Sh9
z2uCQ{*>_@iKH|XxTY4m}WQ5E13L7qSbJNJ#nXq`I8=MOgPBSizB+SSzF%c0;M>lLJ
zcL4C9F?2(+r_Y$o#n+|k_gxvIa=`C5z(k;9W~OFlwm@RVhCGAwR{aqu&ISj0FIA`9
z&Fw=#v#rqBc$x4Kt{+79E`?=`yC$TL$kAwweD2jj5U@c-MYa4hKQb~>ZqSJdLKPr6
zN><^N>gwv;&PW+9_6sGYD57PiYp(U4-(~!&z9-hBHh`~4t3YZzlqC<TV4yoMk$bj|
z%I%gFrkxEK4GkAFbPRI`Na%4<ZI*h7z&f>ZA*L_V;w8<>G!)2fz+<$9EF<F0B9qtk
z4`#~N98J5?9(IF`1P3+sssCub8;`tvC?d{_xIi>R&?uHrD>XcNGq;@5;c2)FrJy4w
zfDKL&Js1Gc$O%%PeJm?`1T73sCNHT1Hg=XmU;>X7S>x$&@o1G3tBZ>Zp;dFmWqWY!
zkDot#X7d<gST(tGE4|;{_`Z;w4v?d2#*+lb3&@#DStu{~Yxzwr&5Q;j#8olOfr>-F
z$SNq<|0LQP4^#qF8qjCxKadGZXlqYnpu7VJ40ibXcqLkFcl%tstdz8T`|=(;H#fKG
z1VA?xM{e%?gHI{62zYlqV8&^e7J#TQuoo7*z)bB0D*%CJg9%t^czEuDm7%Xc7lj6f
z__rA51Sls}^BzYIz-zkz-#%q0y9OcRA*>0GeJ_DOL5>WLm_{LD0YM^NJbeJFT`3p$
z1;atxTarLAmP<4cQ0^c-1rr{+8~CO*%x)Z9oSfT*Rvu{i!!1V*M|aEd-93+~&X*Fb
zuu=VvL5>bhqXR9^W_LmQaDA*5HU^}n0*o(C58RJxXCU&1h3PN+%eGuJfP|5cFFrfP
zQ{!|<YmtCQT<N%-)w8)st2V<b4)Ka^!pl=g5@<(u6QW>e4~M?;v97Mpbj!F@EvKO1
z^+=HxU>e<?p11=KMVYcv%PGP2@%n^95RN?cqiYFgpZS7RqL5ub%BQ7gZhs5%^YAH5
z#%F55*5q?LWa@B!h=*1*dB%Vp_yh8U66ruw53<fh293&<N<Y8LdnjpQZjw>*XCCh%
zh=ech=gvWggZg8_N2{#lg0V{$HJV8&ri2TRFkH%)S@Vl-Q9<3(n{R4N5)@csg#&Wx
z0qZJpk=J8p$bavzL@q469)bcID>nvD3f%kh*Imj0NhDld`5kNT=MVGWBc`UJDsw$p
zRkuKO0}&5U=EN6^`N$Oa?YzuDyWR$mGv}4eD+;VTAYXn5)Eblq?`vSC;*b4lVClix
z#qq^u;5J+k$TY;EF%c2i$<A{q0EaAVBQK`w-N3Jd8i7zkvijW;2naYYC*5CICV&|H
zxBTLU@1+KkFnS+waYfUnxJLkfb4vTVHHc!u5@m%-*)g`^*Z?P@FEG<@xvbUyZaFBx
zL<k}}C!y8v+xp@@U(4minY66DA*{(jV2@~K#)K1E)E$2-Em_})2xTTC#?W5*TQ}bZ
zrlbhE#ekZGkP&U<Kf#xOcN;D1B8d?{6gE^8KqJmqEp9M-84c{Uwib@ohbhZO`eLO*
zmFGfMWRT{X`%{{}u(YV{#5J_NeW!M2Mjk>XxBdKZ@$j_|546@@xoj}NP%B;|egu>n
z9cjSir(K%2z5<e3Z|^)K^Lg0SjxY+an4d4~_Wvls*mXG=R(0O2@ye{I@GU4HjbYJh
zd3PI6txyGCt!0(lxFM%i24qdHX=r?Mxi}bspjbRk*Mc0Rldys4!JB}n0;z!WN7xmb
z1&V-qBI`w;qNe{`sbtI?9K#qq#+b;zb~6HRPoXq|d7Jdu2{MEMQ<eyOPbFn$eg_dD
zgjPE?#%DhDH5zms1_u1yMOLrHvbUep=s=i)3=O0zcMF-@NU=gCc23-}b>$IJdkVyR
z;E|#uof{o6fa~z?>5=<lJ{@vEoD~4#2xM>qf`e<<gWkttaIk1oPn*!^(!;ue-04)_
zmoFcvG_e!7Exs=0<;=pkND>YkHcR@_l#ySX02&i2Dw?!hoy&+SroRt(T6uYxcBlI*
zYua)7Sy`WeY=daja(b|4bz0Fdb6ZUVrKAX=cIgEgmG55?B$O3uR(+sNIX7^1{R(A&
zSNL?`F32>{jkYh(YT*q4mQ;TAs?F=7_{u@$sIt7=$AmLTCUa?6RcCQ8<%%>jvr3YB
zFmrn)f;QRt-PJ+k<y)_d5ijxl;a7*h<>llSgfFtM+Nv5ak|r)i&0wUGK6XS8mGJ3b
zZk^xhSrr19`3yp5gPGk3h6XCZZ{V{@hqpW+$_{aBNb`2_RQ=1LAf4F)>9o;CFY*=B
z74zwel}{~<gjRlbm**~Xb8`UV5|o6O$5s!=#>c-mHC+QKhyp)Y>$x{xs02vRspaWD
z2Wulb4vs1?uJX%qykBQWCXDsHfcVSnF6?m)O+eyD?8KqC`1sn-ui=4Dvc66=ZIV$@
zVXxw<46o_IEHG<5g?+cA$i0t<RUP1f119!^{m+#<7(KWx##_R?&T%~V)5&)>ti1ez
z6>>^Ta|35F<`M`JdjOt5;N9qv$8B(yT_=+7gi1WrIUU^6D^)8RP0xS`tR(rxx`o%E
zm5B*LwF(IhO;yQ<ci>z<tzL))%U8>Gzz1ipE{|MJSGH`o>esRa=^^9`Du4o1&|-q-
ze|9MoJ|h1~I_h=JpCeuo#4RpXnk6z)L`R0jF{n(i`|d9N{QezAvB^)!GnN<eaYBqY
zk>7#1c8c91I03*B#4;fJ_cbdwcBI>!+wO{%Tc3@1?282naJ$URnt0&;pb=(y+=ZOy
z^~RL8wl?*(dSXm}6#rH`qpaq{{H9&kdN9&chgPyk3udk35YDIDmX`mpSv!Q+t@Ws(
zCnH1*r>$xA4UB~yXBfos_K#d(!{PY5YJ;n$Laxt_Q5(R*<)*hcBV}$GloGnue|Y!K
z{n4Yrn)-iyImr(%Qk8RIY(Js!?wqH@y3;e>MFB*#*kdD*$YSm249F8{zwi|YBuG6V
zw+UiH`S_OnKM(4T%^N_TrLZ6cqw}P#Bb2u&-D+t2+Z_mKL1q#n5Wo<P0SZ#_@twpp
zHtf8d(*z@X)cngE)Dl>JgA6?6cgi7_Y^r%yeX3P0X!*VSdubeVxgADaco!_4V7VQw
zXF}LKUGjtOqj)K4+4)OI0e$T^t0WFp33hUJ{&k~MidU45^b^PRUx@`x8g<U>8ygmp
zF%yopQ|pr_WMHdhm6h2J9zt*n+6_Ll250`+3OuR%_mh6o=2cb2s=|)TmyH8Ng!J3D
zIOPkZ&jTjXGPaY49@@}05w<Q71o#V8(&<o}e>;0aRWDCoetpFP{*&-A>(wIbXyqFM
z0BNl7m?N#M{#AY+SXfxNZ#N_7yg-FB2~Nt!S29{~MI-^E0-6T;Gi&YK_X@e@6$@-Y
zbanOCEUhVob++a{&jtbbb4JEO-b9B!``52u10>QRe36=CWaat_G8_oiU-OGOB@b&@
z!=<NaGX>Bn)%q0eVuU46iO|Y715;pA4z3-%hT3(9^8;e5ZQ+K1iB~Iog3C9UBD>lm
z6@{er^ae7pOhC^Ap9X@5S)9e25ZZ!A4g9hEYq9p$cE&ZdCz;AShZBvNxw$J}1nBdI
z&A_+KlA{c1*9Yqyu&qV?4Q(pyRQSIlX%*!PjM&L^{kC2GmtJ@~<O&??TrTPyLypK;
zzm(4FG~4U@kfLz!SBE3%gXn)7QXs^H5XoeB1EF8VGfx06cDmNd4q<{+2w9n-_q65U
zxB6N0%1KCcaA`SdqiVw3?~Y|o`63)OvX+^jPm(aqoy)p!KCpHGq~e)UDb`bjYp*`k
zpe?QhSwD(tlZCb}9I$t<!J=%(<E^?@$eCLmADGjTJ|M+RkEZukjfpe)q-+S*MLUk7
zd(wvwA9k+)Gy|-J&{G}O)*z4z>3d@bVL>vU^z?_%G4p_0YkM5Lyj(55T>bKxogwI*
zHaS?Gpf6WXoVLwC)lXP8%0tXSV`EFmcfbU(3YELMrLwxQVF)}E9CXG+LAUpnZ$@@2
zi#<YtN%jQ}lmAt1u*zG(C~=w*f-h3D)o=z3rk*lv;^!oUN;ldn4|;J!JGM_eZn<CX
zTp``K<s$;U1G)e#S*6jzC;b4MyQG+%T=L{EaBsuh2JmI{qGlMF*Ap*c*q<=rBnC7N
z9%iTn$D7{oH?IpbGxwTQjuTZ;$S5ir!%7ySXzTViAHQHJ5Gx+Rfw+=?E6=${-Tt%*
zm)%4DPtULa0_o>vq)w>BQ^>+(q?}xM0*WKy@znoy<YaAqtx=fG(9jolJ&V0AkaW)H
z2w)ejzEI)A{Pmd2GI9-Zr5`30lSLi*#h7+~eo(&1>!SQ9li4y)q;GG(AG5U#f#2g5
z#guvQ2QIJ1-1fHfiq59b%8%mhIs40fKl&m8G^Ajc_O^J-uKSq#!~IT&IhqZN@iXnJ
znj;~_$jK@T*OWsSD`g&@`lB!Mk<(Nbb8VPD*9|1trKMVku|%=4BYt=zagyGbmgL`?
z-XfPAxBhrd{*K;O{~jq>-!HOX>TfpcM^1}oo~Wr2g2AB++k*?sR?laqLB{&`Lx%#^
zbRa^;$-`CZdBG3Mv|Ow=x?x}%E2K!o7{cL(Y68W5Z;xgg5LVAqqaGh4faL6VG1$$?
zTJ(wZdW>52s^;zat@#C8hQ-YF)nnA-D<izUz14dSv*$DC9BXSl&bU45>rKQU+n+3o
zdmHvkUF*r}wZob_k;(rNY$2@E-+$N3>&oRoiunHhOwbd)1xV@Zr)6J|l$Vzem~ci#
z#aQpQNuvuPNxb*Bp`{jNBbqReD>2(^YvpDNI@41^QBy6}UW_;XH9OBy|NWkLI6~<f
z983;al+)1>j-ZnN+I3sxGc-v^A?_M`9d2ZAQsxc>+|9Lnn;i|2fWDMQf~e-y|FQs0
zKYp+w`FJWP09F)KLSqVUZ-IVxyxlsZtfkev*H5qw{_@b!Q2Ksl_L#?WP;UPHFg%rt
zNB}wx8w%2jvJZ`)_SLmNBqJhfwk(f?)jds35`;Lrw4{JK!hr)bv)XPM^RH^c!S=(y
z{ZSB=^zrdA*6*#b;uKcB*;x>n(2)ee{4P}RSI9pIP)mCv_oC;a*0tq#6N?=)Gbx6q
zrr*kXetabA+UVf2v{<m_wU))cL3YphL65?dC)eJ-eS7Qn?R$iTQcwh`l&ORs#$@B-
zF@~pS$ZYXQ>ZOM5Axkcxph1{IUtiw@QW5Zs!0Z7}*>dE(W6qI3>CU^)zB4X^Obpla
z!G(bK2>X0;Ae5t^1d2&WKp)-xF)ya%{y?bKX>-!$b71ns2Esw8yExopmXhf^S@2xT
zK%t4+$&Sv;V_?zx8laLNzgvZY5uXxpI|j1HK*inn2P6@aKB(pmR~L<00DwM!ra&mu
z0K5?cZdk?_+WEg9--jfJdTAhd5fBof`$n70^YU`faWgKCP<ymKAtQs@oAxJIn}0%X
zGvyD<xVhEt!uWW4O;q%0U{DYZBV*$4-&i+%ZlIu}xAl)~1tQ-XS*>_o`0W9#i*5%<
zgEbdB<Ttd~Hfyb0HBmaYq1M>}SJ19s%EtLxWmP(c>ip)7fPerBhDSkv;!NP7-9bhD
zq(YRaKnsq63FiY4^mw)UhyoSgQD6}h%Q!ppawpsmZpTTPSozer?pwxW;KYk{TM4gZ
zvvXc^B_lsS0dj#<(k&w+u^&G^6BQLzD|?NSn@iu;-i|{`T1O{03V@valSK)fiieUy
zOKUn_PD&`kU=RZOK+cl(h#SA4uzB;YbcaNMf<kw|-qsu70q8EUXSfxmpWW$aZ7Wya
z8OqcRyLfwGjYj|>)Z;uY@w{*qefkvAM-Q2pCii}qOn}0<gIxk=zKjWs?{Zzw@&DQc
z{2Z17`4|@MI$wo+YGL7&l9Cb^!Nz7?3a(+Cr>@r4Smfm7Ardny{q((0AzuAsaIj@#
zqAJ`lesoj~qLTqZK@T21c=!9`$4)b8E=@6|wm)y$zU?z-LvSEH9bZC%33h;ce0<`N
zkP#u~!b%LKhgkQ>q|IM31<GQpM^rDQ?&$@$i^BRNtWDt(;H3_m>d9;gpCo&U=;pbk
zJTW#V0P`oIm%G^RrT^3iirpsOF|^*a6?v8XXaI09sB{pP-{vNZhUNsM&NCp?JwCoh
zi0+Dx^e`|G1%L?-35l$`J85{hlA;&|sMKz+KUP4lS@~H>q_3~9xLqqg9^O4%T(Q@$
z7jEin309B@?@w|kghH+{q^(U1ly3a?t!OF;6EXw^2a_|nYZP)b#rRb<?H(N&oE=(Z
zWKbDA!hqPu$^MXHwD)_68lYo-0z~#2mhOE0{IH3MC5(+}!5ahbjpaLm&wu3}6!g7^
zh=^=<cd<LJ#|Ia>aR%>?J!wfvWLOJ@W+8`K{s<3}Ul0Vcxj1zIGXez_RUd+9h>eud
z(b4|GPC#CiY-|BdO%y<mqFk0fcZunE2|I1>|Cs(I0!sy|#VX++KLW%@g&K$~7^h)s
zNJ~q*gNfO(Vyjm*My?yRZE{@}>y~?nP>4nc0-NKeyRkqF;g(XOV_bG;Z(95TF**`t
zCX4Nl@fcOE>6V7T--8nqvcQWnGpS*CHa^K8#v&sVfvrVw#wuwaU}0cly55#fV2E=w
zFt|zXc}xi_ymJA5AX&9Zl6$miLbg(^k2UB4tQJA(M#jW&TM3QZfyxjJiwm$B$3^cm
zBbI>j&YMaa_%>xX_PY|ccR94061b4Q=~%R7fApva=>E-{H2=hH;!`O(;MAxwrNj5*
z(64kYbDaF@5X>{}>;oHcFPk9A@bAY7DV%)zV!CCxp90OJ9s@2^xaHK65{P%agXZQc
zkFmyM9%ri~9b;;4^0}-m;_wSMw3o34i^tNkkpUk*T#9zA;)#csY+JXqv;r7!GUpze
znvyqOwg4#?u_O$V0(?;jSf$=khdx0{+5^(ylDNsBStZ2LZ4Z0Yk>qJ6EsyGY?i#8e
z-cfs(gJjc44w?H*)&eHC9@^mGV7zG6>><PwB;>AP<My~eopoEz^3d(3r*O_Z2tXf5
zPvXE!^H5ULO9BaJZR8Evn!Gv4aXC4<p!$FuvKx4*T{R)^;&OYrKLweFaRBG8hI&jm
z6-v1aR{rq&;^M8AnGlVQ>TYv7*%&I9i{2G`{YUYHU@X0U#LjC16><aC^4s;*Iq!nL
z|KCUS7#b491|h7BpwhN>w&ZJ~4yvjIAe<tU5>PRrn{;W^*ecUB!v3F(jBIXR!7U1>
zd5IX-eyT^0WGpJ&pz>&VdE<am!A_V6XiOu;I?FvP9Fuh}kpMQ!+}mx%M8vX(1Rwo6
zl9Q5FNM-sQ0vr@f=5rJ^$H)AIjst{pOd92(&<WufRfCcZd>0lu0^y`0(wUHMmxs42
zX>PubQ@)Mx$`)<uF-Y#+Q+V+LXf#)(SOm1Y59C+A!so45+ihzHKT>^0Hx4^pF#s(>
zP<VKFB)R&ZO<X@T*q%kI6C;z3=<u2LgC!XMF>}C9fRcbADvC9^62jtQP7YmhwV`I!
z5}2r=t+UT&5KrOX4@ExM(1o6c^P9X0E{gmtk&1yK$m8s=6jpH{82Iw#OUNO+qNo8-
z0Aw7aS*0PRpa5(A{y?;)lbIQu0PjGb)i*EzXD>CkCNS?i1JsiQJUEE3CLYBZi8PCd
zNlL;dcGr&|KkBGzBB@Xkdfzwy-fO$LrC5@4s{lwoe6JQ>uge>il_NC{YxtL!m(YiH
zL#^C>K%j!0XJBBUIvYVTGU5ii7e!+XU#ru=WzJk(SZxz{?OZ()5LmA)?lqbRf?jvB
zq`DLpV_yU(&H7;VC2Lh@=Ua(~k2qgv<@~{(sIq$Q?wfK8&UGTMjedkalQcFU`iyMj
z?TML7q{uE#%eP#d#c#LBXq%gx3qTi8sy5Aoy8ZW~CmFOZ=&_|C>dty2Xz&o2zPUf(
z<t4o4KNOfV;9PUVvBJrE_op-n`lEHG9BaP>pEltn_6voIiHqMOBI=50qkf^Hq6}ga
zzzeXs7{L<Sb1j5r5t}`uhp>hXxTbH+>^rzbm1As#`zDpaPfK2L-ay1#uFkx!tXOnd
zI?O5~K7MRMq}5dlzr$voi_&co#JW)X>~=;@lwrKL_Y;tT#9$vDc1-rhRbvdnxdDR?
z2fm3;FRik+dd$goYZzoev_CevrqO}rCn_#}f`d~6%_>_lB`Z6-Q}!)zf4G+2_459E
z*w`&^Z#F)^T(Y4gbu%LifEN^|_3_1jO-o71#T^Kq<z%%8tOFx5Hb8;UTrZ{@gkX~@
zFgTQ&j_!^(V%0jCH_*^L3KBW6?|{1Rnl!Ff{sik5eB%%l3=~ROq6m&<v4PY#P-)nh
zo2t{IrKa|S@-;InDrIsYx)*-Wjz5KVZ;lS2B|Mn)4FdU$%8wthpqVv8w}piB`LNCf
zp4TqzRrvY&xj3!%b(nM3HHqJygoTA~+`Ku~=yer!|JRDWEwHBbL)!UGhxrZ+c%Rm_
zQ$fR{^P(|=mcEejcvWrvc6xf6b~iy&Q<I)a;4!^W)Wil2EMUTSzd66S0H!R%b_70A
zM@NBTZ|nBq2Il5u&E7Zjxc?3o{2Yt}yJ{fy1qKA@!nfGjixIKp=S>12Q|;YrW(X;8
z;E}Fhx2)&dP?*vm)*ZK?f!LovH!SdbZMhTnm=UHD6d|Z#K`=U99ybn{yf*yBfF$Sh
zs#ZJQSKt#fXsxBCr7kv-iLCgcw^gL1?pQUP(ENWieFr#~ZTo+uQj|?7WMq#JGBb;e
zRAw?lA(4^2(vX#mlwC4<P-c-$Ms|`UD@l@!$X@@?{r-;sa~$t+yuI&Jecj*tI<NEm
ztW(f1kBGG*+`#p$&VC8k2i}Obc7lHk8JONJ|5*q67O1bKoCeiUzHz|Ke?E@kCa!PA
zm2Z81)7LC6&dqQA;8oJU!>vrsb}^n@pPq({#AL|~sTdLvcl?F5PITqWx3MUyp08hj
zkEaQ1(pFbj6Fvkq8USMcE-s!Y%fmr|rjDu7A5zjX%wezr#BTlcs37!rhYtMlU`1~V
zGGivoq#;WuiK(H%SY6$AE8+djcX!!GbvYd5zZTcLsp3z7;o^6D)otfzNl9T652i|M
zP)(V->Aa(8Oa_fEQ_;YQP37vks{yYBHQyiD6U?!*-i(FaBv<`OXINI-E~kl|uXu02
zAV~@6^mlP|+!A`Iu8)yiNn8VtS|i53E8!gE2P2!j6XtMljR(ZpGPAR%ofru(5|WUW
z{cWJAPtUtcGgT0+dBxC-!DB=$-L5Nm8f(SC@Q8W1B1WxwdH@2>*t8m*!XQ&V#q~e?
zP+Ew~1}qM|rD>@$4;n8smsXm-QdIp(qfCwJ*<>dPSC4n6g-ldLdwj>mpW_=Qg2i@k
zdsS@JchS&{=9#{A_&h6;+Z)z4X*Q?0D#mC)DFls)uI>TARDMhSS2vbQYsNojazY7`
z%grl#>B8asnwpWe-=KvqFh}^UwmILx>!n%u@sx&!%#ClJgbWUV*e*Xle+_;8Xxidy
zgfG;_Mm~N|H18FY-S|t1u#6tJ5VwiWVk_0o;7}pF64R!~f&3Y=(JpTZ1I(<f^vmp8
zU0dU96Hk9(WoP%Ku&|r=!*iQgW^e1Ej-^8DS3l{FaF84SGhkyX(aqi&@-DPzvSGBq
zy3iOLM@E*&`4KV!0X=@`_z3kF*Qp15^$$77iBC8iuw1cl^zslcT0+2szG~?34!hTt
zaP|>Xt9JwnJT%AbtB~V9091$5ShHSH!zC-b)?$?W<I1mE=U>}(>-T$%Y2t0E{V9Jv
z&ZM2Ir&yUCzWq%??(az!w)ADWzz2&C$4`H3KQ6><L-McfT<B^YRkxbL#t@&}AIhz%
zu7ex>UDU_Fw)Dl4DDTi$s}W+T*^nn)uD)|nQKIw~rgb<{@`XA(2wVt;q}8tNRV7$k
z-tFmB+MLuwS*3y$HEFS|0rT<pn#HzYg~xQ_0vNa1z_jA43ZDmXW<7_!Lx4hie!lH#
z5>GIU+v7DGOguYYagY;+M)vKM<~#T9b+!qkizKL11rdwtyid=2CVxahX6v;+T{>||
z=)Asuf)WNMB#y1D?<sKXuaBIX_K@o}%7vfj=#eA0_OnU_3ArJW#>vyYz^zZjpwb7!
zLc<))pe{fE*^68%yVKqLUZP80Vq)R<q7UG?$yxA^|Ml1%|7!9!LszI;#;nu3BYx#A
zZNX9SLU_)JINwA7gT+Q;;8=(Bh|Tg06O#_}p55}wHM~y0d|4OYPNa?*)vS5GvZzti
zB)p`f|0d5@E`By*%F8<y#a3q{exxl`+;8%Xg-;$GhWSH9#YTdk^?+ayfF8wB%l%YT
zQO(umW46D2@)r#Z<9X%M)6zt%tD9U%&RBYR{R%Ckc6codLJ7*Rhke|J*$c*2D{Cyo
zL~SYdZEWmzP2diwtYYn{(H{;>8%o@I@k0JCRDe#yl`ZJsD&Ih^g6~kt`IwmE#+)jY
zCrT}GgN{;4#xJ6|orcQQ(P0H~=7{dLVK%#z5@BfA!N(YEe6Id|q28RyvRnN&<ccE;
zyNc~49bfk`3-hP+I6rI>dVgR-(j)Fh$?IdL=Bw{VWIe|B#gC<lujFq(e%iN8H$Njy
zlV!i;jQASZ0ZOG@D~2Fjl6S3n^F|82&qSPlpPl5#zCP&*@zRnKzW#h(8IL~)VIxk8
z4I*$ygaB2&k(QQ*ovIxK?L=m^pnw1_bQv8T9rFSFGoc=b>)6{53T3B}>W-KUo`s_h
zN{*?Y5NU;+98aGaF-iG5=J{6vovy5|q7%gL+@D)iY*OLzQovl{3MUum7wD1G0@nDM
zM9sOy#BP|Ywco!_zVydC8Y7SEpIE(LW&TORNzz^K8P3MrqP7vkL!POn0&S$mJNPz`
zh>{K!FZ7H1vvpIx4|xV|N9TG|C|g+M_XTlCN@~BY!~DEjHL|Wyy*cgd`!ZFp!C2O7
zjw8S7O`!6i7}WA70=MK3gd8^Uh>ebJS?E$yntjE;Ov}o83Z0CtZ6@?RF6$n9)YTsw
z7hmZJwy|=&etjGQL_5jv8FlV3P3_txhr1_F4V;npTj~c97zB+|Pp4^VnYQ8>jTl#W
z2;-u{&kuE~ij~z7E345;REX0tthnOTqm04BWaiQ4=H{Y8u?KT#ugbqd5+MNh6eg0k
zE5(p1v}hL2yUPx74YkY)C+TKiG}~#SuU2dAcZZU8tm!mWPw*j;j8t<oB-=SSbNzPS
z+P%%R+)>bd^1~$can2&{Ta?Aod$@Butm}?PQtV@ioG8_$6lhk8z46v%X>siLRI{o7
z{%Zz2-P<E5O7dTS`&L_9f1O^kZ@l`+;q@=~@)LWFks)H<&<mIktVZP<%nO9~Lrbgm
zQ2Fm4>yFUesfHZrND8`%7(X~AvJdIy(w${^)EEwv%+nh7+?#(2d8%0s9U?#obYz6?
zgpExjzYos-J83oSoBo?CD+dwS1@qtiZx5+a71*cz^v6Yj4LA<GiNvemGWn6G{@uHY
z#_#;-*dU{QgiAx;KwlGf>wKJSwL2kt&dHpbcoiddj|FqzN)icAH~Q)*g^vRb0(tQ6
zNR6}RZ_-~J;o*7YIlQXSnJWClcVt~ew6Y)xWf5s4qYt@3Y6OT<<P{W%<?n4Fv@6wq
zCjf}X<`obJ895Bz<ByEkd`+m><l&^GDiY$-6i3jZfWCgz$fh+^V?^5oz>p)16X|&J
z+kUs2B^Oao#p|=v*s`;;_f(nB9v72liFEw(aO}}!9YA0uB@5e!)2?sDk~%1yy1BzR
z(N99X<;^1uP8a!gB!SyO{ISdpO-<-fdYK}^S|=!>qk*L0K*+l(6DigMhh?E6ADWQu
zzkQVN;heMI;jtV?n?9QpnyxYh1{Ss}b5@5G{K<nBO@((VZ5=Cb;5le`rE8{M+|IgQ
zbEP|~8%C2~XOD=HU)TQAKEkwBO*S(S-u3(V`x%#R4f-2NyO~b}9J;8Ja_SpPq2<<i
z4cD&{we8Jqm4A!=zr2CMAiZQRZc~k39Zwq9TEI!n3_DPRD?zM5Bc?|i#jVLHDe2xs
zn#UKmv6}WjQ8C28+7~bMOf5oxaHVte7a&oPw$0ZFzaQwgJb3=hn|I;XN>As^$;tU?
zwnWTOhK5UOm%T6;weLrO6utmtZZ>#pn|<!-c^AnlZCBN+Wf7Rh`A<~Gv=!eTFt~bL
zh$!2{^-&uXin^zrMPlx=Y8!m}c5jmj2bnUV-GO1xW$lwaq$ofC&?+h0*l<64_6#zB
zLfK={I|8?v7dtRhkdp44TQ=I2r+ND}S_V0o1O;=u(`AmVk+<f|0>;G<^{KDVZA=4Q
z(bjHAw9Mz`K}cD>e!y=i_s!)<rO}J9`}cp`Wip5BVE9_Cgw?p)#)`|*w6A^4qYr*W
zgss)6$^WoOI!r#DDRPevQbxt%?)&`e(is;Q8X#|gsqrviM-IrVw+6z1ikDpSb<(3k
zQ(M~xsx|XiJlFf(zj8}D)HpnoI^I0(XM~Ix(zlcgPI_sy2wSD{`_^&OH8(xINCr7a
z;8*8w9m~lAXw^?wEOd$5hWekZpY-*k-_EQ@XBYj5=CP?rd8o8PMY*xLPPsLZFF(7`
z_P8)vLE;+8ncZp@TwA;3Xl|=`1zq~|Y0tj;vehr`ud?^f@mkI*k}lUgaCgqrDOLMb
zy2bUbX7jke(#DZj!z4DP?H`;K8<a%Tg8FKL=N>NC)YK?}R!`w((Wzuj0&n{`aFxl{
z7jp$T9Ap~$YcahJ9lIiX-vW6i0J>mM)V)}n81+LEZwkI90c*ME-7mO}%X}6E;H6kl
zd^R;XIr^g|){U_Mh`Nryf7RND3;N#skai0BKP@ApEN?EWW(?$sS%rnss7II{Tx9-`
z?cP=B=Y90A%id%;qtl=gn3<Vd{lvt@X_0sNe+i8c5X3~MK9g!!2^TjN26PB^63WXn
zsDzbev_AD0E)+YO>_g}bKuXbPK%k$5wB!DWNsX_55y8{Yc=mp0=fJ;zgGc!I?t|_F
zei8cM!4X;6dq8Q3UxbEQ!LTon$ecf43U@|ov>~MtfT`Ks{L<2r*Gs9^BKZ~zljq%Y
zO{~te2M@~F@A%0~X@G>^$!60|pxix!gAv@CWka;<OI2!SW*fo1=W&T8B(wqJf`E@V
z?a&wuXHnBWc$ts@3zb!sQ&DK$uhvJoU#f@Xax?kDa8JQY9Or>*-EgH?*e<PHaL{KY
z25ArA9r=$00i2znOVQ1yB$6IvK!79@LtZl+-Emh>-?%Y1@j2!62Oq|8vZ{dj=5MzT
z9=y|)_?oO{(||Ie?IB6a<wnCT=2x!<+z#o4js?fbRCP6M6`$NqcAG@fRL!EMA}wf#
zduZsLs^#mKN49<4_qvb=`Fh1|ZLcn$-t_NTVdi9sY?_jYzT_ov%kJ{PX7ix$V0-Sl
z3lGwMS5lRKHnVqjM#>&t5X4ikNu-k-g3Viy9Zr4$X>>gV4L`ZX<Wh9Oq{M0WC@Lya
z6rKGqJzZQ0jun6qG&Ir0=5g`dq40(+!0gNVScGBJUDl*ZHVL3<Y<}SM6u}JIylm&>
zmW)$g;Lx2IS_!{(OPT0xE?x|8yle#_;qUJ&betR<L>3t0sQ=5$OI+B<FD;Ei%u0G6
zE83mr2u3R#lb=h|@Q5@goBd+n6Rx7g3Oi)Y_VP9oF6J{H&)nXloqC!{ub#aamqzKp
zOUjct$I8mevAZ3P9XnRJ^z|A6Fh6mBfGjJdf{Z|ym)IN=UGU%JQT9N;gMe*?+U|se
z{Kb2seeq@r6#)UhxxOkXIcVwV7?GHG4O-}Afc)Nijr^LlVCZ1ZX4n2L5yMTx#MJE8
zr4TYpFG5d3Lb9^9Ch%8!_Z}VNm!BJ5ZaciTd(EPu@yDDo5LiQ+lA68^qE-Rs^Us`Q
z5;6ORrzQ>|X0hx9j0j*__S9O66+ohch>~B&y*3BMbjEnc%J-917O?tmZj;rz<T70K
zWcvDLr8!{Qdy@AS$$cs}k`I&Q8?6^3;^W0viUW7$?!Ts^@L4c3)yqQfR@<sy^H|~m
zP@izPGt8Bp`Z+#MsPX~>mC#1vbQif#4oX~x2;tn>v$dn|LO{r%YVM<;=r461$?NpQ
zWJ1g}5oZ<_7XHi6H=6OqOx+j`K>fgh%cTPo^xLgA;x4(bAFZs&5Jnpt8#39HfOCHc
zK_iX}N`n4RHAThUpsj!la~wI6`pNb`hj;SI;Khuq<zt~jL)bTw%9Ph>4NzU)_nvZW
z=woxA&?irxFn4e@=UHE{c*)5r@AujIL$lNY(?k0ibdY+CP6toX+aG`KQUBoJU~p_l
z`S|S24Tr91p#!)qBHAX7uj71TZ)p1aSC>So85w^bKS)OgD;FH!`ymdE%8WpdiqMdH
z1ciB25I#S&5)=YK_tW1aKf(mjC%_YOdz)Z;iZiHR#p!mLF*^dpYu>dV2t=gD8B<f)
zYkjZ$zgd2?6yqa}?w2q5#%N5XO0AZxJ3+_%A@{`#p+6_3Fy#$ZcnS{Saqx+XroAS=
zffj>@&g{YkdR*I^_08>csw&;t860-<v#Lm2h-vxvDbC`9gYx=)W9{J6b6ZB;O={k2
z!5ww)6Jk#v=uPw#0Qri^K+CWbnzxvwq};LvLK}DfyoBiI54huTK@0XjU0+{6;qSA=
zxC*QT|Mw1Rdin<ZU`z((Zuk!OAiES`q<VG_4Gqn)sRj4p85>9UI6%1&0S2d~DM9G{
zu~Jl?leFSp=Z6mu#q&QnyiUE=_VBIWiQQGo*PrkrLD>@37X-AE)Nu06;1~)fkq@5+
z+ksHy;q^ycmT8S?ziOm7)|q1dIp3o67wui)eI$ZNR0ENpon^iF&)s*g@v~=E=RNi*
zQxhR6Ul2f@(d(p4UFMb@f(}_RK4i4eIt?j7_l4Tj)YJ&D*-hD6R;R#YS^|x3N^R{)
zHT?|mEA#VRKNT2!@mjO{&E0xzJJ@hXjFU4N$gtvv4fNg?h1v@Ze5|R4zi02(_mtk@
z-vy7H#`Wo&wjW-mr_Z~OO-;2y_}J3gO8WHqb97PZCMMeW-I36)3e4sL3pkbXMc_HR
zzUCya+@JS)Nh<s=X#T$zAZbbRbGl!V;j_pG`Wik<UrlYoS$_BmAPlg~W$fO(f|OLe
zL9QVUdtjeHesy&!0S5l^;e_n8a$~+N=qkx8ME+n-U{1ScXLldDwuo5XaS~Z-XbLG3
z?`QyUThiv<O#uW;i(KmbSGljq698DR^r?(XQG`vO8Y)ur1)kD@$f)iRTmH6B8YX0b
z5FW;-YMh#w)=Y58Rt19x_9f4xD)Hvj=Z-Ht|Fl*aL>|W>A&7`#WV*B|6lVJco*1O0
zoQ6I0{3vJd$jB^@<=v1Fg&PGJ*xwmd&oi<IZXAQfDwd5AZ`2{Bt-TeVH8RfQa$a-P
zIy!WP`GbTV4k02+(dFQ_f`duH2A@<{-&G{E0Oxsez~=6A=O>!gdL3T7&V3^R4OwM&
zR^hV|3xdg7QWbpJy!_BaXgt?1tYu+lwz8~^wRJrrya9(flBuo7nskDARin8zA$2?^
zY4a0r+5L3H#JxkDA@%}*_r;{k#&!$$ZBLX?8|YdXziR>PD(nmuG=ifs(S@-Oi6N&G
zzU&LG4HvCGed<(*X-ll&(86l+kvsRs*4rKa80B+D_7|Avs<SX_EF7$7l%Iw&%4Pix
z?*UWL^eARzd-fnmDij?U{1|xO3F-Vbb2ZDrz(67a0IpUgk{eT>h2aR09m|V^qzg%X
zf+MJ<;Eg^_$d0|h$0KT^u&}VFYB0H&1G9lRwJu|!!l-N9@+bNQ8D(l9aKQPv$Vt&6
zQ&K*8Xghh!?Wi&}QlbEfa)ogsxfgX~bZkuZ;wjckLs7KaGyISHwkOt~G;hrZ+HC5I
zyre7=>-bKgEnQtZlt>d-`1IXOt*n#=nAR)B{ZnrPO0~tsN$b55$Vh<t^5wrcNv{qY
z83wl~r=3=Q2f5n_-m|>i=bz7Egr=M2R^Pi55fmZhE3|a^vNbss#r9<i;{zcnc$yEZ
ze3zq>lQ+}`NwdR;oSGfk&dEj#wT0d6jyrsgkfB}$#Cb@;=TqBFRcbQGSwJ9dRPHr~
zTLB&iF-gfE_LeW9(hTOk?0>utP#w6?r{(1=nlUg1q-SQ{MXH52O)v2q{n!3wxV02%
zpFK-G)#uE?&He9IGi&6q$Q=;)qoXw7LW7qYx3)--(Hc045a81>+4*nHbm1jO90}MQ
zjDv;!f8g-Z(6|S^$HdFh>iTfe8N*Y%aWU`w-^>*UH16BPMDwp-FTQKY(2-n2kiMSM
zW?Z*^RAUF;(D$va>AAUQZk`JGM!mQY@@vZ1^X!brkpV}l!!H8$?HwG>8NP-6?0#hA
zE)EDFKYTa~cL5$NhM%r8Vi!|%n-T1WafC4If|d{v6ujz}in$!Q%A_al?D&W{85tR4
zhJp(h7zbE2;_)%Zf!||jF5(tkyC=W3b;4`zTXHAWjo+XC^ZgO(yDQQ6HA^Ulr;ndL
z{kT~v)OJY0)%`0=kG3tJZQ!b6@0FwZcB9>yb6>@iuIa3oYG10M%FmtbOOziL>2rAH
zQkC_yn~R}t>g0}2i?X@<xq}S-aoa7aZ1%+$Pq_VT&)|ZR#ml4lfI;m(O3Lr=sFZ#X
zv2`GQ6joQg!7+5n+VzjwplH=8)5UNfA!jRn1=&YP;x$}73mwh@5MQn~RJ29p{(LNI
zK@fXTbEdehg2~|%-KMM<y2q{g=j)=piHTt93qC{$BA}5JQ>C4g(-ph-?YDNa6FEyf
zyBi!~<{^xcU}`n@?f|XG!((u)C-BDl3S=sLj2Gz`jdio#lqFWdZvoTPMJzA09_;K~
zXjuU}LZ)r}QYu1&os^XH!_*=O9Y9a`FKJyt?E(T8ACzAi=Gub$L2eo`ViHV<*I!Pg
z_!Nm32GAq@52N;})oNt(`T6@B{B_6~O^TvA=&!Wxe<tQ(UJk|RCeo~JpOIUln?=+)
z335F(Z`=@r+6R3yd`SdLmQ44ZJou!$yCmlFY`oy510A@f(QqUzeJRCz4v~nq=1F1!
z1F{R*&m`;PieB}q@vRSb94?BE`w{2z%tRx#AoZ=@%jj_zO)I}T<-)RPg%L94{)L&5
z=B{z!CtgCnGo@A03cnMsDuypFj{kMrTR$Wm+`Rj0nDxxetatSg!+2c%AXkCc#REgR
zBrRe0Ikoxcz4Fp#jt3=D$NX@iOXm-4+c{<A&7o8haMw4m%5Qe5zi(D;UAH1&d|hyk
zty^>H(__ED&DJs7^87S0+s)ygi974po8@g|Y!e_-0>?1lpzxFNZV*oxCxNv8{d?@3
znI9I2v>j6PL!;1E+@*L13<<iXOY|;9fG3Z}YMKnul8r`_#`A@N>?nf!3`6KNPZ%DT
zrzaCukXUvkzIH2$gYIwKvqtVD#HPU<#FezQnNN4{;J=2c>nON+=2eM+E1(v_5OV(9
zxq2MeE5fuogQu0YSG%&?<quL**Fqv%dD&R|Zhk(n>HsZYd_AZvjAr!v$jON`Dk5R{
zvh^6XItg{$g&)fSN8gydB!WTb{+gK85vtoOU20;>xI$g<n$*$-W)~K^n35v+dJJ(!
z37nQ-cILpu=9p;DHjW!M$SY^7CSPlv%jOZQ_M4saaY^sCW0dTM6p0rTMC&rhC!Ss}
zBJd$J)U3i|e-OYMsBL?_TYfo8vh-JYiXzEiJEGSi6Jxi<_FCV@G~%F29*7DEocr#v
zHy9#fHUx-ho%AGn^1D|lMQIsXnqZ{s-No}LJ7>0>M|$b6nwZ|#i%$9xWkA)?!QthH
z<=AVnSF{^J)M*<n>;zhk#{Yh3=xG1dG;2W5K%u&14)*)OY*4vzRzc)WROdbXzW%Cp
zb-*lveai`Z1%;5^ZU6~KEIP%hA_EALAzQ$9jTml%=0#J%sNTe{q!jqZX-G>`b1&H1
zC)zvd8B;-_3m12mkGj6aEPP1XSxVPr+r`Djain@PdLp5~y?lU8s{QA)T}%aloRHiv
zVPeu#Dld7Ea%Fv8)S>hue$s>^nzV>G(@B!TQUM}^3f+iV_5T=Dh4~dsFOX(EvSB9j
zilQ9`pI5LvXbjHHW*ubXKSjq9Cev1?LHg}arD|K#{-+5|6^8LNks9^eb5wgbzRu=G
zy08<$_$#-SYP-6eO@nhao<D=ukbU#lsp9G@XfmXohskjF+`+(1xm%C%6*^GZfWi1-
zQ-eN2y89hd7d9o2xu8AgfjbMu{g|w*maQ!h8mFj;Vq2*sRMv>^OHY!La1b8!y~U$r
zWTFf2YW*Hs{vu<Z{s;%h8S&sChg;(`)P6H`avO*|=i#^KC#)QBnwh;lr=h9IPiV{$
z9Ave=beM4I2W+nG5}ulyqrOsXLi0>bpPP^GSM+}5j`kN?%b~ntr%oz8MegfGMo@*R
zu|YR6fdI*;w;IDO2nFIw`s9k7hx{3!?L<yEwy(7Crp-OduRMN@<Qi0Q7F+X7WoE;L
zIW{Iu&CMabxqaHXfjH0k|DFbOz^#Cmg~C(`gG6x-#dJ^6d_F(|E|@S)fIy*(A<D9O
zl@<hP$XujSm;vui;DNKjh@YAMul3)0S)rkUnBSi#mqRixmeXIl(C&E3{k_MYrK!c9
z3nC}~EjJ$*&y#A{`9!mc=l^Jo3HK^k@bM#%7?->A!vOzxZ9R-Z5K~}1B8<Ab>rh-F
zbxC-v9f)5+X*z)9!UJ>M$7PX&4f_43YE*^yLKbH^34-eHBDL)G0*19KB$k8WDv|ox
z^OB%Xi?}00x-aa?3!&6jgOLlJ5Iy}LZV9&&5MTV<-bmYiw?4c5FfZ@X#-hdc@&5i?
zrabMcDu1m?lDQX?b#*;%-4s+&->R0k;s`r0FmZ6L(9!9ib2VvEk){J4fUdND^afHM
z`1$xOfrjg~pSv!=$;tVp)R`UQIeJeVK_Ws5$Qi*k{xZYMsi&l%Yi&IiN2=>Oe=cBq
zc_iiYThuLvzK&22LZ{^LTEuP>us)u6SXvr8p17LY9ng}{fvjFev`qt$ULvIqe-y48
z*R2ij;ehq&J#1q)WXfE}_aV@@^^QDHMDwEq;eK{ErGGZv7CK8ta&<s>_x5bf<=pBA
z<m9^#-)kSY<9z;8RwChNKoFnM6{>3WV=YM{bd39p8=XqsAZN6#6OfQ_i%FWf6LfQ;
z^DstMbi-&GXsD^hCPs4lriegMd^$jaV%RC8uive@OGq<j<wvX@o(m}lq(jg|be!=}
z!<lamd9lwpDg(O!Q1C)bW11Qo_$anwxPfJ0IQ<Ye8;pa9!8P5udOsy444|vE%lCxN
zU}mIcU|8IBP`cxgz5L#Y-zqEQ_#i1MDTid{<On7J_&EU(Ee{8BDh0Bl5L1|@!l}_H
z{EjX{4+NXt*Q}Xb!<Uyg*Y=xLo5}MteuirGg#Q{n^ho6O?Xpb($pDO=bNYa3vE#(a
zwJ{FN_?c(6Q{XU!S%Ap*fd36QgN?@R4-j<#*^#mR3EcU#Zl{J}-uuMQZ)a{ENwy1z
zh-6o3Eo9;2cxC9?;xh>aM<EPD{Kbb`it^+K5z1&lA;j>g>lj2DxGDjphPT+HqnF%V
zUse*^-39OytE-x)J$BCZ<_A0ifYRm3vzOz<p@>WT3pQQMfu}|m&YpVHUocV`I{t~d
zNvkbxM`lxR|H}0gRZ<4#I~;+e7Cy{YvJ@Ic5td?4^)6+3%>0*lj+&qxQQ>-6=m3on
zRP@vo7%l$4FLw~h5GE=JUZG}SP)BwARv1%c`f`EO@)qyU&x8v#D&QX%3<Trj;{f|H
zIjz>Z*uguGToXn@mI$QXKrSk*mGjgmJ8J41ku_=A0qTCoE(PsUh9c7K-xp++mOr|F
zzE{s(TXJ?|SdNU0?88uRBe9N7qq&KCcEBQr8#EW;_(bTyUL<~D*obRtYHIoWxOr-5
z(iNw+ju`f!$z>{6$Wn1Vb3b^lVIno$^{1dFEp*JTZ&ydD0#BksW915_#=i&^`a3RK
z^S8k;5jgjwP?GoL!S|j%t+fvq#k)rk;KYkyPb~#@QeXcVI+>(s0~3%qk3}wBwzDG-
z!q@OWb;gwsNg{xtfK1=UXpIbhLOfqECxpY?<Kcn~D`?-&ViSSp;m5WK_wS$fm<@3G
zYAb$_(p=<DI-f{ckZtLJz5QFnRqR80Tj1&kA+;hQWN@`5NM!Hbzklou41pyQvB)x!
zzeHRDu<-J{cSi}wsE>dj@77JF&41_7>h9=jwN_Rpsjg1jru6+-aCf+7weH;>8@jZ=
z5%OypWc6DsQOABdCNAEzJ;~0d^+#z-;ycsRRQ7c;J^IyRrF98xZD`!})*PxaOszq@
z*6!!d#n5g$1cP`62v4I<gJ~@%Cp|ykdH8$luHCz19zSj-CShzY0=)JeDNj`N1?Uaa
zjV<7UMskU|t839$Z(MxG=2tca9MleZG{>!jYX8vPJ!<>xc(2Olg#r15F|yg~%ZIlM
zt}n9%=G#dOjZKv6P?O<pA|-_%A)=k~bDsIwxF$C_DO&WB;c|6w4tU<siIb6>i;2Y&
z5u;A$;(p{HJjV(nG>16(ctOzgA)G{qxO^8+ojUavEC|%7*yM&uadLKcxvZl@byN52
z(ZRGmbJZ6jmmTAGd52x2SY{4ri@mh|<L}@-TwAHuvmIM}ZDalWW_|88HaDXYmH992
z#3fxaeXs8<@6u@sj<R@R-5|0EpqGil@EUjd^*<zlM?qF6?-Ki3&eMmUA}_Sg@W$M~
zmmvVFF&FS6)f(-K=&tk6y9s^#t(|+I2TGQ_??YX)J~V6RQ2w9%j$ywA!SIi34oFV*
ziw1Ry9beeeH4aVXpD%xf&{(G!Dl9E7{*6uBXp8ed{lBbQE$9qoo{M8soZ#gzcXlCm
z_e*eZ3w8&B>LKpNDD7v$VVtlH*ISCsl?JX$<FX;F$|lEe8tAVcLT)Bl$NeDr2;l<s
zgE~4LaioUe6Cpydad)Q=LZL=4_OFpfi7ms_+n<TLgoTp7Ma^$Qj?8~CWi=sD=5UaJ
zWe1iHu|95F{9@$AqnVzSbq{{L+Ts;h*k5g2zbp?kA_na54WV=p6`?!TL2X8*TM$3m
zYbQyBKmbH@8E<6(^MI;44%bvFcMMF0xJd1;Ei}kQhKE~2(%8c$3dU3Z*>cZHF<91Q
z-R3yAC!Mc!2_4Cy10l@{KHp7l=97zy_??c4Xnv@2cKBuy<!SL?nMc3IH&v3kTgf@j
zd@K_aHwmIq*I%uX7kMz~d%fu{n=<vRxyDkQ+vNb+BOm3B?<|mWnZAr~x~D<=eGD7u
z7WZYlKCfoW?dXKxK0O)R*RaS%`^Jp|+V!_!SYli5^g4_~ns++pN2vFIeb7#f@~jO^
zdjKOljUHAuCBAc@6wc#Ck?b8scRW*+(gSws&wb3S{B1-{ZNsdYdWil~yu3=RjQL~n
zqEqi<W?!Ds8`AwV(f@;{FmAI+RETSY;bW7;Acgqp`Jv^(J3%~}wD^2|%d=P24L@g$
zOuD|MrvC6BqcIz#CpsWQiTRA5|7}FiB)6RJaVYMkAU>vJWR%MM_WFZz1D-NAbh`BR
z*hxwpc)-fSQjfrb6_m<fH{EazAh~&bJWc<+Tjb3tw!=Y{SR9A#g-~)lv`qjxhZpXL
zt$xWDKp|Yx(ji~JvfK+%$lX8D$$<ED>^vZWq>AYOoj&QDxIsxlVGZk<gvj~-;eVo1
zNes%=(1RKaaZwoZV!!2==>wl%rBn;5A>kdXL4qc3rSCHaYNLk$NNv~!5QB^rSFVse
zAU{fu-);fM$44zPwTLi)+->dhg%}Qw6&4!z@`~JMC+YbwFT{iBckcX<Ez0$+|LR`v
zW!7KkdOme7?rOrFgU%)2<s_aDAd982o4+Ae>Y2Pic0A;rGN4SHkwodKCUB8NUL<N7
z=}GLSk#-t<!flC;%>_#XGW{^i3!!H?$9i6UoF6L~ILL$h7RF};+qrJ6?pcX_BNUnU
z$OHwaNyz%b1p=2s8gku87khS3<JIvrte_%R+HrD{lDv44m+|^_qz<bz4a4D3zA(wb
z|CtE}Bi|BXGTnB_7fL?ib}uiF0n%tb<}Y2TCsgf>4<<=Dk=8BbVf~&JN)VFp8QIx)
z9tepoU-L1&=oDQe90g=5Swvczl?B@ft*0ffp;cCnv4sryt;frI<82aHN^y!>tz4Fn
z5NNn+?ZYwr<t_`M=n2oQLo}l_n6_Dv-RZr0l|EH2M5qndj{s*CciX{8MfyKN9AX?h
zOyVbJS-;gcQhYMc=Lx*1h^-VCg@v(ZG<B%(_Lnjl8x!0YX<dqf&Ew@$_4bfl2?%@#
z{zt^hA#=>kNxJc|DA!xFJd2vtZ{IArHPcH<;?O=xIe*o@aT<=<dl1(k#i}13bKLRd
z!L?9o^AS5Aq(d?_*_LO3uR*fHEh2R<dC<(p1n-BwsbLuOHS2G#+r2i;Gi2@Udu7FP
zkVDj2aEpWl+Ot=EX4zF$xA(Jtj@o8RDF13o5L2jom@bs*)nPKmO=}6)j8+j11SO;W
zD(4<&;%&aAk=r<V!xafPkS?PTsB?vJki#ONfAjUep`9w67=Fl<)pCu53wC&IZ?d~0
zqk>G?22c@NiN{C)d7MS5BGOhw>n<TA<k&<~J0T!>IwphMI3OTEf-&6Bf75I9F4GD!
z0?q)iemuuPPAcsr2>@`?ey=eR<y2j*Ya;B27+Xy()1qJO-ysbc!$#K&2A9-hxY6P^
z04LFF-Evccu8VNI8ozv_Ng5Btt1*m0e*KS|govJ{B?taC<CoT21Bo4~6h-?e;UK;@
zFwJ?)W$k))-xM+4T8hcLwEeFYKLu+a?inDHN*XcTz~pfG&(juwqQx>^Bt!2kK<J73
z<M(4OocDIqY#IfIjx0??U(&_h-Ip`DR;<)`^J?uvyS6Up68jcY(|*=o-AolUk7yEG
zm%*~bnI`wrnu3C2^=n_Xlrw2O;2I1^T$sA+>Xbp0;6}%y$F6HNAP<RI-x^AE#qDD=
z7cMBCB8_h*Z96eOvP)S_&|xnX6|eVy%_{FN7k=<6rb+71K=c=6@uy?d@b}?)zyPJ@
zy&I{t#FDJb=Qt6Lwu5jAIgmliO;6E-s8C6kXrz+Z4xr%upXBbLt$pgjNJwoEONG}I
zB+aiyAIyjFUJTCs;^4Rck5=M?yv+S9WmQM*yvsv2hgtC*rM`G?hdrSRvkv-Ci*mx|
z=5C;OMizO5u|^rT3ZVIisFV_C3I?;9jncs9Fcay2z)G&$PG*wbcE`CzbqwqG@^F87
zk(>X#`dRA!6RZec`!rIftgU_MEvvM^j1#`Xv13#Uxu(d%0E^^?awsk#PB_EG#Hf*I
z^!m30CeSmnk7-%HcQfO`VJ8u{u+F)FWwnzlsA#qau8AF_CcE|I*TIh;<tc1|SeVHK
z-l~1dktu?Xn%I#w?E#b31HLez=zzrzvL5C+apEo}FC5Q`+@#bQCzw6y+t#+W6k%n6
z8g)T78^$tOSyTG~oL%^O(L-cpddI9mTTWbjL4i9a?d|W6ncnGlWJ^1LE$fp7o@DwC
z+38q@0>+L!qugG^o(o`+qs1I-jYL6bmzjg-;Q}H><VIExAcJH?)R<Z4D3;<atMVQ_
z>S6d@`P|2I)mw}CD3FzYk?Q(~jBbnSo0zPAaC7$<{yF$o1(+YCYjB}~X%_(|%(SVB
zfj1>Z&f+ZTH|RG30`o>i$PwZ^7*ctA--3LE$6{%~SxKzy?>oZv1~s2Og{G^7j9sGZ
zqaQ9}N8&=i{}bL95)x8itRcqY&tzLEK_Iw3N)9L?k>YlQ(4wr&*oA?NPe33&BLiDh
zMxX&<;uTDwOn<8UxPZ+6rz@pk@iCj1^?}RgD$q}ao~2qU_qnIIa@;Rk{N&#A{zm6H
zf_Vde-4Psakhxrb!NA}c2Iu^7Y@K}m^y$fjKk&)~gBjuvWQTGPe_5!@<CGK~505ur
zA3`b_9hBKS$tF`eXI%G<=Y~$h-PDzuqW!bicN9rhcL)!8R$jFVxI7YiP)o}%`>(al
zYN})Pvf0IWZgkBDLU!i%`sj&E^!qB=Gz#miRs{7)*^c<ONd#QijCYB0?wQ)Z!s6d;
zSAy$^h#_^EVvnL2tg5+~-xnPp-*j?&2k<aJK16Ij=$F2YU*>%`a2$aQf{9PfFeV`h
zo4HZiX`87#yQyP-Q_=i!?11KyPqy}_#yjDaAeJ0n3p9Jbe#sqGVKF@P(}jsX%h7qY
zdS2V_=w*+Qftlr0H@}sPPbY$gMt7K;l)HP9r%hZWn@*Wa@o!_dI+rC6kMMZc_J$l4
zrMB<3Wk6SI6=FLpD5*q~lOFfWOU{(~^@T`7bZ6)k0nPxJ(4#+p{%C4d@`aU@4C3qh
z@oa)mpj;ax5U^*61wi(le-ujCkYRiow{k^17GjUt%WBZ%!3cJqQrs3jJ*(6YoM5@!
zB7m1;W7d@uqg)d5^7FL@G0oz8lB}|@@`P>oo(>5QD+9=a*NE+3@a4=Up7eBg=Nsut
zKA{Sz_*-H7mogVFoNkLet#Z95w6|~X^;$gcu0Q%d$i~y^#Kret8W${{2)y;26Sh-R
z+upHwrCvh0^5#@L=WfI%0Y=?RLE-z2S4`}ziTo56Dnd>tc=Wd!isMxTQnfB4nNO6W
z5JWc|Mmc`ii&jxo4gjxc1(V&TDw&Yg#_IQK(KFA)I@I@g%O8m0RasVO*7GitZ!Gjo
zivnKczWfiWui$>ipV%$qxAC``P<6(~UmRoZJ_VP-`KA(<v@c#^5|bgr7PQK%kLvwZ
z^g*Y>x63a-p=mkhdg4*Ut>t~UF&M+Gde(nY0Pr{-8N7?fCS^%2i{S+v-v^T@;U;Hb
zxXb-E`^AybXE&FAO^>BN9O?v}3O*4n{sVTU&S<VvSf@>iY7C+n*dL4?b(8KsUsay`
zW{J(`k}tc(ZaP6FP|-F?_JjZ3$M;1ML!&UB=9rN45_;tbqZquuE4}$!{&lT(;)_7V
znx1b&WzR=ENPQH<Lrx0(dJJKu`j^=T*MnFONxF>I_p$uv`~S57-o$tdd0WO9F<agk
zl}|~z1SSJo4@m969xENEnuoDd%Bel``SZg`9Zm2aLlrUso#T`FCfuaof#07e>$(lY
z6h^&BO=h~RGiT4@ey(m+HICtK|5>-IxE3zZGBYQ*8B1;#8X6jMXujHY^BM6DiNh3L
zXxqjJ{+S5xhAT1~ojIZokl~DL)7qw?_$iU~1wBT7Q{zrb!6I_+DZzi$+tzBeF3)d`
za<Pvk)%g3|uFgK>X`wW)R!J4fsO-F=vN3ZiYC;?C%JZw$u!Rr%FJFRyCyPGQUG^{t
z&i`d_0S1Gs5#(M!3R|TxCsxiO^AD1W?@(5N(RPD{$~<7z7QY2oJzT7$1@4pQpk1!<
z;(Q&}dS2LaFWlC!Gm4b5PW#W=Me;pp9=$ojCF*riUBTUnUyr9;{<P)aVZB3C+=^1N
zXcp7--KJTpHvJF&gGqbm;lqc~*x`y3c}$@hf5P>LHLh49O%L4Q{D$M8saAlROO_AM
z-mzeWma!vlXM82QnZWX@H7v-KpZo~L^rL|h<I$sSC0}PYwq}!`*G78Rv&=I-J6wPB
z!+&I4xNwZ;n@s4=y)vwn5;j-f-#EDY=Z=Zmgz%P|K~@#ttP`TF6t6r<xI^;&VUX(k
zRGDyV??_?qSYeaT)urVZt2_<-14}*`o*6QA<Uc)H<u<PM>C&ae&le4xd)IR(Ruy=$
zV=)h(_*1g6p#9kkk6^TWR^^J6K_)DV(t2QM#ML(0aO7sXz**D3f{fwVP(|$cBgo34
zbZSa6;*|JQDY_){>`m)xLwj~>Amtsn%e|B<bVs+|$sK){6`MZpb@}T<uVbAvl<IFY
zlR^Z;VxDBPXxqJ3sU?N$x8-r+8{_t0i?S-uIIb-GGuj3PjZ891Ii?HvqMW}3zPFPk
zC^{ltK8N#}ZuZ!a=g9bEcHlcOV4lx)GbU1fA7Ewtvz&c7`#gtkGT{;|b`q<Zt&W;o
z8kjw65y)&{u-X^6$&UdEb(P!p1d@Q3YqqPbMxX!py(`Bc_e+uO)c(?UPM(&pzn)8f
zy0#`imL;jn00%SP4tS|L_Ii6Amy(+Q9xjSiI5GPESavd;<~UeJ<S@DZcuxZy#~Us4
zmqP13zzK9T@0!*GY^FW%UYD^ec6<MI`$%Qi#sfQ5E(h(SD?gL&<V0D%`Z`}d?D@JT
z>iwhLpNuIP#Ny(g9!qG{B$1Ss=56E%0@d^vo}s7&Mq>L25tdId4}|#xA;DVDpTcZS
z$j4B%b`T4aplrBY+z*b;C|0x=CoS!tG!jpX)oVXO&`7?!s=>NYpEO?MxpHpr0gXCJ
zd&*d^Q4St8CQVzjQCpjjr{qPra`$2U9R1yUjh_C8+mvwWS@YC6UvZX5JMO(E`g5zB
zl9wt7T^cb%Gj&?jIv(TI`F-j?13Dl3^al<_TAW;PpSms&d17H?7uJeKp$JYbWVY8Z
zs6T_)2(VFlw*#LLpCvkf=8mKGh^z>}>P4jM5a8w2Wi}+`6&h2zef51VX-FZ#ey=Vp
zoRBL%GO|!M_}`N4O1mRSg+Vf4UCx7j(Wm2vhuL0Tx`yH<tD7vZ&P|(Dk|D`Mahg+A
z`hdpU{Y;%7*=F5+A6^N$cjvu3n{iPxc?oQ@_BGqF^aV$?usr!7CGyVgu3b|#;7Lrn
zI2<eX8-03QtfqkhgdI8tD(!7;W1VudKTsW?3(y{*qy*pCk$Hy6r2L-`e;Zufcr$d$
zE99cNLy=VS=6Z!~E2oRby;l#8e;Xo(@yp|_k0g&@QZ?WrT%*rVjliuhre5N^&>{Wb
z%a=L>eQ4+rlDM4p6-mB@#_jx@vx`eLS4rbBF?J?j4%nQ!4ep}YZ*@MR?ak}g3q5ID
z6xgDHH=}D`m&C1;Hx?i<DS{mv(s|O8SeJrfL0j<?34HH`{_7qz3CAdV<{4zz%hFzL
zuScq_)%LSg?xDJDS}Z~n{ZEl+?sJk_{-<}AEYGe_h)4zsRNIZ=YfFo$6;JBG9}zh<
zY4zF^YhY=JH~PZeoR^tc@UBnd7?%DOls85`?D0;m9s7F&bPnKJb*Pq`0fO98ll+1@
ziP@TV&p8gSF8#Oq?8->g>$cNYv&;<r_i57o*W&K<LOT{^8(M+SUf(|;KfmxG#!4<R
zYRk@KPASLJl|pAqFQsUxzG+i`)7FRs3V8;iy1C4d!B^a$Tx6kI6W^2|pA{~4p1J+b
zp?@EW`W^q&TvgiM4ByT|Bnd%W+DZ0}D=<79bLG?$VtHP^W~_Cu=j`IC{H{~L-Q_%I
zpsfVp7kMwKy`J^2ju6W(29{S+Vl}RckhbXhJb56)RbcFHER+c~5VlD%!VcDjdw$NR
z3ftuh@IS!p^EApNu{!o#k5Z=gbR0ML&5e|Z#xa;vOK1#>gy{C3D)^cg_>X*>nah$t
zEN1@bnRr{oNg-&?QY@%z;yE&Ha*bI3%dKb_Z!;o1+Ox}Uf4?^_tJOla$;Lnumz3mt
zp5Yr>14N)nLlY#4Xbt}KNN!C6fO!`(a+;<fCl>}R<Kxk%=`~{r$^;!%YSEq-qt9~x
z`~E0Qs5!etk4N)67yH|&Sl+0fPwx9`M!FCjbqyUuiX_K@yDE?<>v0R%NBp{mMOpA_
z^}oKhtn-@K-&<_E{{ZdYBZa`REyY$+lebk<kBe6Qqa9iMEbTp47h+ssN%FtFsbXIK
zcV6J95Di=vBu)gI_YYBMG}|&FMs|_b{b`(ZCw4s{Ht_@oG@?(^Q&^;jEkdi>3{-J*
z-n@#R=m(igHbVh&TdFl&?mrJXNIuwJ?y?$iVSOVv;}<yNF5gspM)a9iRj8uaR3TO!
z{o_6N$W;gtcjyCJ!crjSXFkgGJk1DVizbBl;)QD2i`6)SStSp(pg_Y}{Eod4){s^q
zd?E!|gc%ln01o;qy^HyVIjJ}F%l=+<(&XaVmFLb@VfvEswz8f)qpE(KT3jc&#N6MK
z*LTjxNSclxEKJTgnqC3a0Zk_%q-;--dyLRCw{sqnm?;K}x7i+4y4?Xv1pRV5c2C_i
zDL*~tnUJXL5q+>VoUHcBYke~_ZJ>9I#V7P1oqcq={Gw*)-$T1~CigTpJ)3mF@<sZ6
z(T^WLKKDaIN86P*W8L=8N}aD7;`AML_V2Dr?Y;IH17zpkb#={{2Gk`2!ODDW&hoE5
z9+Bqff4(*Uy3wxrCp<2vvEjM?KOg_EQq5y7_0eJ{X8!Q^92#aK{fq<?O>=Wa1opjp
z^~&b%6L>3rPvsN&wY?5U<m5ssE6rd$gqs?N?N-H1`L(YDUxtPro%619l!98?d5r_z
zpDPw_SFd(BKg(&o&&MGOJB=)k6qf4U#V$+pz|EU1tob}z?u?AC6DgiYJ3eFw7w*=&
znWI_|gz1)0Eyn73Jk7cB@0VP%qJP-*fG2=>FDw$q{bp?XObjU!A;@0IV*i02%j7l#
z9|{EEHyH1iwQE^qsjBB6<ztoWhLYb;iS65nsYUI<>c9iI6mX<uV4TIjjX&WD{$tmw
z<wZdj+)oHZj0-RC0-uCzs;+>g;vX`=t5L@qr#-?Su-vj=G)<<GKe_VfvGQFze}yfk
z)aLSOM(%a`jEmKM&Ror2GiR-j`PKN4&c^PEzI|keP_B`+S@nFC^Td99`mA1#H`tMg
zltSKZbHpTuGYThuaF%xb5`;-|Vu2)-?>!NStJRo+PSq7oU1ERPvmAl^%6ToVK*CH_
zKaGu;5}Io<1yAq^wIyjgI4qg(H(h>Tlsu=Lcwyh@m%TR64TQ?vCuxauh<^6ZNKIN^
zUJKKj*unjnA9ra6KR=aw0^B_uIJ#LM{~J@^I_;q0qQNq%mdUpdvE4`xvXs&j;i<-<
zM&?)+167ZANaV-OrYAe8yx~?dBB(xi?on|1_S=(T!a>yD;nORJFA80uSJ$i%(bS2|
zJJ;j=Ao3;~<@(EXKDFIAGBk$RW`fw&#*OAQ-_C<1XR>>c$MNoQ^*<hDi(*12VwA2(
zpqnC0%mYpXEx-x^Q3zwMdFS{BTzW0hjcF!WjI5~YIcwP=&b`s5eDyjuMLE44H{6;F
znK%)!VcZ&kUK{u;lqFa2rV($B#MMr2Et94(02A0*8^(WLr8&GP@c_xS7dbzOg*Tvc
z{+6{9S_rA@T9sBC$cK)W`ec~fTjVmP0m8tq=lW?iOB0A()SidFe*>pR{pbf@pTk1N
z1wp81+h@<6YZc6|Z2vpPi)hu;=Qy>lf$kY}X;z%6n5~Y(mX%xI^bm@)&#7wBkbvh=
z;4uca%!2OgzzcWB+qXRAOz!@HVSnQQ&-=GZqc|IFNV}A}YOj|@zutZ)ee8&4+>R(u
zkEeTeW&)1bm)IRTkiklG?cPji%N`d7o<m26hi`4)o%prc>3P4@^XN>WuG~zQKihSu
z;XD~bu*W4OCECB`<GJ@%3@JW)EDKf<k41LX)83vR?nb(=o!B9R2n+mcI|xJ4hY_XM
zz1UeO0oi9pdU`No5LMoF=H9VwkU|)HgSs$v`<wNL+*WN`E7MuN%CCAI5NLrIL>rT8
zkqAAnFy2fW0rB@HF=`;<z&!wGhPvNV^?UD}_ouZ*);X^=$P1t4Rr;L`VpURb7K+wA
zeB?$;QAAz?0)})o!v4FxdQbJTk*kY6-wAycE=>qAGL9#8ai>RW`^PoOzhS`@@j}rj
z4k70S$eKw?Op-fvz^P(diU5^SX>UXBmD_8h_@hOsH=@q*w@i9`PR?yMquuU4LoZeX
zxoMU4>GRHe)g&#F8h)>{+zSd|bwLo;>Z{%9cTugQAa}#G*kky_?XB60pCZ`yY{`B6
zhpVa-TIf~OHGhmH?#VqeD1>8t74%TDyf<ef#ToR>2gFX{YGgvKj|1n#&HmZrPg&#%
zx2(^5lno@pjdF6d&x{!2CA{3Y4_W=meq&#-p;Mj>eTSXts1Ss?JNK|XiQ(f$!eFn1
zPW5;kcj!ruTKrRzwp=8XE&qavk#UZAJAG$2#n6X8%gv=Yy^<+>sr<r|Hxy)n-wClH
zDo2)D^bWrtT<p(EN-nsc-HqjIV`VV%IkW3{SzU5^x?Pi2Tj`-t9>Y2-{dVNca$og5
z8%b?}@smau)H<`bu6(SI`Lj40xU{S`^@>}7@@VUEn%O@bbHA^VMIBZ>ocAK**l!nw
z`3Ym^5npSSeqW?sZZrpO>_L%gy)(^x!2q4w+y_-E1b419nt`5x{#a{xT~JVvcSD5_
z32c}T+>m!-bN4?uIswx=wd!&XNmwCE<l}(TeFFY;8j^K6MnP6gpLG3f?LJ#Z1l7J@
zWLg%8Ez9Q)c=~{rPs1oJmOC^t(NvNpRBmY|g_hkGzObA}f_Q^|e!NRTPtV9{FG=SZ
z$a33-|44xSaZ_NcWuM(6KO^s|jyiDV9ed0h(Ra(@Z%f~8T|NCZOk+%~{zwU?lTdAC
zORjaD){NRCZTp&|*iYs5xU>nEYHMD2g^nBXQyW{$U9Mnz@rE#4=*Mb=JgB>=TIe!U
z6MdQtb}d~w(^~8{d|xEbRTOyTQ;#+*4lvNNQz`uWaA(2!rtIOc^c&NrV<r2H|9T}4
z4_9Y=d72}DyzcWJ)Hrg)Ixnmz&C1GR+!Px>pWEvSu$Ks9hBwdzv7rxN+%I+-GIcu2
z?)zy~Sd%$Yo6c&pXDki*#E9w$AZ(Hr7VOv-K&+4?Bi-5ZagJ-xY9y<)Yn$*umd!et
z*E}IPq*xjl?8#Mq0s9U7_sz#WtMt>m`5>M_q`tnvYnsPXYx{BZT@4N6>B-4@&KMXl
zfay3xQ-G9Z!V?vzWUPJa({zbgw8!s7W6wVgK2wIf57gWxBcrO({5F>>!?sqe>FjJ5
zuC=|xe|w8P)~U!|pQk@<Wuay0`7N(Aa+gN7t-g|ud@Km@^4FDmrL=x<Ms&kjQsP_W
zq*~`O0|O%e-1!$<wS7;P%dR}EBf)DO)5QR77=%YU4srzA1y!idnk5kthM!MY>jrC7
zWu-iHovn~}C*;YKhdP9UBBQze^rRR?G&jGACC?=?Ot*D*;<QoCuPl~7{+Oxz{qT2e
zI$7ic?ilXS8vwC7lAb8tWzY!^3r|oAR7B^<mjgLtLQp{F75b<P2AL3gozK>VN$?DO
zFd#Ed#n5hDis9zTVLzbZ%O4h;&exYl!OhQq<!-PtmW7<vAhUjv(C~DxZvJr=VFiUD
zsEvvpCqME>K1{*#2ZF77WJKFbc|Q2cr%=HibSJ{U|CK>%==EFWM@X*m1)U-{eg|Ft
z6+C=<u9xBpWT=JMT8$AZ`%8h<Sjp>-u}SP~$Z(X4R<TPg`d0;A$I^5~EZV<pH7lU3
zzynsHxzhA2kPXs!KKl7$>3LIAsh3o-TuN!L$c2I+8ar=d(t<>CY@L-~{8SgwcE#3~
z02TBmm##})3%e|$Ny`>>)pVB-rh;dxHDkYfH9&90l72qRb2@q-H+5x=zQOUvs3R<q
z59I1PT3avUvmTOlCy#25J){Kbk?hRUy;cs@`Yy}7oMKff$yoaV-!LIw_Kru!^ajz*
zo1rc(=l-lO)5&Zw?uhPW^9wm(HB?P^$)!b&ip4?bo`7Jif2WR#`P2;#`=jDbxg5>s
zvyLQW&}!~1uau7vH2hY!{7!~_P39xFze<y=QAjxd`J0NAO1#eW;f>-)M)Dx#0Q#D`
zQStPtrONJ^cSx+b-YQXt2Ded_N<s1`>+)z{|5=tMWA2YGpNTQrX0Z<8(75&`X95mX
zsj~;{?_xP(%HfrUHB`9gS}gBC65e8`63?KbNfFyJbSXnWxYv&Ern){924<y6NZ1p5
z<S~do94#k^coih47Ue}Gsd)K&l@EJpYGsUjSD7C;>=S=OrUBO>MEJtc>`?+aFfzg#
zn?^gy%i>ALelN;gdKEEDB&W%5{&Rjf*80o*4BhJtU9mMCE_=wAV4&4M5x5n{l03EO
z!zgO5viWFtrQ+<1s|aN})i9W8|L?=8k6-wsvCPp}KHl)@IfF-t7uU#5p)q)MHI7hW
z5Z&k8EhGB7puo^>YZ)0$l)QOaTFPr1^DO7L<-TX$t#2^h_rDQ7#{b`cCz9rakfI#x
zT@jWJo0}}~BPC(dX5Z`I6+x2yO!ZYHcHo)kORqkapZ5cg70H?l*f{ay^u?Ekl)J54
z6@BFW`KkV#k3?!TY_v$3w-z$@+kEK&E&*)<Ob$rG$+5X`cVw2AEegTB+rJn%$Vlrt
zpiujQsIzY6v34Yx;8#tXLoZ8*Te+g->j4V^S;3z*Mi-_c0=&9z4?8zYB%IHD0<Fox
zuX-l&d=C?VAC%dkW9{h{oPT)NmD%vWT33v)5l??@%6s(Wq`PtHZ0&08sZg`BCtmU7
zA))3~;#MSQ#1jjK1LW4=vKXP`*&Epyb2dgiR6i)A$J(@nyn~GL2hYm`Jw>)M%l?*P
z`^8if+^b0ZyANp#8pXVx?c`ss1mb`;@UW0s8MS5%X5l&=?<~qfpG5~YuWuwO&py0Z
zPlPV}tbhi_5`c55AC{0BnCiD<q`>N3#;!c*ivA+w{^RiQgPa_Q@*eP?#zcpau?%SD
z?oFxdbmfLj*xG2Zy=bujeSbQ1)kU@xFta3_wHWkY9x|))VL+@g3U*22cg#1MgQXD~
z?TivoY(B5<b9fEe6za{h+WkIrIn3wY-|L3$T&TIf%*DFMVdO$gN=hU&{+XE!#r=}6
z$UXhP)Ho@NYSOYqsu>yaq7);8U?=oX@~cTOS`$gqz_@*G-?HQgQqI5H*!qb0+3ibX
zy}Y05a(>l1-2EtlRGuF0!vf=W`{$3H;*o|HFKp2Sk-z=yFQiers#RqE_27Ql^&~MX
z@Oz<`DGAV!N^?<oF*2yH_&iDD!}W!!BPI<yD69W2+DFDs{yB7g<4U1%)}Ty4cY32A
zF81_i&r(lw$53iM=H_P>u`eAUV%thCJY)8Gm#F)KMf^b;QkZIzVC9X58=TA9=43zs
zTN)ObiLvOMFP}bge(Rb4WA4z@)S09JiR8WNnv`CMLBx)U%;bwm%ReN)!VR^}6`(LJ
zb(et_U>iqlZ0tKU@>v&7Vmh0JiZn$paIMF0ttU<QfgGI`)~ETw+pW!<f0Wo_fz+ps
zuDCtFi5*;uONCLycYJh~L_$NR&#u6KPS=sW(rx9bY~e%%LoVOSXV3dLOStThzxXRt
zp8h>lPo?P4kFD8liE4@I3(xwjYnGpXnJuF`J|<-#d6Yui`dCs!@n`9W-1Z+QhKvu%
z+z7%rmyzemIWZ9_<~d_l{WsEiFzzW5otu$qQ}x5mM4pX(!8V;+6VuV(Pv`Y`mR&g4
zi6C}frB%(X@1k3m&rJ)8i0Cr2-gBhEX=G<dGDW{l)70Vv77FNyv1Y(aUYFzG!I-m0
z|BK1&7W*_cwY<8SEbjvIoC<arDGIO4T(~1lzc0PVzB#H^Yr(`>Pw(T157uZ19g6<>
ztJ8(BR|^S&>vCD0x6L!0+Mivj$y|J}c{Wg8A1fJul(}Wunjh5!!^U7%?jB0`^J<F3
z6SAJ1W;Wn0aT^n6+;{iB#&dhFJbLxIi3KZhma7eeZl6cG(bB+VQ$uWD0WmaWKcMUB
zN!4xV*dUQ<)ylpP9=L%wL-hc(diL>RIt5IRWX0o|wn-tengyB{IPU9C{cUe8Dh}LM
z!0wxfZu8QiH?eST>b*N;<=%fbs##U$P-?vY{*ZT)(iV{T**?G1hgu5V--#RFp7Z}K
z!gPfWYaV*(qa8y8&UM9((~m25k#%**=1q3D9lT!9XduMDWXLn?XQ+n6=(xLE*s=Ue
zQTDy2?Nd_eqKOhMd+U7#e`soO#q`{58f~A;zh>H)8y);HGBtlJ<=wlkj!Eh>H)V%(
zPI2*Y?v&)0qN1~(n)9#hp1LVsbJ*1Uz7c6%yK;69G_P2E&7Yv4+Lz~YLy>ylz1_~z
zu?IyzDZf?MnRxi*o7r{YAq54*xa<zU>BDJXVtW0f<t6>N?%>1*FZLTH<%zdycR&lu
z&SOpjaTsP^m21~nAy|j|sOai)>R$E#9t`{3GpmdEY5bfiqR|?<tdbu3=|mijkG5z2
zF8M;zRiW6V_u?*{TnK61NsXvn;>Dqcp1SM)=?#;^=ksy|hz<4d=E1C%Nk-aZbnI1b
zZn&|KLFF;so_F6jRu?c{Nh6dJsN0WsAvrmyUxc^cc6`IW|BX-^+3mU#dtYqJOnM@b
zJm6FjiJ2SAf0VQuTxBo${d>MAYy3M_o!s{K-|TP=tdV{hx^dwfwrgPb8~|5%6|~N@
z@$vK5*-2uXPlB|wh?D2cixHETjDs_S%+a@nsPP1Eoy20KqvCxZzTNnLU43^vmVN(z
zg9aK#5>jT0>@pJBduCTw5h^1J5!sQ<tgP%E*+Q}_QWvR=5)u;SMhN*HXFd1xdtSf4
zdUfA*bve(^c)ySLF~l$yK-Oe-=S7%c+Xz_*Fa)}r-(@`&S}yS-^R#lXa<RF9GW>l&
zZ^c7WACNlY*Q75i|4uzWBpvb{{AgIp$SNHvAK7yYnspez0J{Kej<S$l^xetdTQcWA
z<XRiKrT>Er6@^GEI^&c|ucQ1thxqx!(O{NbUjC<(|4=5=O6T}o#yh6&F8=0W0bq7X
z-@fgm4wfz;FN79_aLdv$m5M`5WkYI&&-C4c9TW&0kk269^UHoWl|6Ql5kmQzF?v(C
zVRsA(qUc`WPJjy}zRybrs-8J|Nlf$r?p<$y$Yy+w=5AZ|m6Z)>ns?nqfxb4)bC_fv
z#X%&Hz??$(<X%6`6Gk_G{4$5Dvx2LuVMv6SK4HBDCc^#u4+$d+Z4;9=AP!D7NoY`T
z>~I#C?xYU(pM<w{bu}5};>08V^Q7aKm81MS#*h*L>G`dmN%iHAmXFK@7!>2{3TSDL
zw1u>xRz^wPgIqQ!z;ap|1ZX22;~m9BY>ZMB?X-ozL&*tPV|RGFapeV<z>suq;1qzF
zkg8*@Nn-A#pVNTIb#d@nUdjD5*JF%3fPeXzM<7f1%Hiukw%nb^o&IDu-4`Oeoj0kq
z_u6Vyz(+5B#ZkCQ>+1u^Nz>joKt?&OAzEhe)DD@-?b5$l*VatI^KmFa^I|(-0I{Iq
z!?XK1&F`*y9D$9<h4HR)%+g+xd6otFFMy-vRQO*-fCpTAr%xf(22@*4ItohxvFwGV
znSrWs@0BX=oT8#*?z4z7z;zLnb0sl9Ka=5+oe1OO+FFpbC<yiHJ?R0rmQg2sqm{+)
z?PFJU3=Ng8Op6mQ6)s&RCZyfP*8g$=j-G`o4H_GGZ0I8OS=q{pJdUq-c-GCR!OPE$
z;=jw>dV3TH3>mfQ$Z^y;IpwcIC)jd9^@3$4tqxapjE;Bvh#+L7fPp(xM^xR}XWb;J
zsK}mmT~Eu8=GlM~+q8IZy-awc4F*7;7H-h4FD(C9S$pP`u>V9Q_zNZdTnTs6HP@OL
zRpg);T-dS0O)|xnryQjz`j@f6WoMD&rmc2mY5VjAxXMEF!e)K)^WEy9A<EX7JEyO|
zos9Ny@lICr4`Rrl1$>tXN&6h&5nGd1o?GLo>zmPi$i|LkmVj^tdAB4cBRw?BRuBwg
z2tv4hDTUSTP`(!vLx@?Jj^`$BBa&+J@@((jiV=tiCwOW*4|pX5-u5tA@x-w0`uC47
zX|*IZC0QUR1g6zqYAeJ{2L2iLGS9x*djc2lI-v@xbQqEv%U*76ZI%4znVIDyBg})!
zD=SIw-C{1iV+8kN6NG{AX=xZn*Z>_Joq+dmPpBbs3zM{rjT8RA=WkJFEXY7Lzwk5c
zmnjA=f@ih``1%=hN6$W0<*&wZfj~;|?}<4^iB`ro)(#=a+PFhxWn`#7oJ&^MJtm6)
zk-PHlz><SoE!`w<l5J*nkvKZ6Os3L}0o9komlup!jGDvF1)A~r_1Y94Hyy)$2>qm1
z;;Hz!R3#9h02e|=%%Q>9ctq}grikLT7YJZ~dJCX(p^j7}h#`y1*o7A^)>^k*6PY3-
z`b46(4`X&ulwK**azq}L*7F-dtH~FPH4O&lZq&nLIdS#dMd=Jz$B2C)f!#JDxYi>v
zahmd+kUL<B5*D;w>~gNNKiwsX<S*Ragm05nDnttQsonE6xH!a^xmdzUZ{KzeH#_`*
zTa9IChy+bh(G?=`n2UKQgf1|Wphd0HD6kOt6SjgIppJvB;E)x3Zix31W;}UuWVBY5
zgN7ApffNBYa05hF@fqj~k|abY7s&zG6C{kGwHa5;vbGYlEbB7zWIoc)43dkS*I87B
zb#+smUHf-##gLX$jmGM!hld9csC2mFsl|PTxPp!V;n9Ji%|^+^ld|bQ7YLcJu`h>8
z{89SQE$l4ehl<?Jca#p>{-iRP?HPJ{{0+59U`y7WYtODGL3N(UcR`RI7wy@as{!z?
z0AMyTVRLbCLm_qKDHIrhR}oEt7R(yF+0PfoKRHUcVZdI4>D>-W`Gmp;nUOnGs$2s1
zNABSqwmPfZ1SEwRan#iG2y4N(4eP%LFr>4ybCWf97Qdc$uS#k(u&ni`8)Ej5XME|G
z+p`U*)y^Pi#j{Gn21Fhfrt21a<+tW&HFQL>I}i>_AxbRE9I3zF<YZs=fbeK^GADHe
zK!Kh<D+3kI6=L^6(xDu+rQQ@RF4{Zx4ySaT80ppFMIwd=sKr(15AQD<|5*U-VFUJ$
zHBjoGarn?RReaz}e}y7;n`qYFmP^I;FW~I>3d`@y7X<rN-+ng`Zd$CwB9g{LlNCs|
zAF$TV#QQ`Lg}8WMXKF))$4sB}Sld6iX@?<2!JR~48-m`y-xR{GEET@}Rm5`iZ)WWu
zIFVc2;5+nPDMyW=X3FIT$9Bs{Vj*l`Drg~`O+4fd9<A8cW*&W%9Bm+c<_x};#taHB
zqzIouYT^;xsvwx%2g^jzbw;wh-}s8zp?}ZP)#oS=O5PW`)j!cOtpfg5`#}>cZ}+wl
z7Y$i~i0%&Iko4CdNsUS30XPxV&$wudFlJO%(j#*fSHx?M&70c#Qe4jQLeCCZLCSGu
ztJ`qvxZu8*mxmthFFQQUb8X>kuSyn2nEZaFCe<}J-vJ$cjKyF`@d<|v^GTDJAL|b!
zMLoFVLbr#L><L6i<&4v}W7W=o^Z2rccs74u*0T>u7)d8>`wIH4aQ3bDq4Xuh#GYe2
z5@t?axv~R?7A{Y)Z6+rs;P1Tbz4defJ@TN{q3qfK9d$JQ4UD1VMT=?t8smAx3rRNB
zD>FE(6+t*8VqgsJ@OHF}3bI9VV?!VvSA<bPpU}*}z4i3lU0BJ4iu@?J3#h|2Z4Zop
z`2yXtrU;`|qi+DwNg~G<46PgWe8xiO^l3>ZBpW0g9iJGY<6^uE2o=X|kM)*nhoNLd
z)U})vQ=m5GiVg;pBgwKWKOYWWXz$|H+b)PC7!)w6C*BKgA4Qrn=tLx&begPmjwTH?
zHMK(`r=FK~0#w)@B!_gt8@kiT#XzbgjK&B&@OWab6!*WcwDUNX#O9r~gr3s3(fkFo
zYC&&QOw6$fmCu|fPIyFZ+QGglm}YK|aafsoqbx6dNBOtt{CwEl!463d>;g)9zRy?B
zoeO-H{pj6!0fzV!4t&;SoZ`C}EvpS@cT|eJlvmqNSCC14CBba~?3ncQC?MmGUpi|6
zlsvL%o?qGnkq!`z+=2o*{TQc3{@Lg6*(_l+6q25~rzD-k0Liada+j@eZYnRfErO^=
zT^m;RS5#b_UD+Bhd~qkt>2?7|%>ki1^_7wsb9zor9B_r5sjm{bE%LZQrT&dRb3C&y
z9VIE8H8CWE8H<MMF@VAP`q3=Lr}L^+qeQM`PZKm2w8MUzvtiD8!Xi>R){0I+Mu)~>
zr#(l@H9}}0wbWRD`mrV)&9RaD0Omy87D@~iJML2GqY^NDquvxTWnR9`!x!Mna(U25
z!Ps~*v$Ed2ag)>d-hYd3YgQ!4!Y)!yZZq(U!vs;O(XNb5WxI|=_ne}l(!M;qrZY93
z_U>)sUvb6}>@x!4Dz7Tb&nL(bq8~l|m|b^*xzT1L@PPY;e@pZ*i=LdN{@F0I@bbsO
zw+rS^y7+ad=?0n|q9~elF##r~2rwOC9$ctzA!S#o`k}hEr>i$5qp&a(;7pddcR5Hf
z7F7emifp$1iCkPfcTb6)y)^N4Mo~dyDjd<nNU#TR(;0usrj5b~>#?*{w`q7cMz7YR
zZRit^C<g1^yvYJ)7<!4(k=8q4|Jr{2%10bH$1lGZt2wO~)g8Bz^z?fmTctIV=>1>W
zy>?r-%A72VVS2eS4qZNoy%oThBH&*ztF=C_1xvYt&P37hg(*>hNQyX`R1gtXUSa%i
zO}%^7iDXkj=$*scC*ho!$)V4HVKT#AE+8$lMxIfJy)#-_klgm{)uHy7=<Dau83W=t
zZD`ou8@NS6qQ9^Dk)5p1LChG=_x2`<d(a!l_C2<uM3*yKY1`6yWYj*B_OpGM2({{w
zTwAKHtYCXXMw9T6o=HMmBDwsznDAZ&C(|0~G1#h3h6p$lhx#1?{|6xJKatusHfAN3
zoRN_Zw>PL!%PY%62a*%A3Z|xK7Tp9G)r}N0w;}EhFd^gJh`2;i=Py5mmp2VPzQ$BF
zc$Ej#;SPq<*yL%p69cVCo*`BnfcLYH^^^<^;}}@Ug(+aZ?YB!x{>+&Z=ru2G*|Lei
zbk^5~8XjQU3A?{Xwx_Yz3Y7)j4iEu#`Km&h2LFA0oh<0~h8-XD+JS<|v{2u7xy`O@
zWMTQXJUi`P-3@xdbdieeY1~3tV^;B<+2ztiDiGL4fUVd%oIrn7b#|tNE)*SMM5@^>
zl3TsGz6fI$Vm&(yzrx6_eRTFg?h&r&!&YHeKRDYa{S41@P*JZ8VI3{vn;xIk{r*NL
zQF9=Qn%eSs;tu35Hro@v`sL-7LJl)$VcY{GMSVL9qVrs?Un5~?j(3JG;d#f)th&O`
zwoC7z_0`}p7iZpylX3Ff1jC~>64%W0bJ5*#`u9eKOdX_+jgRxP&(kaDM73_=qOqAZ
z70e(n1R=xn(R=^7qeTWhur&g*gLX^luQtXqldh8i*Md8uu!7sqyGgEj{wzz)f{fDj
zqj*!JU&EKGhMV`hyLDlw)r!lEFh+oh52`K*#pN(Yv9cOTqW|wMXNuwgR#9TrH!+7B
zNL`WkC+`W&u;I}ZArig2BvbEIt}?f;VuppLj!qkI7)FS63KkfnBmE#FpEz4DsZnAn
z9qfn=Xte|Ifn|>SjSNA3a!^a9)zu|oN(7=->WI!Uo#CU&HA~m}RFji#iO#eMWs<V)
zJdKL4j$9F-Q`or2pvd%PZ(9K&Hks+t#BA*@Y`yF%(t9T(MupO}Z~6FaX#BMBq;BeA
zEvQ);d7P0Js0~VW2H<Cc#A8PpZ|a62>GW}->tCLQDao?Y+(j4UR6D0=%ZJ;(kAwu8
zgtIa}w$_T2@Cg#0JSMU)w9S9=7g6OLlA0YSo-Fs5uH+7^)|Q6V4z1gY?4HAvW}ik{
zDVfFpIbQtamjvPYknH+#_;s2bpjhHmLOwRgw_W%2Eit<Li&p@Xv@Y8hC(Pu~5NE_2
zXzKkTO2OeZS}?+8uxChmmQMB<>y{mR{UKs1G2*TMP-9zUPp1H$X8}royjzq6IL595
z;<`T5uYmPhpRu{&m68AhqHJK9LeyITJu#0RnSF15?I%^}p-n2$zm~XE4#leMi)EfX
zlx1ym@z>B9Uixyz9X0l%CtSa0I|y{Y{Juy_75I9x9Es?t=KLGYKeYspR+pZ`m7%B4
zIWU%9R&d8$ApWhVG(Z0?M9v$!iViz6AZNw8&?LK=uxBto_$-Puqs$xRAFL7t-Ke<V
z(GnH@sC}_Ca02@t!;>ByI?=e!(7vIIeZ+8>TzCuMco48*;`2#*eNOs$N5@5sc2Nq?
zHSYQ-gNaRhnEu?asi3@vQf;7!JnOHxwA3B<#J_o%ZvX!)0CZ8=xVzW9c~jo=eIe>Z
zC%X!AjbdUHBw5<07Oa07u_S!t;%vFcDvpMp)cYY|V7klbsMPLluit*`0{t2GU%C0J
z2C+Nzbs*2n!pbz~mghxP)GO0qgBO;7Hpc_5_zOGm8iPG#!rypw%S-2dU)ZRUBa_E|
zzFBD7xqbK2Y(xUvM_VVaK5}_Z1LOus7G7~W_wy$rjCac`s@M$;&kG)s%!OQre0doK
zFbR2W22Zcn&{%@Yv)8NP03^WoNP@i0qEYfFA0lo1S64}!-adW00oQD2L2i6E+cnXS
zl{V_fSy=%WQ{3-ExEUK8Ykk#u&Zyi*5YdQqn_|t|P~T!o!*udwJ)%RT;Uj@a<bYPo
zG&q6743eODVY=`qmGgS#9UQMAVYL4)Piq)q1!5Qgu3r6z4}Wj=6f33TccXL54-H3|
zic~^>o!)Ql^9HG0_=Zz&A2!=4o0S)2WJGHVDBG$4rr1GYbiMqszqBE=u*T=*KSG%c
zC<Lnd46-o!6PYBtdlEmrGCespys2els0QNAKcC5d#?>$sH1r*pIi2#h=*jeWEIB%=
z-NEv(SAOnpX>ZC>0SmO=-WY?95|bk6V$pA$!oQDc9fyDZ{K`?Q_pM=Jnn*bWYbm3D
zY;k2<pzMWL2iJMr4s0{Rdmz%J$5U2LB14o99b)*+Z=>Hn)W)r@Zo*GG>vmSpfwof&
z6|l$8Z)te8?FYP*z>UDRyd`hr^jCs}J?;4;Fp5(`<KGJ?9s@!~h|wgHQ^Fo<(jPTV
z`sZfshrs6-nnowJ<ln06sqd>a!m<fTk4Q}oH;9wEkzJ9s@1U``b`%FZ3UISn?3UUB
z{$h@B<i76S-T=)UHQS3L-n~pO$m{sdW^qlZh$b3}1=|!yZud_xOz_G>Yz1aj)jlF{
zrC7KDkH^oStE1Z+$F5q4Ft+ik$3;iW;Y28|+Vy2urOAu(@JSxM>n`Pp>_CcC9;U-!
z%0t+wBhwiA*vq;oAD;m~wY^KN{3#Ezv$x<f9+z4a|2b3d*VNQ>f$w2{zU4i-P9rc4
zYSw?RAg*IQnaK8#{Q3TTMl5v`rkH;#@E`4pR!0UC0z(JON=qlcF(1`7;x)ho3if8m
z7rgi4+Gz2K$QSQOBp;wwG11AS_5AuW+S(5(bm<$^x!7@D&7`ngFz$+%08)iF2(xzh
z-3cA>y{OC|`R37z+KY*!$Eq1sl1E14tWR{e#A-CC9=kGHe6BX|`Q_}A635O|&?fQ2
zAr5D7X6mL5LKg@21MlZJ`A~7>suha*AM+kjw!xG4kLKC+*CWJr-u>6JR~-kgg~2Wm
zQ%~zV34?WY`>yS=v+i&ty??KaNbkJ7;MCvqIfEYGgWSJ8D()WoKAdr2Z5vJlg0|=W
z583|x(}(xQaj1X>7z4|uAH%zb)Wnz>MXra3XoAiE_rSNNJ1v@1Zq$ihp=Y3{`P9LP
zf0FwybD>?`*l@tXd!1}8>rG<~KLCc58B;v7w<5n)o|=Ew<hQRnk??qnxxfU>f76Pq
zT;&65Nw@+)c0s?CafV%^u%mlY`Us7(o$MeZ*DVdLvmy7Tc4)`gFWrk!c8l2aDj}L}
zXEr;{PNc@&<vEd9SU9Dl&4s(C17lsJrDv{P(bSZM339zWTjZ3EL&kNh^4>bds2}e2
zOLbBx_!O-Q6f$)B2l}5s?@DlLq$^W=O0mUn_2rP{i;Qro7gbLdv;gS-@Y&^g>%i)#
zt&m~+oHD<KIPI@fOiTa7HFQGnQAN;h!7*i!3<iNM?bHv`5*rd50oY;GNF;wu?DjFI
zPb1m1J$%Z(Azo+8gFjDgRgg3}U+wG+v!Vwc^7rFSi-Du2<sM;iSSPzCZ2KcYzG-}J
zc+MR)DFantx!U65;^^Y?;Uo8>L;E*Ha2y4@*PlI7u;<;a3%#;seWB@@Rgw*?BwK--
z0<3lF$#$GptLj#nKZI2qUX-&>l8Nn!;J~FP^6+jXh~C9j;l8V@_pI)2*$ikMg98T#
zhqm9Y63HN-ZSrHY>A$H?>jPx;Ceh7CrJ4z(uN^_J?ByVh&RLj`!RBq|Zd5e~;0~_F
zU*B?a(erl0uY!imI!`{RpfMp9LnP1o+O86}SxW%IxkW|8U^|cDf8>DaTIas!Pxci?
zY?du_>O45S5Owi+fCj00V=al>6?E|WxkH4Os?f>GMR6c{5+<XcU#AT{LUW6uHJ8Nc
zgYo1LQ_UexS>v`*0t>@d_B5>|F;)q=Kl_!2|JtCqI1p+8mhmTC?O)vNK~8Cf$@T^R
z2rla7ERxN{%vUbvVCWK~bxuKX$4VcrV`L=oHRYV1p1)>J1`bFN{zBco`FVLWGpplg
zQc}ug+uiz=Za7`gw%6%BEbxT!iNWF1?26u*npc$+zlAs3UxeGtc|M*Fp>rQT$fr}I
zi&soOpqWEkWDxXfTy4c8SrdV9BgEyt_S^p&vUY9OBRDYSxFCe0raR>TbSH%0Wc8(s
zZ?_qNKmc6xm7`da=kK{xt=Q<x$fpU|o09XlY1lxF#NcKFl5R6d`Y__dp*8eR6D9wT
ziHz_N>eB*6(0PnLYG)Zm!H%)t22|7wjn4kQ(j3+s<=LolU*%Jnt@OW>C)dF8e2pxG
zs)n^Ikq@PIJ~C2Y(xo8w_9<B#Mw&c{9(jqrvTrDPd-_P?!4>(rR0?mo-|zn!IvJGg
z$70N0Uw|ehW|~b}I{a7Q4K%W`Nv|@SCtY;Ky`>OJyL2<}a1agckBqKQtj83tpC4t<
zdUZ`trzY|ff9k6sHXhtM%<u{MQVOI70zLU7){JS-@VZ18VMOz)It)q{(EO$rvTyF(
zzyC~Z2-IIQj-mGxkBf?)Kb{!9WnDPf*lOg7<FG77LeX@^Aj88w1v}Ed_R7T<rGUR6
zC^s;mcO7)via}Q9liw7eT8-wXf$bsSU`h_xyY;7pMN`VnpNqF%9IYA*d1=kbymQx`
zC%TbiV)*;Xn&Bjyr#2$k#E9qSqzS&=MYqj*V}3r(&L9P)zsF#_0M7WNq@)Yw9ggf*
z?(`x7GbQsjJ8Op%AR(^k0NyiP8`ZMPYAKPG_0pPYp-9HT9HaWW-@h!-ki)U+RkZ@X
zr<DTvUcRL9rq;PmGl5imUfP#XrjxNGJ3EP%0+l-XXbxN1D-8Y8Dp9xP*_cmUhaWBd
zC4*v?#a8Qvb~!aQO6QL8sAK6A2RU@fu2k}RN*%z=sH{)YE<yfYEC9Ycy>?6$FXfLZ
z92=dxgawpMy=7+mdcpCtn&Z_hrruQ6F}JJ)7^-216YU+;Y=IEI9nd0zN#9u)EwLZ1
zq_D0GUEMnPTfjSSIe3nyzOVNXUH|E@t_YgO;?TPB^>6PtW-6Y`qPpjY+F^Jv`|eJ2
zwn$~w3t8q@srJs16Q|$O=@xSVZ9Tn~$(|gdWrcirIr+H_$G{llJ2~6WXJx}W<`+F4
znH;@f)z{6pmm*+x-W~QJ8dH(jy4-w+6*5SU;9Col^WO4pk|hvPP@<dgVFV*KiX$!G
zFH-${g?O^>F}=p!@!i$aE{VFr6DD8%y;9ho*3WgFz^Pt>8DeRWd>s{-?P`~p8qFib
zIVXGOF{g*AFJLxsIJ+r~Dhqi#=_ySey!)N4xHs`cM;IBHN~z8Rmv&inDRY@D-6wva
z_hlvJ%_X1N<+{4rHVarXi$1K10pR}9+P4wN#nzmj!`&yzdN3WtP;I7qbSJh*8?Dx=
zFENFX{Jk_vWK3O{)E#!<+}&vn=ibyG_1|x5xwPy6(zwNswP@V;PZ=USKhvrgL=K`I
zn9;X8oM1knOU8?~yeB4CNHlV<)Yng|>~s_uCp$_cP3)LFy^f#Z$Fe7{X(`7Qf6$WA
z)@}=s6%i33Izy`LgsdrY7zw5;P}<gA2h&gsj-f)A>qq6y?~;o1IaqfX#tG^5B`%R2
zy-C@vz@3-N?q5Wuo0JdcPtDu+&k$^b`86>J82w5$?0B1UZx?NNJMJ^(+E9zv&(?6#
zxz(FO{Hk>He1*{#9_n9@`bxVm8)^*P(AV$D2CAY8hQuQy&o-033G1od>D&~C#mBp0
z$ocr%6xZdHfp_neU{ViOWm;jOdT6V!EnfSG%i|?f{LhJFwC|Ny8*!y9JyV()_1VgD
z@KF0v36+EK#FT>d>7Q-+<52UPtx|t$SnKG7x0MEOAMuTO<n;LTOeRI#;@lXO30NE@
z9%R9C#cKTK*IZrMJA~|Ae{r+w>ti~qk<I=V>MFRM3VJjSJC=G)gJ>xs28HDV>Q(s+
zS0Vi+CJWVvA3r-OGwE|?oy+0yxi|7T#j6%$MpW+(2i%$T1K@VWF&);{Er?L%>Khds
zBPOO*%&fEvdXXJ~t8U&y(l!oV$pu(mS5#zGq((+`E7e3QOS@#wHrP2hG(Qjx6^cz2
zoSW*C@aI@94*Sbd8}Ro=kcRk^g(8R97FsQgom-7~-L&8eqo;pmWMybwDqQj(6|dzE
zGhD@sb3xLQha{B#E$o+QIY0r&gh!(P1fJC4Aw&uS--V=r_MiTH9D9nZQqkCbm^oFT
zaiE}kAaS>PQrP@HmC~f#P#ae<6eWa!P@7w1=I3}<Vh7edF419loRMwXC~|+6EN4-O
z-EL$+wZ%&1aB8NR<`^VF=>=tXl54N)>WN$&oG4J{=tNY!Amd9+AYLPH=aDxG>(^5|
z+=TVmm`#APLkX#v@TGvGn>nb8n}K3@ZUtWPDzoR*%8wF0DelWP)z%Nz_jFnkvvI-?
z1@4igbcIpAA()1p4?o#<QEE<l;@h`R5G65-UzHEN@4HBcAcN*z-l0&Mlw$ioSsg3>
z5gG33Z)oa{sdF>F5dH$)erRd`mbP`T@LK5h(2vY5BbGvIx$E+ABA))>?gu<sx)wKl
z<fBC97u+K_(8KNH*?j}kVvX$3NZB-PF;+*qxDu;5YdH5oVD`aj5jHfaT-w0M0&~-d
zSL$K#%6pSn%6khF_tbU=nCAt1D_t>uY2zhTcdV2RT?>IE@)#C&z8{yvg!S}k;T<~`
zh@@(TJM_bgD?oZD$HV9;t;@DI45gAhj^Sk*5d;^8Jw=B&J!WqV9TPo9y*)@N^9I+%
z)ZA-qC0pBl_xF8mm%K#yOA;!0qFl*OD&<9LjGz?kfoJDZ*O@TbSRIxA&2F~#)eufU
zr(&%HuTp%RbFL~WiD+0()S3gocMs-7L|)(>ey`a(n5J&`xPOH1Z!3Vu_Gt}G9_+4m
z8S4Q4Q%RNzuAbhoVQaC9yo_G%B@x6ULUlXI;MsWeuEmvyo1{{AsdQUbnDi)0=jGoQ
zaIdn0<}LCPWq|C`nis_#ig-hv=Sd@$nkkY2hvXOs9OrLrW)d~fLA%e}aW=0(*38VT
zZLB@^w%!|yNf>~eJl^$lYwp*HW8q;>Ri`^n9X#c%MmbxpjxhpirI$9<#bFZzK$y>0
z^ylXH--hiMWAMoIV)k0m^UE;4qGi|yRzwdU`pMU}BA7sp%(|HzC>-{9|K<%M>A;2I
zw=%}TaOn~!AGCI&u3t=r%a$#N>OFSB+hboGDwgq2jfa-Uy?<Z)wQ_RMYh~+FZNmyP
zC?LBtvOa~w8TZS>+8Il*Z&v4X)|*HfN<g@eTnX6zhw4MDx*xw~qc}>POOZY5a^2-Y
zzFkOe1h|pS_Bk2#$>^<;4&~aaj<=o{+%n=srFc>?T5$R&@F+aTpyq`62u+oHr}bC8
zq)9e(K_7g2?TB!hKjZ&rC%!wjf2^myKRS40s%QhYguga-3ljRGSB}+9%Uw;9NY{Ee
z&}z1B{VZ%%akS)3qqnX!)7kNo<9YcZ`P>;&YULVom{2r;;$wTct_|T^?=ktv)vLRA
zUUJDyL5Hw1M#Rck<$yhmiNc{=5;56vGi9DFwb2}Ytldae2F!y@xHm#Tll@3goMJW`
zE+I;%ppi+fN=&YCK11lt&2OlnLySk+J2bsae|OefgST^Wiu^#ZyxVN+URXjqs0Ozy
zb2LHyy42<R<YiykMbJ(;2^B)kSnM>yW(0L9teNpflEmlu5L*#9LG|zR{iNmp2kANi
z7fCoc3KhqUiS18LmS@xXLbnGP11xF|G4DLBtsOyE4lEllR0+A#itK27h(2t<uI??w
zoM|Nw1g>yJgIo-+EOF-$nn~M7+%I1i-BN`UQk!x{q&(494$LlletzsMc%~DLy1A>^
zv*%rk86?D`Lp-dRt!7tYqNk9RCr+KId-iP7w{Soon2K0gjWOH}R9V3Lhj4g5a7YUa
z3sf%?-{gyXA_<hGKq1x&cfU3rXV$I(iT>?DvYm(T%R&DH1`N>3yx%xsprOQ~AU=!g
z2C9(1_QQiSq}F=C(zUgT*cb=~>Mws~Bccifa#)#xj+>kJ>98Xeg#k6>sE$;>5?V){
zVE%>tD?0jYmUTQeq%*>yue&?p%r|=jN}~wv$ZHbkbZuc0^EodkWh4kuUHSQrgpx3T
z-_r+^OhE~X?4(Es6?>QcXdlsM%E<Uk=>7XDh@wVOV2YeqUmv$e8tkF8xu5OwUSiiw
z64Qi|1<4AcNB})~K5%|JnW!A_L&$0mJE;CWoBwk}s~;_sO4a;>Ho<vd(0hFiQ%^_O
zUjIO8U<OX=soRFWwxg&QBymgn%gT`U{r9NC7f@$^yx6P>ZY**(UqLJW{N=T|nBvvV
zTl`F!&o!10&n+)C?t1bn(cwueE;&#cvUaQeDPjL}MZ|M`6b<_V^42Q|KQ9klzq7lT
ztfvdrsH~P2|K`n`F`e*kM=ImpSL3w%+5f#)@ZB3l1g1`w1ftl$yP>0}Pk%>AttN1i
zJH+?hRO{GPf*m`X{M|DVLGXlp8S@UU$mr20w&z{{_wkXvb2r;=1fr}T*u{P@=neuc
z$(gV`fPtkk1SW^%dH?s%XK$NE<Y{m56Ejyn5bXvB{o8xOn}@4yWmxI&1lu|L>JSX2
z%Y9b~`v|zH)m+U=0bc@@0bIJ${r>mKSF3z{M{Pt*1SrjxQUO^0h5<BEUI#Raip$EP
zF>}|^q2A&rlXTLT4&7$>$)Dkppf?!tzFlvqUr`F?3CY;tN!9GJv*~s~fXro~tSvGn
zf<|ZJy#{Yp_Oy_P_Ds-QQjgMDp0_}pra9;P+fTZh+M0YwZ)7&<oR<8&Lhm@t;?+C+
zTLly#jF;fe<-d8uh7Ars_`-t*hla-I=S5&d1^bnR)KqV0_ZSXLlWJ>+oM8qC`s0ly
zIoQ_~SH--1>2SD%?SsAO%8K-9ef_A<KYt!a@VquY>WA$288V~<p3cvIF+1zT!ot#R
z@BLzMa6B|Tyup3CtgP(v<-#1^j`@`}UOql+mvY+>w=bVQ8NzGI+4(ejCC+5cLn@`^
z^R379@NsTwZ5{)!pFLAXW_ej<<xjJB^|iGfA3uiEDZt|=At^OA=Y#!4yRRATkKL`W
zEnEVCR#smRgd!S$z|=v_og5!h{Aq8KH*`}maND?N71`HUqpq(083_Y-73IWMdlwe2
zL6=a5%i5rTb2ssil9SWpmkkv^WypwM-z03J#4oo@PMdGUuiBR=c!~dPr{P0<`l+JB
dCB$!YD~1%rWXN^~M6naURZ-BCe<*A2`+tAh7sCJm

literal 0
HcmV?d00001

diff --git a/index.Rmd b/index.Rmd
index 7f8b322..b2e7ff0 100644
--- a/index.Rmd
+++ b/index.Rmd
@@ -24,8 +24,8 @@ is.html = !is.null(output) && output == "html"
 document.body.classList.add("index-page")
 </script>
 
-MI².AI is a group of mathematicians and computer scientists that love to play with predictive models.
-We are spread between [Warsaw University of Technology](https://www.pw.edu.pl/engpw) and [University of Warsaw](https://en.uw.edu.pl). Here we have workshops, seminars, here we are forging new ideas, creating tools, solving problems, doing consulting and sharing our positive attitude. Feel free to jump in. 
+MI².AI is a group of mathematicians and computer scientists who love to play with predictive models.
+We are spread between [Warsaw University of Technology](https://www.pw.edu.pl/engpw) and [University of Warsaw](https://en.uw.edu.pl). Here we have workshops and seminars, here we are forging new ideas, creating tools, solving problems, doing consulting and sharing our positive attitude. Feel free to jump in. 
 
 #### Mission {-}
 
@@ -33,7 +33,7 @@ Machine learning is like atomic energy. **We develop leaders, skills, methods, t
 
 #### Vision {-}
 
-MI² is a group of experts supporting global initiatives aimed at responsible and sustainable machine learning. We support the development of future leaders of responsible machine learning through internships, PhDs, postdoctoral fellowships and so on. We seek for research grants and business projects to conduct both scientific and applied research. We develop and maintain software and infrastructure necessary to build responsible and sustainable ML. We develop cooperation with international teams working on similar topics. We support companies to implement best practices related to responsible modelling in their operation. We conduct workshops and training on responsible predictive modelling.
+MI² is a group of experts supporting global initiatives aimed at responsible and sustainable machine learning. We support the development of future leaders of responsible machine learning through internships, PhDs, postdoctoral fellowships and so on. We seek for research grants and business projects to conduct both scientific and applied research. We develop and maintain the software and infrastructure necessary to build responsible and sustainable ML. We develop cooperation with international teams working on similar topics. We support companies to implement best practices related to responsible modelling in their operation. We conduct workshops and training on responsible predictive modelling.
 
 
 ## The Team  {-}
@@ -642,84 +642,79 @@ LinkedIn: [tomasz-mikolajczyk-ds](https://www.linkedin.com/in/tomasz-mikolajczyk
 
 ## Open Positions {-}
 
-We have open call for [HOMER](https://www.mi2.ai/research-grants.html#homer-2020-2025) and [xLungs](https://www.mi2.ai/research-grants.html#x-lungs-2021-2024) projects for following positions. If you are interested in any of them please send your CV and Motivation Letter to przemyslaw.biecek at pw.edu.pl. We reserve the right to contact only selected candidates.
+<img src="images/positions_mi2.png" style="width: 100%;">
 
-### PhD positions in 2023 {-}
+We have an open call for [HOMER](https://www.mi2.ai/research-grants.html#homer-2020-2025) and [xLungs](https://www.mi2.ai/research-grants.html#x-lungs-2021-2024) projects for following positions. If you are interested in any of them please send your CV and Motivation Letter to przemyslaw.biecek at pw.edu.pl. We reserve the right to contact only selected candidates.
 
-<img src="images/redteaming_mi2.png" style="width: 100%;">
+### MI2 policy on benefits and obligations {-}
 
-**Call for PhD Student: Research Opportunity in Foundation Model Redteaming**
+We are seeking highly motivated and talented researchers at different career levels to join MI2.AI research team in the exciting field of explainability and red-teaming of AI models.
 
-We are seeking a highly motivated and talented PhD student to join MI2.AI research team in the exciting field of foundation model redteaming. As the advancements development of AI models (especially transformer based with attention mechanism), it becomes crucial to ensure the robustness and security of the foundation models that underlie these advancements.
+There is a common recruitment policy for all positions.
 
-**Project Description:**
+We are looking for
 
-The selected PhD student will work on investigating potential vulnerabilities and biases in foundation models, such as language models (one PhD position) and 2D/3D computer vision models (one PhD position). Redteaming involves conducting adversarial assessments to identify weaknesses, uncover potential attack vectors, and develop defensive strategies for foundation models/transformer models. The research will explore various aspects, including data poisoning attacks, adversarial samples generation, and the development of countermeasures.
+* people interested in research who want to create methods and tools that have a real impact (not yet another paper).
+* proactive people with initiative.
+* people with proficiency in programming languages commonly used in AI research, at least one out of Python, R or Julia.
+* solid understanding of AI/ML models, including their architecture and training processes.
+* good communication skills and proficiency in academic writing; excellent problem-solving skills.
 
-**Requirements:**
+We offer:
 
-Proficiency in programming languages commonly used in AI research, such as Python, R or Julia.
-Familiarity with deep learning frameworks and libraries, such as PyTorch.
-Solid understanding of foundation models, including their architecture and training processes.
-Excellent problem-solving skills and ability to work independently and as part of a team.
-Good communication skills and proficiency in academic writing.
+- the opportunity to learn from experienced and motivated researchers at different career levels as well as the opportunity to work with a rich network of international and interdisciplinary collaborators (for example Harvard Medical University, Vienna University, LMU, NTU, KDD Lab, Bari University and many others).
+- a possibility for a tenure-track position after the completion of the postdoc. 
+- possibility to present research results at international conferences.
+- access to our supercomputers, among them Eden with 32x A100 cards and 1.5TB of storage, a perfect place for training large DNN models.
 
 
-Interested candidates are requested to submit the following documents to **przemyslaw.biecek at pw.edu.pl**.
+**Salary**
 
+Salary depend on experience and type of project (applied projects tend to have higher salaries than typical research projects). 
 
-We look forward to receiving your applications and welcoming an enthusiastic and dedicated PhD student to contribute to our foundation model redteaming research endeavors.
+Average rates range from 50-60 PLN/hour (equivalent to 8-9.5k FTE) for a student with experience, through to 80-85 PLN/hour (equivalent to 12-13k FTE) for an experienced PhD student. The base salary for a post-doc (no obligatory teaching duties) is 120-140k gross per year (this is a base salary, extras depend on activity).
 
+### Low commitment projects (1/5 FTE) in 2023 {-}
 
-Apply before **June 30 the end of the day** (**June 10 in case of interest in MIM UW**).
+Ex 1: Any object can be embedded with an appropriately trained neural network. But can objects be embedded using tree-based models? This project (an ideal candidate for MSc thesis) is about exploring different embedding construction techniques based on random forest or boosting methods.
 
+Ex 2:  Analysis of champion-challenger model pairs on subsets of observations obtained using data segmentation. The project consists of an extension of the EPP method (Elo for predictive models) based on segmented data (e.g. by k-means or by random splits). In addition to the performance analysis, a PD profile comparison analysis is also performed on the splits.
 
+### Mid commitment projects (2-2.5/5 FTE) in 2023 {-}
 
-### Deep Learning Engineer {-}
+Ex 1: Development of techniques for explaining segmentation models for 1d (feature-producing models, e.g. survival models), 2d (image-producing models) or 3d data. (CT, hyperspectral images) Experience and knowledge of classical explanatory techniques (SHAP, LRP, CRP) are desirable.
 
-**Required:**
+Ex 2:  Development of red-teaming techniques for LLaMA 2-type language models. Development of test suites testing the vulnerability of the model to a set of prompting-based attacks.
 
-* Background in Computer Science, Mathematics, Statistics or similar. 
-* Experience in Deep Learning for 2d/3d image data (*torch is a plus)
-* Interest in medical applications
 
-**Scope of work:**
+### Full-time positions for post-docs in 2023 {-}
 
-* Training of machine learning models for tabular and image data
-* Responsible ML solutions for the healthcare domain
 
-**Offer:**
+In addition to the research tasks listed above, responsibilities include mentoring junior staff (students, PhD students) and leading small research groups.
+
+**Additional information about costs of living in Warsaw**
+
+[From FENS](https://fens.org.pl/en/job-offers/):
+Warsaw is the capital of Poland, so in principle it is one of the most expensive cities in the country but living in Warsaw is still much cheaper than in the majority of European cities. In general, a single person in Warsaw is able to cover all necessary expenses (flat, media, transport, food) with ca. 5000-5500 PLN per month. A typical cost of renting a 2-room flat (30-40 square meters) varies between 2000-3000 PLN, depending on the part of the city. Electricity and media (TV, Internet) are usually not included in the rent and have to be paid separately. The cost of the lunch in the vicinity of the main campus of WUT is ca. 25-40 PLN, while a monthly ticket for public transport is 110 or 180 PLN depending on the area of validity. Public health insurance is included in the post (this includes also admission to hospitals), however the medicines are paid separately.
 
-* Excellent atmosphere for work in a young and very active lab
-* Conferences and training budget
-* Short visits in cooperating abroad group
-* Access to CPU / GPU clusters
-* Elastic working hours that can be combined with studies
 
 ### Research Software Engineer {-}
 
+In addition to research positions, we also offer positions for skilled programmers.
+
 **Required:**
 
 * Background in Computer Science, Mathematics, Statistics or similar. 
-* Experience in Scientific Programming (R and/or Python and/or C++)
-* Interest in applications, machine learning and explainable artificial intelligence.
+* Experience in Computer Vision or NLP models
 
 **Scope of work:**
 
 * Training of machine learning models for tabular and image data
 * Interpretable solutions for tabular and image data
-* Interactive interfaces 
 * Responsible ML solutions for the healthcare domain
 
-**Offer:**
-
-* Excellent atmosphere for work in a young and very active lab
-* Conferences and training budget
-* Short visits in cooperating abroad group
-* Access to CPU / GPU clusters
-* Elastic working hours that can be combined with studies
 
-### Post-doc {-}
+<!--
   
 I am looking for a post-doc to join the MI2.AI team for one or two years within [HOMER](https://www.mi2.ai/research-grants.html#homer-2020-2025) project.
 The ideal fit is someone with (experience in Medical Image Analysis with deep learning models) OR (someone with experience in AutoML for tabular data).

From a4cf9b566e227e4c388eeb738e8030dce29a5956 Mon Sep 17 00:00:00 2001
From: runner <runner@Mac-1697150430135.local>
Date: Thu, 12 Oct 2023 22:43:16 +0000
Subject: [PATCH 2/7] Render book

---
 docs/404.html                   |   7 +-
 docs/beta-bit.html              |   7 +-
 docs/books.html                 |   7 +-
 docs/contact.html               |   7 +-
 docs/index.html                 |  13 +--
 docs/mi2redteam.html            |   7 +-
 "docs/mi\302\262education.html" |   7 +-
 "docs/mi\302\262research.html"  |   7 +-
 "docs/mi\302\262solutions.html" |   7 +-
 docs/open-positions.html        | 153 +++++++++++++++-----------------
 docs/papers.html                |   7 +-
 docs/research-grants.html       |   7 +-
 docs/responsibleml-blog.html    |   7 +-
 docs/search_index.json          |   2 +-
 docs/seminars.html              |   7 +-
 docs/software.html              |   7 +-
 docs/teaching.html              |   7 +-
 docs/the-team.html              |   7 +-
 18 files changed, 142 insertions(+), 131 deletions(-)

diff --git a/docs/404.html b/docs/404.html
index 74bad58..60051b9 100644
--- a/docs/404.html
+++ b/docs/404.html
@@ -92,10 +92,11 @@
 <li class="chapter" data-level="" data-path="the-team.html"><a href="the-team.html"><i class="fa fa-check"></i>The Team</a></li>
 <li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html"><i class="fa fa-check"></i>Open Positions</a>
 <ul>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#phd-positions-in-2023"><i class="fa fa-check"></i>PhD positions in 2023</a></li>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#deep-learning-engineer"><i class="fa fa-check"></i>Deep Learning Engineer</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#mi2-policy-on-benefits-and-obligations"><i class="fa fa-check"></i>MI2 policy on benefits and obligations</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#low-commitment-projects-15-fte-in-2023"><i class="fa fa-check"></i>Low commitment projects (1/5 FTE) in 2023</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#mid-commitment-projects-2-2.55-fte-in-2023"><i class="fa fa-check"></i>Mid commitment projects (2-2.5/5 FTE) in 2023</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#full-time-positions-for-post-docs-in-2023"><i class="fa fa-check"></i>Full-time positions for post-docs in 2023</a></li>
 <li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#research-software-engineer"><i class="fa fa-check"></i>Research Software Engineer</a></li>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#post-doc"><i class="fa fa-check"></i>Post-doc</a></li>
 </ul></li>
 <li class="chapter" data-level="" data-path="contact.html"><a href="contact.html"><i class="fa fa-check"></i>Contact</a></li>
 </ul></li>
diff --git a/docs/beta-bit.html b/docs/beta-bit.html
index 79dcc81..2d7ac32 100644
--- a/docs/beta-bit.html
+++ b/docs/beta-bit.html
@@ -92,10 +92,11 @@
 <li class="chapter" data-level="" data-path="the-team.html"><a href="the-team.html"><i class="fa fa-check"></i>The Team</a></li>
 <li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html"><i class="fa fa-check"></i>Open Positions</a>
 <ul>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#phd-positions-in-2023"><i class="fa fa-check"></i>PhD positions in 2023</a></li>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#deep-learning-engineer"><i class="fa fa-check"></i>Deep Learning Engineer</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#mi2-policy-on-benefits-and-obligations"><i class="fa fa-check"></i>MI2 policy on benefits and obligations</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#low-commitment-projects-15-fte-in-2023"><i class="fa fa-check"></i>Low commitment projects (1/5 FTE) in 2023</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#mid-commitment-projects-2-2.55-fte-in-2023"><i class="fa fa-check"></i>Mid commitment projects (2-2.5/5 FTE) in 2023</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#full-time-positions-for-post-docs-in-2023"><i class="fa fa-check"></i>Full-time positions for post-docs in 2023</a></li>
 <li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#research-software-engineer"><i class="fa fa-check"></i>Research Software Engineer</a></li>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#post-doc"><i class="fa fa-check"></i>Post-doc</a></li>
 </ul></li>
 <li class="chapter" data-level="" data-path="contact.html"><a href="contact.html"><i class="fa fa-check"></i>Contact</a></li>
 </ul></li>
diff --git a/docs/books.html b/docs/books.html
index 9ca88cd..f65a7b3 100644
--- a/docs/books.html
+++ b/docs/books.html
@@ -92,10 +92,11 @@
 <li class="chapter" data-level="" data-path="the-team.html"><a href="the-team.html"><i class="fa fa-check"></i>The Team</a></li>
 <li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html"><i class="fa fa-check"></i>Open Positions</a>
 <ul>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#phd-positions-in-2023"><i class="fa fa-check"></i>PhD positions in 2023</a></li>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#deep-learning-engineer"><i class="fa fa-check"></i>Deep Learning Engineer</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#mi2-policy-on-benefits-and-obligations"><i class="fa fa-check"></i>MI2 policy on benefits and obligations</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#low-commitment-projects-15-fte-in-2023"><i class="fa fa-check"></i>Low commitment projects (1/5 FTE) in 2023</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#mid-commitment-projects-2-2.55-fte-in-2023"><i class="fa fa-check"></i>Mid commitment projects (2-2.5/5 FTE) in 2023</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#full-time-positions-for-post-docs-in-2023"><i class="fa fa-check"></i>Full-time positions for post-docs in 2023</a></li>
 <li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#research-software-engineer"><i class="fa fa-check"></i>Research Software Engineer</a></li>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#post-doc"><i class="fa fa-check"></i>Post-doc</a></li>
 </ul></li>
 <li class="chapter" data-level="" data-path="contact.html"><a href="contact.html"><i class="fa fa-check"></i>Contact</a></li>
 </ul></li>
diff --git a/docs/contact.html b/docs/contact.html
index 441b1f0..a03f7b8 100644
--- a/docs/contact.html
+++ b/docs/contact.html
@@ -92,10 +92,11 @@
 <li class="chapter" data-level="" data-path="the-team.html"><a href="the-team.html"><i class="fa fa-check"></i>The Team</a></li>
 <li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html"><i class="fa fa-check"></i>Open Positions</a>
 <ul>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#phd-positions-in-2023"><i class="fa fa-check"></i>PhD positions in 2023</a></li>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#deep-learning-engineer"><i class="fa fa-check"></i>Deep Learning Engineer</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#mi2-policy-on-benefits-and-obligations"><i class="fa fa-check"></i>MI2 policy on benefits and obligations</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#low-commitment-projects-15-fte-in-2023"><i class="fa fa-check"></i>Low commitment projects (1/5 FTE) in 2023</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#mid-commitment-projects-2-2.55-fte-in-2023"><i class="fa fa-check"></i>Mid commitment projects (2-2.5/5 FTE) in 2023</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#full-time-positions-for-post-docs-in-2023"><i class="fa fa-check"></i>Full-time positions for post-docs in 2023</a></li>
 <li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#research-software-engineer"><i class="fa fa-check"></i>Research Software Engineer</a></li>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#post-doc"><i class="fa fa-check"></i>Post-doc</a></li>
 </ul></li>
 <li class="chapter" data-level="" data-path="contact.html"><a href="contact.html"><i class="fa fa-check"></i>Contact</a></li>
 </ul></li>
diff --git a/docs/index.html b/docs/index.html
index 68e3d2a..9acec7c 100644
--- a/docs/index.html
+++ b/docs/index.html
@@ -92,10 +92,11 @@
 <li class="chapter" data-level="" data-path="the-team.html"><a href="the-team.html"><i class="fa fa-check"></i>The Team</a></li>
 <li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html"><i class="fa fa-check"></i>Open Positions</a>
 <ul>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#phd-positions-in-2023"><i class="fa fa-check"></i>PhD positions in 2023</a></li>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#deep-learning-engineer"><i class="fa fa-check"></i>Deep Learning Engineer</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#mi2-policy-on-benefits-and-obligations"><i class="fa fa-check"></i>MI2 policy on benefits and obligations</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#low-commitment-projects-15-fte-in-2023"><i class="fa fa-check"></i>Low commitment projects (1/5 FTE) in 2023</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#mid-commitment-projects-2-2.55-fte-in-2023"><i class="fa fa-check"></i>Mid commitment projects (2-2.5/5 FTE) in 2023</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#full-time-positions-for-post-docs-in-2023"><i class="fa fa-check"></i>Full-time positions for post-docs in 2023</a></li>
 <li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#research-software-engineer"><i class="fa fa-check"></i>Research Software Engineer</a></li>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#post-doc"><i class="fa fa-check"></i>Post-doc</a></li>
 </ul></li>
 <li class="chapter" data-level="" data-path="contact.html"><a href="contact.html"><i class="fa fa-check"></i>Contact</a></li>
 </ul></li>
@@ -173,15 +174,15 @@ <h1>MI².AI<a href="index.html#mi².ai" class="anchor-section" aria-label="Ancho
 <script>
 document.body.classList.add("index-page")
 </script>
-<p>MI².AI is a group of mathematicians and computer scientists that love to play with predictive models.
-We are spread between <a href="https://www.pw.edu.pl/engpw">Warsaw University of Technology</a> and <a href="https://en.uw.edu.pl">University of Warsaw</a>. Here we have workshops, seminars, here we are forging new ideas, creating tools, solving problems, doing consulting and sharing our positive attitude. Feel free to jump in.</p>
+<p>MI².AI is a group of mathematicians and computer scientists who love to play with predictive models.
+We are spread between <a href="https://www.pw.edu.pl/engpw">Warsaw University of Technology</a> and <a href="https://en.uw.edu.pl">University of Warsaw</a>. Here we have workshops and seminars, here we are forging new ideas, creating tools, solving problems, doing consulting and sharing our positive attitude. Feel free to jump in.</p>
 <div id="mission" class="section level4 unnumbered hasAnchor">
 <h4>Mission<a href="index.html#mission" class="anchor-section" aria-label="Anchor link to header"></a></h4>
 <p>Machine learning is like atomic energy. <strong>We develop leaders, skills, methods, tools and good practices so that predictive models can be deployed responsibly and sustainably</strong>.</p>
 </div>
 <div id="vision" class="section level4 unnumbered hasAnchor">
 <h4>Vision<a href="index.html#vision" class="anchor-section" aria-label="Anchor link to header"></a></h4>
-<p>MI² is a group of experts supporting global initiatives aimed at responsible and sustainable machine learning. We support the development of future leaders of responsible machine learning through internships, PhDs, postdoctoral fellowships and so on. We seek for research grants and business projects to conduct both scientific and applied research. We develop and maintain software and infrastructure necessary to build responsible and sustainable ML. We develop cooperation with international teams working on similar topics. We support companies to implement best practices related to responsible modelling in their operation. We conduct workshops and training on responsible predictive modelling.</p>
+<p>MI² is a group of experts supporting global initiatives aimed at responsible and sustainable machine learning. We support the development of future leaders of responsible machine learning through internships, PhDs, postdoctoral fellowships and so on. We seek for research grants and business projects to conduct both scientific and applied research. We develop and maintain the software and infrastructure necessary to build responsible and sustainable ML. We develop cooperation with international teams working on similar topics. We support companies to implement best practices related to responsible modelling in their operation. We conduct workshops and training on responsible predictive modelling.</p>
 </div>
 </div>
 <a href ="/" class="mobile-logo"><svg width="30" height="30" viewBox="0 0 30 30" fill="none" xmlns="http://www.w3.org/2000/svg"><path d="M28.6799 1.36802V28.68H1.31982V1.32004H28.6799V1.36802ZM30 0H0V30H30V0Z" fill="black"/><path d="M21.6357 21.0469V9.04688H22.9556V21.0469H21.6357Z" fill="black"/><path d="M18.5639 9.04688V21.0469H17.2439V11.0629L13.4039 21.0469H12.2039L8.36395 11.0629V21.0469H7.04395V9.04688H8.95195L12.8039 19.0789L16.6559 9.04688H18.5639Z" fill="black"/></svg></a>
diff --git a/docs/mi2redteam.html b/docs/mi2redteam.html
index 1eb03f9..dbc5823 100644
--- a/docs/mi2redteam.html
+++ b/docs/mi2redteam.html
@@ -92,10 +92,11 @@
 <li class="chapter" data-level="" data-path="the-team.html"><a href="the-team.html"><i class="fa fa-check"></i>The Team</a></li>
 <li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html"><i class="fa fa-check"></i>Open Positions</a>
 <ul>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#phd-positions-in-2023"><i class="fa fa-check"></i>PhD positions in 2023</a></li>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#deep-learning-engineer"><i class="fa fa-check"></i>Deep Learning Engineer</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#mi2-policy-on-benefits-and-obligations"><i class="fa fa-check"></i>MI2 policy on benefits and obligations</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#low-commitment-projects-15-fte-in-2023"><i class="fa fa-check"></i>Low commitment projects (1/5 FTE) in 2023</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#mid-commitment-projects-2-2.55-fte-in-2023"><i class="fa fa-check"></i>Mid commitment projects (2-2.5/5 FTE) in 2023</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#full-time-positions-for-post-docs-in-2023"><i class="fa fa-check"></i>Full-time positions for post-docs in 2023</a></li>
 <li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#research-software-engineer"><i class="fa fa-check"></i>Research Software Engineer</a></li>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#post-doc"><i class="fa fa-check"></i>Post-doc</a></li>
 </ul></li>
 <li class="chapter" data-level="" data-path="contact.html"><a href="contact.html"><i class="fa fa-check"></i>Contact</a></li>
 </ul></li>
diff --git "a/docs/mi\302\262education.html" "b/docs/mi\302\262education.html"
index cbc1632..7715016 100644
--- "a/docs/mi\302\262education.html"
+++ "b/docs/mi\302\262education.html"
@@ -92,10 +92,11 @@
 <li class="chapter" data-level="" data-path="the-team.html"><a href="the-team.html"><i class="fa fa-check"></i>The Team</a></li>
 <li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html"><i class="fa fa-check"></i>Open Positions</a>
 <ul>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#phd-positions-in-2023"><i class="fa fa-check"></i>PhD positions in 2023</a></li>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#deep-learning-engineer"><i class="fa fa-check"></i>Deep Learning Engineer</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#mi2-policy-on-benefits-and-obligations"><i class="fa fa-check"></i>MI2 policy on benefits and obligations</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#low-commitment-projects-15-fte-in-2023"><i class="fa fa-check"></i>Low commitment projects (1/5 FTE) in 2023</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#mid-commitment-projects-2-2.55-fte-in-2023"><i class="fa fa-check"></i>Mid commitment projects (2-2.5/5 FTE) in 2023</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#full-time-positions-for-post-docs-in-2023"><i class="fa fa-check"></i>Full-time positions for post-docs in 2023</a></li>
 <li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#research-software-engineer"><i class="fa fa-check"></i>Research Software Engineer</a></li>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#post-doc"><i class="fa fa-check"></i>Post-doc</a></li>
 </ul></li>
 <li class="chapter" data-level="" data-path="contact.html"><a href="contact.html"><i class="fa fa-check"></i>Contact</a></li>
 </ul></li>
diff --git "a/docs/mi\302\262research.html" "b/docs/mi\302\262research.html"
index 3492ced..15a9ffd 100644
--- "a/docs/mi\302\262research.html"
+++ "b/docs/mi\302\262research.html"
@@ -92,10 +92,11 @@
 <li class="chapter" data-level="" data-path="the-team.html"><a href="the-team.html"><i class="fa fa-check"></i>The Team</a></li>
 <li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html"><i class="fa fa-check"></i>Open Positions</a>
 <ul>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#phd-positions-in-2023"><i class="fa fa-check"></i>PhD positions in 2023</a></li>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#deep-learning-engineer"><i class="fa fa-check"></i>Deep Learning Engineer</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#mi2-policy-on-benefits-and-obligations"><i class="fa fa-check"></i>MI2 policy on benefits and obligations</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#low-commitment-projects-15-fte-in-2023"><i class="fa fa-check"></i>Low commitment projects (1/5 FTE) in 2023</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#mid-commitment-projects-2-2.55-fte-in-2023"><i class="fa fa-check"></i>Mid commitment projects (2-2.5/5 FTE) in 2023</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#full-time-positions-for-post-docs-in-2023"><i class="fa fa-check"></i>Full-time positions for post-docs in 2023</a></li>
 <li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#research-software-engineer"><i class="fa fa-check"></i>Research Software Engineer</a></li>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#post-doc"><i class="fa fa-check"></i>Post-doc</a></li>
 </ul></li>
 <li class="chapter" data-level="" data-path="contact.html"><a href="contact.html"><i class="fa fa-check"></i>Contact</a></li>
 </ul></li>
diff --git "a/docs/mi\302\262solutions.html" "b/docs/mi\302\262solutions.html"
index 7a7149d..80d96a7 100644
--- "a/docs/mi\302\262solutions.html"
+++ "b/docs/mi\302\262solutions.html"
@@ -92,10 +92,11 @@
 <li class="chapter" data-level="" data-path="the-team.html"><a href="the-team.html"><i class="fa fa-check"></i>The Team</a></li>
 <li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html"><i class="fa fa-check"></i>Open Positions</a>
 <ul>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#phd-positions-in-2023"><i class="fa fa-check"></i>PhD positions in 2023</a></li>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#deep-learning-engineer"><i class="fa fa-check"></i>Deep Learning Engineer</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#mi2-policy-on-benefits-and-obligations"><i class="fa fa-check"></i>MI2 policy on benefits and obligations</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#low-commitment-projects-15-fte-in-2023"><i class="fa fa-check"></i>Low commitment projects (1/5 FTE) in 2023</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#mid-commitment-projects-2-2.55-fte-in-2023"><i class="fa fa-check"></i>Mid commitment projects (2-2.5/5 FTE) in 2023</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#full-time-positions-for-post-docs-in-2023"><i class="fa fa-check"></i>Full-time positions for post-docs in 2023</a></li>
 <li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#research-software-engineer"><i class="fa fa-check"></i>Research Software Engineer</a></li>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#post-doc"><i class="fa fa-check"></i>Post-doc</a></li>
 </ul></li>
 <li class="chapter" data-level="" data-path="contact.html"><a href="contact.html"><i class="fa fa-check"></i>Contact</a></li>
 </ul></li>
diff --git a/docs/open-positions.html b/docs/open-positions.html
index 580ce2a..b9f1240 100644
--- a/docs/open-positions.html
+++ b/docs/open-positions.html
@@ -92,10 +92,11 @@
 <li class="chapter" data-level="" data-path="the-team.html"><a href="the-team.html"><i class="fa fa-check"></i>The Team</a></li>
 <li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html"><i class="fa fa-check"></i>Open Positions</a>
 <ul>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#phd-positions-in-2023"><i class="fa fa-check"></i>PhD positions in 2023</a></li>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#deep-learning-engineer"><i class="fa fa-check"></i>Deep Learning Engineer</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#mi2-policy-on-benefits-and-obligations"><i class="fa fa-check"></i>MI2 policy on benefits and obligations</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#low-commitment-projects-15-fte-in-2023"><i class="fa fa-check"></i>Low commitment projects (1/5 FTE) in 2023</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#mid-commitment-projects-2-2.55-fte-in-2023"><i class="fa fa-check"></i>Mid commitment projects (2-2.5/5 FTE) in 2023</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#full-time-positions-for-post-docs-in-2023"><i class="fa fa-check"></i>Full-time positions for post-docs in 2023</a></li>
 <li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#research-software-engineer"><i class="fa fa-check"></i>Research Software Engineer</a></li>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#post-doc"><i class="fa fa-check"></i>Post-doc</a></li>
 </ul></li>
 <li class="chapter" data-level="" data-path="contact.html"><a href="contact.html"><i class="fa fa-check"></i>Contact</a></li>
 </ul></li>
@@ -167,100 +168,94 @@ <h1>
             <section class="normal" id="section-">
 <div id="open-positions" class="section level2 unnumbered hasAnchor">
 <h2>Open Positions<a href="open-positions.html#open-positions" class="anchor-section" aria-label="Anchor link to header"></a></h2>
-<p>We have open call for <a href="https://www.mi2.ai/research-grants.html#homer-2020-2025">HOMER</a> and <a href="https://www.mi2.ai/research-grants.html#x-lungs-2021-2024">xLungs</a> projects for following positions. If you are interested in any of them please send your CV and Motivation Letter to przemyslaw.biecek at pw.edu.pl. We reserve the right to contact only selected candidates.</p>
-<div id="phd-positions-in-2023" class="section level3 unnumbered hasAnchor">
-<h3>PhD positions in 2023<a href="open-positions.html#phd-positions-in-2023" class="anchor-section" aria-label="Anchor link to header"></a></h3>
-<p><img src="images/redteaming_mi2.png" style="width: 100%;"></p>
-<p><strong>Call for PhD Student: Research Opportunity in Foundation Model Redteaming</strong></p>
-<p>We are seeking a highly motivated and talented PhD student to join MI2.AI research team in the exciting field of foundation model redteaming. As the advancements development of AI models (especially transformer based with attention mechanism), it becomes crucial to ensure the robustness and security of the foundation models that underlie these advancements.</p>
-<p><strong>Project Description:</strong></p>
-<p>The selected PhD student will work on investigating potential vulnerabilities and biases in foundation models, such as language models (one PhD position) and 2D/3D computer vision models (one PhD position). Redteaming involves conducting adversarial assessments to identify weaknesses, uncover potential attack vectors, and develop defensive strategies for foundation models/transformer models. The research will explore various aspects, including data poisoning attacks, adversarial samples generation, and the development of countermeasures.</p>
-<p><strong>Requirements:</strong></p>
-<p>Proficiency in programming languages commonly used in AI research, such as Python, R or Julia.
-Familiarity with deep learning frameworks and libraries, such as PyTorch.
-Solid understanding of foundation models, including their architecture and training processes.
-Excellent problem-solving skills and ability to work independently and as part of a team.
-Good communication skills and proficiency in academic writing.</p>
-<p>Interested candidates are requested to submit the following documents to <strong>przemyslaw.biecek at pw.edu.pl</strong>.</p>
-<p>We look forward to receiving your applications and welcoming an enthusiastic and dedicated PhD student to contribute to our foundation model redteaming research endeavors.</p>
-<p>Apply before <strong>June 30 the end of the day</strong> (<strong>June 10 in case of interest in MIM UW</strong>).</p>
-</div>
-<div id="deep-learning-engineer" class="section level3 unnumbered hasAnchor">
-<h3>Deep Learning Engineer<a href="open-positions.html#deep-learning-engineer" class="anchor-section" aria-label="Anchor link to header"></a></h3>
-<p><strong>Required:</strong></p>
-<ul>
-<li>Background in Computer Science, Mathematics, Statistics or similar.</li>
-<li>Experience in Deep Learning for 2d/3d image data (*torch is a plus)</li>
-<li>Interest in medical applications</li>
-</ul>
-<p><strong>Scope of work:</strong></p>
+<p><img src="images/positions_mi2.png" style="width: 100%;"></p>
+<p>We have an open call for <a href="https://www.mi2.ai/research-grants.html#homer-2020-2025">HOMER</a> and <a href="https://www.mi2.ai/research-grants.html#x-lungs-2021-2024">xLungs</a> projects for following positions. If you are interested in any of them please send your CV and Motivation Letter to przemyslaw.biecek at pw.edu.pl. We reserve the right to contact only selected candidates.</p>
+<div id="mi2-policy-on-benefits-and-obligations" class="section level3 unnumbered hasAnchor">
+<h3>MI2 policy on benefits and obligations<a href="open-positions.html#mi2-policy-on-benefits-and-obligations" class="anchor-section" aria-label="Anchor link to header"></a></h3>
+<p>We are seeking highly motivated and talented researchers at different career levels to join MI2.AI research team in the exciting field of explainability and red-teaming of AI models.</p>
+<p>There is a common recruitment policy for all positions.</p>
+<p>We are looking for</p>
 <ul>
-<li>Training of machine learning models for tabular and image data</li>
-<li>Responsible ML solutions for the healthcare domain</li>
+<li>people interested in research who want to create methods and tools that have a real impact (not yet another paper).</li>
+<li>proactive people with initiative.</li>
+<li>people with proficiency in programming languages commonly used in AI research, at least one out of Python, R or Julia.</li>
+<li>solid understanding of AI/ML models, including their architecture and training processes.</li>
+<li>good communication skills and proficiency in academic writing; excellent problem-solving skills.</li>
 </ul>
-<p><strong>Offer:</strong></p>
+<p>We offer:</p>
 <ul>
-<li>Excellent atmosphere for work in a young and very active lab</li>
-<li>Conferences and training budget</li>
-<li>Short visits in cooperating abroad group</li>
-<li>Access to CPU / GPU clusters</li>
-<li>Elastic working hours that can be combined with studies</li>
+<li>the opportunity to learn from experienced and motivated researchers at different career levels as well as the opportunity to work with a rich network of international and interdisciplinary collaborators (for example Harvard Medical University, Vienna University, LMU, NTU, KDD Lab, Bari University and many others).</li>
+<li>a possibility for a tenure-track position after the completion of the postdoc.</li>
+<li>possibility to present research results at international conferences.</li>
+<li>access to our supercomputers, among them Eden with 32x A100 cards and 1.5TB of storage, a perfect place for training large DNN models.</li>
 </ul>
+<p><strong>Salary</strong></p>
+<p>Salary depend on experience and type of project (applied projects tend to have higher salaries than typical research projects).</p>
+<p>Average rates range from 50-60 PLN/hour (equivalent to 8-9.5k FTE) for a student with experience, through to 80-85 PLN/hour (equivalent to 12-13k FTE) for an experienced PhD student. The base salary for a post-doc (no obligatory teaching duties) is 120-140k gross per year (this is a base salary, extras depend on activity).</p>
+</div>
+<div id="low-commitment-projects-15-fte-in-2023" class="section level3 unnumbered hasAnchor">
+<h3>Low commitment projects (1/5 FTE) in 2023<a href="open-positions.html#low-commitment-projects-15-fte-in-2023" class="anchor-section" aria-label="Anchor link to header"></a></h3>
+<p>Ex 1: Any object can be embedded with an appropriately trained neural network. But can objects be embedded using tree-based models? This project (an ideal candidate for MSc thesis) is about exploring different embedding construction techniques based on random forest or boosting methods.</p>
+<p>Ex 2: Analysis of champion-challenger model pairs on subsets of observations obtained using data segmentation. The project consists of an extension of the EPP method (Elo for predictive models) based on segmented data (e.g. by k-means or by random splits). In addition to the performance analysis, a PD profile comparison analysis is also performed on the splits.</p>
+</div>
+<div id="mid-commitment-projects-2-2.55-fte-in-2023" class="section level3 unnumbered hasAnchor">
+<h3>Mid commitment projects (2-2.5/5 FTE) in 2023<a href="open-positions.html#mid-commitment-projects-2-2.55-fte-in-2023" class="anchor-section" aria-label="Anchor link to header"></a></h3>
+<p>Ex 1: Development of techniques for explaining segmentation models for 1d (feature-producing models, e.g. survival models), 2d (image-producing models) or 3d data. (CT, hyperspectral images) Experience and knowledge of classical explanatory techniques (SHAP, LRP, CRP) are desirable.</p>
+<p>Ex 2: Development of red-teaming techniques for LLaMA 2-type language models. Development of test suites testing the vulnerability of the model to a set of prompting-based attacks.</p>
+</div>
+<div id="full-time-positions-for-post-docs-in-2023" class="section level3 unnumbered hasAnchor">
+<h3>Full-time positions for post-docs in 2023<a href="open-positions.html#full-time-positions-for-post-docs-in-2023" class="anchor-section" aria-label="Anchor link to header"></a></h3>
+<p>In addition to the research tasks listed above, responsibilities include mentoring junior staff (students, PhD students) and leading small research groups.</p>
+<p><strong>Additional information about costs of living in Warsaw</strong></p>
+<p><a href="https://fens.org.pl/en/job-offers/">From FENS</a>:
+Warsaw is the capital of Poland, so in principle it is one of the most expensive cities in the country but living in Warsaw is still much cheaper than in the majority of European cities. In general, a single person in Warsaw is able to cover all necessary expenses (flat, media, transport, food) with ca. 5000-5500 PLN per month. A typical cost of renting a 2-room flat (30-40 square meters) varies between 2000-3000 PLN, depending on the part of the city. Electricity and media (TV, Internet) are usually not included in the rent and have to be paid separately. The cost of the lunch in the vicinity of the main campus of WUT is ca. 25-40 PLN, while a monthly ticket for public transport is 110 or 180 PLN depending on the area of validity. Public health insurance is included in the post (this includes also admission to hospitals), however the medicines are paid separately.</p>
 </div>
 <div id="research-software-engineer" class="section level3 unnumbered hasAnchor">
 <h3>Research Software Engineer<a href="open-positions.html#research-software-engineer" class="anchor-section" aria-label="Anchor link to header"></a></h3>
+<p>In addition to research positions, we also offer positions for skilled programmers.</p>
 <p><strong>Required:</strong></p>
 <ul>
 <li>Background in Computer Science, Mathematics, Statistics or similar.</li>
-<li>Experience in Scientific Programming (R and/or Python and/or C++)</li>
-<li>Interest in applications, machine learning and explainable artificial intelligence.</li>
+<li>Experience in Computer Vision or NLP models</li>
 </ul>
 <p><strong>Scope of work:</strong></p>
 <ul>
 <li>Training of machine learning models for tabular and image data</li>
 <li>Interpretable solutions for tabular and image data</li>
-<li>Interactive interfaces</li>
 <li>Responsible ML solutions for the healthcare domain</li>
 </ul>
-<p><strong>Offer:</strong></p>
-<ul>
-<li>Excellent atmosphere for work in a young and very active lab</li>
-<li>Conferences and training budget</li>
-<li>Short visits in cooperating abroad group</li>
-<li>Access to CPU / GPU clusters</li>
-<li>Elastic working hours that can be combined with studies</li>
-</ul>
-</div>
-<div id="post-doc" class="section level3 unnumbered hasAnchor">
-<h3>Post-doc<a href="open-positions.html#post-doc" class="anchor-section" aria-label="Anchor link to header"></a></h3>
-<p>I am looking for a post-doc to join the MI2.AI team for one or two years within <a href="https://www.mi2.ai/research-grants.html#homer-2020-2025">HOMER</a> project.
+<!--
+  
+I am looking for a post-doc to join the MI2.AI team for one or two years within [HOMER](https://www.mi2.ai/research-grants.html#homer-2020-2025) project.
 The ideal fit is someone with (experience in Medical Image Analysis with deep learning models) OR (someone with experience in AutoML for tabular data).
-Experience or interest in <strong>XAI</strong> / <strong>fairness</strong> will always be a huge plus in our team.</p>
-<p>We have a young and very energetic team focused on growth in AI. Full time position. Salary above academic average. Little or no didactics. Strong focus on doing things that are <strong>meaningful</strong>.</p>
-<p><strong>Required:</strong></p>
-<ul>
-<li>PhD in Computer Science, Mathematics, Statistics or similar. (Different background? We encourage cross-domain short research visits).</li>
-<li>Experience in Scientific Programming (R and/or Python and/or C++)</li>
-<li>Good track of Scientific Records</li>
-</ul>
-<p><strong>Scope of work:</strong></p>
-<ul>
-<li>Automated model exploration</li>
-<li>Interpretable measures for model performance</li>
-<li>Meta/transfer learning in automated model development</li>
-<li>Automated model validation</li>
-<li>Experiments with XAI for deep learning</li>
-</ul>
-<p><strong>Offer:</strong></p>
-<ul>
-<li>Excellent atmosphere for work in a young and very active ML lab</li>
-<li>Conferences and training budget</li>
-<li>Short visits in cooperating abroad group</li>
-<li>Access to CPU / GPU clusters</li>
-<li>Full-time job at Warsaw University of Technology</li>
-<li>plus results-driven extras</li>
-<li>contract for 6 months (short visit) / 12 months or 24 months (long visit)</li>
-</ul>
+Experience or interest in **XAI** / **fairness** will always be a huge plus in our team.
+
+We have a young and very energetic team focused on growth in AI. Full time position. Salary above academic average. Little or no didactics. Strong focus on doing things that are **meaningful**.
+
+**Required:**
+
+* PhD in Computer Science, Mathematics, Statistics or similar. (Different background? We encourage cross-domain short research visits).
+* Experience in Scientific Programming (R and/or Python and/or C++)
+* Good track of Scientific Records
+
+**Scope of work:**
+
+* Automated model exploration
+* Interpretable measures for model performance
+* Meta/transfer learning in automated model development
+* Automated model validation
+* Experiments with XAI for deep learning
+
+**Offer:**
+
+* Excellent atmosphere for work in a young and very active ML lab
+* Conferences and training budget
+* Short visits in cooperating abroad group
+* Access to CPU / GPU clusters
+* Full-time job at Warsaw University of Technology
+* plus results-driven extras
+* contract for 6 months (short visit) / 12 months or 24 months (long visit)
+
 <!-- ### ARES {-}
 
 PhD position funded by the PRELUDIUM BIS grant ARES (*Attack-Resistant Explanations toward Secure and trustworthy AI*).
diff --git a/docs/papers.html b/docs/papers.html
index a85f5cf..b1da6c2 100644
--- a/docs/papers.html
+++ b/docs/papers.html
@@ -92,10 +92,11 @@
 <li class="chapter" data-level="" data-path="the-team.html"><a href="the-team.html"><i class="fa fa-check"></i>The Team</a></li>
 <li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html"><i class="fa fa-check"></i>Open Positions</a>
 <ul>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#phd-positions-in-2023"><i class="fa fa-check"></i>PhD positions in 2023</a></li>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#deep-learning-engineer"><i class="fa fa-check"></i>Deep Learning Engineer</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#mi2-policy-on-benefits-and-obligations"><i class="fa fa-check"></i>MI2 policy on benefits and obligations</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#low-commitment-projects-15-fte-in-2023"><i class="fa fa-check"></i>Low commitment projects (1/5 FTE) in 2023</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#mid-commitment-projects-2-2.55-fte-in-2023"><i class="fa fa-check"></i>Mid commitment projects (2-2.5/5 FTE) in 2023</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#full-time-positions-for-post-docs-in-2023"><i class="fa fa-check"></i>Full-time positions for post-docs in 2023</a></li>
 <li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#research-software-engineer"><i class="fa fa-check"></i>Research Software Engineer</a></li>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#post-doc"><i class="fa fa-check"></i>Post-doc</a></li>
 </ul></li>
 <li class="chapter" data-level="" data-path="contact.html"><a href="contact.html"><i class="fa fa-check"></i>Contact</a></li>
 </ul></li>
diff --git a/docs/research-grants.html b/docs/research-grants.html
index 8d75516..32149ee 100644
--- a/docs/research-grants.html
+++ b/docs/research-grants.html
@@ -92,10 +92,11 @@
 <li class="chapter" data-level="" data-path="the-team.html"><a href="the-team.html"><i class="fa fa-check"></i>The Team</a></li>
 <li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html"><i class="fa fa-check"></i>Open Positions</a>
 <ul>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#phd-positions-in-2023"><i class="fa fa-check"></i>PhD positions in 2023</a></li>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#deep-learning-engineer"><i class="fa fa-check"></i>Deep Learning Engineer</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#mi2-policy-on-benefits-and-obligations"><i class="fa fa-check"></i>MI2 policy on benefits and obligations</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#low-commitment-projects-15-fte-in-2023"><i class="fa fa-check"></i>Low commitment projects (1/5 FTE) in 2023</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#mid-commitment-projects-2-2.55-fte-in-2023"><i class="fa fa-check"></i>Mid commitment projects (2-2.5/5 FTE) in 2023</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#full-time-positions-for-post-docs-in-2023"><i class="fa fa-check"></i>Full-time positions for post-docs in 2023</a></li>
 <li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#research-software-engineer"><i class="fa fa-check"></i>Research Software Engineer</a></li>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#post-doc"><i class="fa fa-check"></i>Post-doc</a></li>
 </ul></li>
 <li class="chapter" data-level="" data-path="contact.html"><a href="contact.html"><i class="fa fa-check"></i>Contact</a></li>
 </ul></li>
diff --git a/docs/responsibleml-blog.html b/docs/responsibleml-blog.html
index 3346323..4181df0 100644
--- a/docs/responsibleml-blog.html
+++ b/docs/responsibleml-blog.html
@@ -92,10 +92,11 @@
 <li class="chapter" data-level="" data-path="the-team.html"><a href="the-team.html"><i class="fa fa-check"></i>The Team</a></li>
 <li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html"><i class="fa fa-check"></i>Open Positions</a>
 <ul>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#phd-positions-in-2023"><i class="fa fa-check"></i>PhD positions in 2023</a></li>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#deep-learning-engineer"><i class="fa fa-check"></i>Deep Learning Engineer</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#mi2-policy-on-benefits-and-obligations"><i class="fa fa-check"></i>MI2 policy on benefits and obligations</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#low-commitment-projects-15-fte-in-2023"><i class="fa fa-check"></i>Low commitment projects (1/5 FTE) in 2023</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#mid-commitment-projects-2-2.55-fte-in-2023"><i class="fa fa-check"></i>Mid commitment projects (2-2.5/5 FTE) in 2023</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#full-time-positions-for-post-docs-in-2023"><i class="fa fa-check"></i>Full-time positions for post-docs in 2023</a></li>
 <li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#research-software-engineer"><i class="fa fa-check"></i>Research Software Engineer</a></li>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#post-doc"><i class="fa fa-check"></i>Post-doc</a></li>
 </ul></li>
 <li class="chapter" data-level="" data-path="contact.html"><a href="contact.html"><i class="fa fa-check"></i>Contact</a></li>
 </ul></li>
diff --git a/docs/search_index.json b/docs/search_index.json
index d5ca818..693448a 100644
--- a/docs/search_index.json
+++ b/docs/search_index.json
@@ -1 +1 @@
-[["index.html", "MI² MI².AI", " MI² MI².AI MI².AI is a group of mathematicians and computer scientists that love to play with predictive models. We are spread between Warsaw University of Technology and University of Warsaw. Here we have workshops, seminars, here we are forging new ideas, creating tools, solving problems, doing consulting and sharing our positive attitude. Feel free to jump in. Mission Machine learning is like atomic energy. We develop leaders, skills, methods, tools and good practices so that predictive models can be deployed responsibly and sustainably. Vision MI² is a group of experts supporting global initiatives aimed at responsible and sustainable machine learning. We support the development of future leaders of responsible machine learning through internships, PhDs, postdoctoral fellowships and so on. We seek for research grants and business projects to conduct both scientific and applied research. We develop and maintain software and infrastructure necessary to build responsible and sustainable ML. We develop cooperation with international teams working on similar topics. We support companies to implement best practices related to responsible modelling in their operation. We conduct workshops and training on responsible predictive modelling. "],["the-team.html", "The Team", " The Team Members Przemysław Biecek, PhD, DSc (Team Leader) Hubert Baniecki, PhD student Mustafa Cavus, PhD Maciej Chrabąszcz, MSc student Mateusz Grzyb, MSc student Stanisław Giziński, MSc student Weronika Hryniewska-Guzik, PhD student Piotr Komorowski, MSc student Mateusz Krzyziński, MSc student Tymoteusz Kwieciński, BSc student Stanisław Łaniewski, PhD student Piotr Piątyszek, BSc student Hubert Ruczyński, MSc student Barbara Rychalska, PhD student Nuno Sepúlveda, PhD Bartek Sobieski, MSc student Mikołaj Spytek, MSc student Tomasz Stanisławek, PhD Hoang Thien Ly, BSc student Paulina Tomaszewska, PhD student Piotr Wilczyński, BSc student Emilia Wiśnios, MSc student Paweł Wojciechowski, BSc Katarzyna Woźnica, PhD student Vladimir Zaigrajew, PhD student Artur Żółkowski, BSc student Collaborators Mariusz Adamek, Prof, MD Przemysław Bombiński, PhD, MD André Fonseca, PhD student Katarzyna Kobylińska, PhD student Anna Kozak, MSc Marcin Luckner, PhD João Malato, PhD student Bartek Pieliński, PhD, DSc Hanna Piotrowska, MA Elżbieta Sienkiewicz, PhD Julian Sienkiewicz, PhD Adrian Stańdo, MSc student Patryk Szatkowski, PhD student, MD Jakub Wiśniewski, MSc student Alumni Piotr Czarnecki, MSc Alicja Gosiewska, MSc Adrianna Grudzień, BSc Maria Kałuska, BSc Marcin Kosiński, MSc Adam Kozłowski, MSc Wojciech Kretowicz, BSc Michał Kuźba, MSc Szymon Maksymiuk, BSc Tomasz Mikołajczyk, PhD Katarzyna Pękala, MSc Adam Rydelek, BSc Bartosz Sawicki, BSc Patryk Słowakiewicz, BSc Michał Sokólski, MSc Mateusz Stączek, BSc Szymon Szmajdziński, BSc Zuzanna Trafas, BSc Kinga Ułasik, BSc Anna Wróblewska, PhD Hanna Zdulska, BSc Przemysław Biecek My personal mission is to enhance human capabilities by supporting them through access to data-driven and knowledge-based predictions. I execute it by developing methods and tools for responsible machine learning, trustworthy artificial intelligence and reliable software engineering. I work as an associate professor at Warsaw University of Technology and the University of Warsaw. I graduated in software engineering and mathematical statistics and now work on model visualisation, explanatory model analysis, predictive modelling and data science for healthcare. In 2016, I formed the research group MI² which develops methods and tools for predictive model analysis. Google Scholar: Af0O75cAAAAJ GitHub: pbiecek LinkedIn: pbiecek Mariusz Adamek I work at two Medical Universities (Silesia and Gdańsk) holding a Professorship in Medicine and Health Sciences. My interests are focused on lung cancer prevention and screening, the latter by means of low-dose computed tomography (LDCT) with special emphasis put on molecular biology methods, prediction models and image analysis aimed to enhance the performance of lung screening outcomes. Website: www.mariuszadamek.io Hubert Baniecki I’m a PhD student in Computer Science at the University of Warsaw. Prior, I did my MSc (2022) and BSc (2021) in Data Science at Warsaw University of Technology. My main research interest is explainable machine learning, with particular emphasis on adversarial attacks &amp; explanation evaluation. Website: hbaniecki.com Mustafa Cavus I work as an assistant professor at Warsaw University of Technology and the Eskisehir Technical University. I joined the MI² DataLab as a post-doc researcher in 2021. I work on explainable artificial intelligence and AutoML. Google Scholar: I63d1WIAAAAJ&amp;hl GitHub: mcavus LinkedIn: mcavus Twitter: mcavus Julian Sienkiewicz I work as an assistant professor at Faculty of Physics, WUT. My main research area links with sociophysics, complex networks and agent-based models. In the scope of MI² DataLab I follow my other interest - scientometrics. Google Scholar: mIwu11QAAAAJ LinkedIn: julek-sienkiewicz-873829 Maciej Chrabąszcz Master’s student in mathematical statisctics at Warsaw University of Technology. Interested in deep learning on text and images, explainable and responsible AI. GitHub: maciejchrabaszcz LinkedIn: maciej-chrabaszcz Stanisław Giziński A Research Software Engineer and student of Machine Learning at Faculty of Mathematics Informatics and Mechanic, University of Warsaw. His work in the lab focuses on using natural language processing and network analysis to better understand the spread of AI public policies. Interested also in applying machine learning in bioinformatics. Google Scholar: Stanisław Giziński GitHub: Gizzio LinkedIn: stanislaw-gizinski Mateusz Grzyb MSc student in Data Science at Warsaw University of Technology. Interested in artificial intelligence and scientific computing, but above all simply enjoys programming. GitHub: mgrzyb99 Weronika Hryniewska-Guzik PhD candidate in computer science at Warsaw University of Technology. Interested in deep learning modelling on medical images in the context of explainability and responsible AI. Google Scholar: aJeg3IQAAAAJ GitHub: Hryniewska LinkedIn: weronikahryniewska Piotr Komorowski Master’s student in Machine Learning at the University of Warsaw. Mainly interested in image processing and XAI applied to medical images. GitHub: piotr-komorowski LinkedIn: Piotr-Komorowski Anna Kozak Graduated in mathematical statistics at Warsaw University of Technology. Interested in explainable artificial intelligence and data visualization. Organizes projects related to education. Google Scholar: JIrqf9kAAAAJ GitHub: kozaka93 LinkedIn: kozakanna Mateusz Krzyziński MSc student in Data Science at Warsaw University of Technology. Interested in explainable artificial intelligence, with particular emphasis on XAI methods for survival analysis models and XAI applications in the medical field. Also an enthusiast of data visualization. Google Scholar: i_r7EUgAAAAJ GitHub: krzyzinskim LinkedIn: krzyzinskim Tymoteusz Kwieciński BSc student in Data Science at Warsaw University of Technology. Particularly interested in explainable artificial intelligence, computer vision and NLP. GitHub: Fersoil LinkedIn: Tymoteusz-Kwieciński Stanisław Łaniewski PhD student in Quantitative Psychology and Economics at University of Warsaw, Machine Learning Researcher at MI2 Data Lab, Msc in Actuarial Science and Mathematical Finance at University of Amsterdam, former Quantitative Researcher at Flow Traders His research focuses on enhancing classical methods used in discrete choice and finance with machine learning and how to apply them to explain behavioral phenomena and heuristics. He is also keen on finding balance between best predictive models and their explainability. Avid gamer who applies statistical techniques to deepen the understanding of best strategies LinkedIn: Stanisław-Łaniewski Piotr Piątyszek Undergraduate Data Science student at Warsaw University of Technology. Works as a research software engineer on enhancing accessibility and completeness of explainable AI. During pandemic contributes to a system of monitoring covid variants. Github: piotrpiatyszek Bartosz Pieliński I am an Assistant Professor at the Faculty of Political Science and International Studies at Warsaw University. I am interested in applying quantitative methods to study public policies. I am a founding member of the Institutional Grammar Research Initiative, which is focused on developing a new way of analysing social rules. I have participated in several research projects covering social policy, non-profit organizations, social enterprises, and international organizations. Website: https://pielinski.info/ Google Scholar: hnWiaVEAAAAJ LinkedIn: Bartosz Pieliński Hanna Piotrowska Information designer, focusing mainly on data visualization, branding and book design, with a strong interest in Data Science and perception studies. Winner of numerous awards, including The Kantar Information Is Beautiful Awards, HOW International Design Awards, Polish Graphic Design Awards and KTR. LinkedIn: hanna-piotrowska Twitter: hannapio Behance: hannapio. Hubert Ruczyński I am working towards Masters’s degree in Data Science at Warsaw University of Technology. I am also teaching students about data exploration and visualisation. My major interests are: AutoML | Natural Language Processing | Data Visualization | Fairness. GitHub: HubertR21 LinkedIn: Hubert Ruczyński Barbara Rychalska PhD candidate in computer science at Warsaw University of Technology. Mainly interested in deep learning for natural language processing (NLP), recommender systems and graph-based learning. Google Scholar: Wp0wHJoAAAAJ LinkedIn: Barbara-Rychalska Bartek Sobieski MSc student in Data Science at Warsaw University of Technology. Interested in deep learning and hyperparameter optimization. GitHub: sobieskibj LinkedIn: Bartłomiej-Sobieski Mikołaj Spytek MSc student in Data Science at Warsaw University of Technology. Interested in explainable artificial intelligence, data vizualization and survival analysis. Google Scholar: 1u49AqYAAAAJ GitHub: mikolajsp LinkedIn: Mikołaj-Spytek Tomasz Stanisławek PhD candidate in computer science at Warsaw University of Technology. Mainly interested in deep learning for natural language processing (NLP). Google Scholar: gq8NY_UAAAAJ GitHub: tstanislawek LinkedIn: Tomasz-Stanisławek Paulina Tomaszewska PhD candidate in Computer Science at Warsaw University of Technology. Gained experience in AI at leading universities during: Deep Learning Summer School at Tsinghua University (China), one-semester exchange at Nanyang Technological University (Singapore) and research internships at Gwangju Institute of Science and Technology (South Korea) and Institute of Science and Technology (Austria). Mainly interested in Deep Learning, Computer Vision and Transfer Learning. Recently, focused on digital pathology. Google Scholar: eO245iMAAAAJ LinkedIn: paulina-tomaszewska Hoang Thien Ly Bachelor student in Maths and Data Analysis at Warsaw University of Technology. Interested in working with data, and learning explainable artificial intelligence methods. Google Scholar: JkysewYAAAAJ GitHub: lhthien09 LinkedIn: hthienly Piotr Wilczyński BSc student in Data Science at Warsaw University of Technology. Interested in Large Language Models, AI Deception and Natural Language Processing. Currently working on my thesis, which applies Computer Vision to medicine. GitHub: wi1lku LinkedIn: Piotr-Wilczyński Jakub Wiśniewski Research Software Engineer and third year Data Science student at Warsaw University of Technology. Developer of tools for bias detection and fairness. Currently researching responsible applications of deep learning. President of Data Science Science Club at WUT. Google Scholar: _6eQsXMAAAAJ GitHub: jakwisn LinkedIn: jakwisn Emilia Wiśnios Research Software Engineer and student of Machine Learning at Faculty of Mathematics, Informatics and Mechanics, University of Warsaw. Interested in natural language processing and reinforcement learning. GitHub: emiliawisnios LinkedIn: emilia-wisnios Paweł Wojciechowski Graduated with a bachelor’s degree in Data Science from Warsaw University of Technology. Interested in explainable artificial intelligence, computer vision, and active learning. GitHub: p-wojciechowski LinkedIn: wojciechowski-p Katarzyna Woźnica PhD candidate in computer science at Warsaw University of Technology. Graduated in mathematical statistics. Interested in automated machine learning especially in hyperparameter tuning for tabular data. Carrying statistical analysis and predictive modelling for healthcare. Google Scholar: tAQS1gQAAAAJ GitHub: woznicak LinkedIn: woznicak Vladimir Zaigrajew PhD candidate in computer science at Warsaw University of Technology. Interested in deep learning, primarily on images, with a focus on representation learning. GitHub: WolodjaZ LinkedIn: vladimir-zaigrajew Artur Żółkowski BSc student in Data Science at Warsaw University of Technology. Interested in explainable artificial intelligence, computer vision and NLP. GitHub: arturzolkowski LinkedIn: Artur-Żółkowski "],["open-positions.html", "Open Positions", " Open Positions We have open call for HOMER and xLungs projects for following positions. If you are interested in any of them please send your CV and Motivation Letter to przemyslaw.biecek at pw.edu.pl. We reserve the right to contact only selected candidates. PhD positions in 2023 Call for PhD Student: Research Opportunity in Foundation Model Redteaming We are seeking a highly motivated and talented PhD student to join MI2.AI research team in the exciting field of foundation model redteaming. As the advancements development of AI models (especially transformer based with attention mechanism), it becomes crucial to ensure the robustness and security of the foundation models that underlie these advancements. Project Description: The selected PhD student will work on investigating potential vulnerabilities and biases in foundation models, such as language models (one PhD position) and 2D/3D computer vision models (one PhD position). Redteaming involves conducting adversarial assessments to identify weaknesses, uncover potential attack vectors, and develop defensive strategies for foundation models/transformer models. The research will explore various aspects, including data poisoning attacks, adversarial samples generation, and the development of countermeasures. Requirements: Proficiency in programming languages commonly used in AI research, such as Python, R or Julia. Familiarity with deep learning frameworks and libraries, such as PyTorch. Solid understanding of foundation models, including their architecture and training processes. Excellent problem-solving skills and ability to work independently and as part of a team. Good communication skills and proficiency in academic writing. Interested candidates are requested to submit the following documents to przemyslaw.biecek at pw.edu.pl. We look forward to receiving your applications and welcoming an enthusiastic and dedicated PhD student to contribute to our foundation model redteaming research endeavors. Apply before June 30 the end of the day (June 10 in case of interest in MIM UW). Deep Learning Engineer Required: Background in Computer Science, Mathematics, Statistics or similar. Experience in Deep Learning for 2d/3d image data (*torch is a plus) Interest in medical applications Scope of work: Training of machine learning models for tabular and image data Responsible ML solutions for the healthcare domain Offer: Excellent atmosphere for work in a young and very active lab Conferences and training budget Short visits in cooperating abroad group Access to CPU / GPU clusters Elastic working hours that can be combined with studies Research Software Engineer Required: Background in Computer Science, Mathematics, Statistics or similar. Experience in Scientific Programming (R and/or Python and/or C++) Interest in applications, machine learning and explainable artificial intelligence. Scope of work: Training of machine learning models for tabular and image data Interpretable solutions for tabular and image data Interactive interfaces Responsible ML solutions for the healthcare domain Offer: Excellent atmosphere for work in a young and very active lab Conferences and training budget Short visits in cooperating abroad group Access to CPU / GPU clusters Elastic working hours that can be combined with studies Post-doc I am looking for a post-doc to join the MI2.AI team for one or two years within HOMER project. The ideal fit is someone with (experience in Medical Image Analysis with deep learning models) OR (someone with experience in AutoML for tabular data). Experience or interest in XAI / fairness will always be a huge plus in our team. We have a young and very energetic team focused on growth in AI. Full time position. Salary above academic average. Little or no didactics. Strong focus on doing things that are meaningful. Required: PhD in Computer Science, Mathematics, Statistics or similar. (Different background? We encourage cross-domain short research visits). Experience in Scientific Programming (R and/or Python and/or C++) Good track of Scientific Records Scope of work: Automated model exploration Interpretable measures for model performance Meta/transfer learning in automated model development Automated model validation Experiments with XAI for deep learning Offer: Excellent atmosphere for work in a young and very active ML lab Conferences and training budget Short visits in cooperating abroad group Access to CPU / GPU clusters Full-time job at Warsaw University of Technology plus results-driven extras contract for 6 months (short visit) / 12 months or 24 months (long visit) "],["contact.html", "Contact", " Contact Our rooms: 44 (DataLab - separate entrance in front of the main entrance) 316 (xLungs) 317 (HOMER) Faculty of Mathematics and Information Science, Warsaw University of Technology, Koszykowa 75, 00-662 Warszawa VAT: PL 5250005834 "],["mi2redteam.html", "MI²RedTeam", " MI²RedTeam MI²RedTeam analyses machine and deep learning predictive models through the lens of AI explainability, fairness, security and human trust. We develop methods and tools for explanatory model analysis and apply them in practice. MI²RedTeam is a group of researchers experienced in XAI who perform a rigorous evaluation of AI solutions in order to improve their transparency and security. We apply state-of-the-art methods and introduce new ones to tailor our analysis to the specific predictive task. We openly collaborate on various topics related to explainable and interpretable machine learning. Feel free to reach out to us with research ideas and development opportunities. We help organizations to better understand the vulnerabilities of their AI systems, and take steps to mitigate them. Our current core research topics of interest include: [ARES] Attack-Resistant Explanations towards Secure AI, i.e. a critical evaluation of the state-of-the-art analysis techniques [xSurvival] Explanatory analysis of machine learning survival models [Large Model Analysis] Explanatory analysis of large models, e.g. transformers Methods and methodologies introduced by our team: Evaluating explanations of vision transformers InteractiveEMA towards human-model interaction in explainable machine learning for tabular data SurvSHAP(t) for time-dependent analysis of machine learning survival models LIMEcraft for human-guided visual explanations of deep neural networks Fooling PD &amp; Manipulating SHAP for stress-testing widely-applied explanation methods Checklist towards responsible deep learning on medical images SAFE for lifting interpretability-performance trade-off via automated feature engineering WildNLP for stress-testing deep learning models in NLP Explanatory Model Analysis towards comprehensive examination of predictive models Tools developed by our team: DALEX, breakDown, auditor &amp; modelStudio for explainable machine learning in R dalex for explainable and fair machine learning in Python survex dedicated to explaining machine learning survival models fairmodels for fairness analysis of machine learning classification models Applications supported by our team: In medicine, we analyzed hundreds of models predicting among others: survival in uveal melanoma eye cancer, survival in sepsis, type of lung cancer, lung cancer risk in screening, lung cancer mortality, COVID-19 mortality, hospital length of stay, progression of Alzheimer’s disease. In credit scoring, we analyzed the transparency, auditability, and explainability of machine learning models. In football analytics, we analyzed expected goal models for performance analysis. … This initiative is generously supported by the following institutions. "],["mi²research.html", "MI²Research", " MI²Research On a mission to responsibly build machine learning predictive models. Responsible and sustainable predictive modelling is still a new and developing area. We are conducting a number of studies in this domain that examine predictive models applied to tabular data, computer vision or natural language processing models. We investigate the stability and robustness of various methods, work on explainability and transparency for simple and complex models. As part of our this effort, we develop open source software packages (usually in R and Python) for model explanatory analysis, publish scientific articles describing new methods or investigating properties of already known methods, and create educational materials, recommendations and examples of application in specific domains. If you want to find out more about what we are working on, check out our seminar, which is always open to those interested in responsible and sustainable data science. "],["papers.html", "Papers", " Papers Consolidated learning: a domain-specific model-free optimization strategy with validation on metaMIMIC benchmarks Katarzyna Woźnica, Mateusz Grzyb, Zuzanna Trafas, Przemysław Biecek Machine Learning (2023) This paper proposes a new formulation of the tuning problem, called consolidated learning, more suited to practical challenges faced by model developers, in which a large number of predictive models are created on similar datasets. We show that a carefully selected static portfolio of hyperparameter configurations yields good results for anytime optimization, while maintaining the ease of use and implementation. We demonstrate the effectiveness of this approach through an empirical study for the XGBoost algorithm and the newly created metaMIMIC benchmarks of predictive tasks extracted from the MIMIC-IV medical database. Towards Evaluating Explanations of Vision Transformers for Medical Imaging Piotr Komorowski, Hubert Baniecki, Przemysław Biecek CVPR Workshop on Explainable AI for Computer Vision (2023) This paper investigates the performance of various interpretation methods on a Vision Transformer (ViT) applied to classify chest X-ray images. We introduce the notion of evaluating faithfulness, sensitivity, and complexity of ViT explanations. The obtained results indicate that Layerwise relevance propagation for transformers outperforms Local interpretable model-agnostic explanations and Attention visualization, providing a more accurate and reliable representation of what a ViT has actually learned. Hospital Length of Stay Prediction Based on Multi-modal Data towards Trustworthy Human-AI Collaboration in Radiomics Hubert Baniecki, Bartlomiej Sobieski, Przemysław Bombiński, Patryk Szatkowski, Przemysław Biecek International Conference on Artificial Intelligence in Medicine (2023) To what extent can the patient’s length of stay in a hospital be predicted using only an X-ray image? We answer this question by comparing the performance of machine learning survival models on a novel multi-modal dataset created from 1235 images with textual radiology reports annotated by humans. We introduce time-dependent model explanations into the human-AI decision making process. For reproducibility, we open-source code and the TLOS dataset at this URL. SurvSHAP(t): Time-dependent explanations of machine learning survival models Mateusz Krzyziński, Mikołaj Spytek, Hubert Baniecki, Przemysław Biecek Knowledge-Based Systems (2023) In this paper, we introduce SurvSHAP(t), the first time-dependent explanation that allows for interpreting survival black-box models. The proposed methods aim to enhance precision diagnostics and support domain experts in making decisions. SurvSHAP(t) is model-agnostic and can be applied to all models with functional output. We provide an accessible implementation of time-dependent explanations in Python at this URL. The grammar of interactive explanatory model analysis Hubert Baniecki, Dariusz Parzych, Przemyslaw Biecek Data Mining and Knowledge Discovery (2023) This paper proposes how different Explanatory Model Analysis (EMA) methods complement each other and discusses why it is essential to juxtapose them. The introduced process of Interactive EMA (IEMA) derives from the algorithmic side of explainable machine learning and aims to embrace ideas developed in cognitive sciences. We formalize the grammar of IEMA to describe human-model interaction. We conduct a user study to evaluate the usefulness of IEMA, which indicates that an interactive sequential analysis of a model may increase the accuracy and confidence of human decision making. Climate Policy Tracker: Pipeline for automated analysis of public climate policies Artur Żółkowski, Mateusz Krzyziński, Piotr Wilczyński, Stanisław Giziński, Emilia Wiśnios, Bartosz Pieliński, Julian Sienkiewicz, Przemysław Biecek NeurIPS Workshop on Tackling Climate Change with Machine Learning (2022) In this work, we use a Latent Dirichlet Allocation-based pipeline for the automatic summarization and analysis of 10-years of national energy and climate plans (NECPs) for the period from 2021 to 2030, established by 27 Member States of the European Union. We focus on analyzing policy framing, the language used to describe specific issues, to detect essential nuances in the way governments frame their climate policies and achieve climate goals. Explainable expected goal models for performance analysis in football analytics Mustafa Cavus, Przemyslaw Biecek International Conference on Data Science and Advanced Analytics (2022) The expected goal provides a more representative measure of the team and player performance which also suit the low-scoring nature of football instead of the score in modern football. This paper proposes an accurate expected goal model trained on 315,430 shots from seven seasons between 2014-15 and 2020-21 of the top-five European football leagues. Moreover, we demonstrate a practical application of aggregated profiles to explain a group of observations on an accurate expected goal model for monitoring the team and player performance. Multi-omics disease module detection with an explainable Greedy Decision Forest Bastian Pfeifer, Hubert Baniecki, Anna Saranti, Przemyslaw Biecek, Andreas Holzinger Scientific Reports (2022) In this work, we demonstrate subnetwork detection based on multi-modal node features using a novel Greedy Decision Forest (GDF) with inherent interpretability. The latter will be a crucial factor to retain experts and gain their trust in such algorithms. To demonstrate a concrete application example, we focus on bioinformatics, systems biology and particularly biomedicine, but the presented methodology is applicable in many other domains as well. Our proposed explainable approach can help to uncover disease-causing network modules from multi-omics data to better understand complex diseases such as cancer. Interpretable meta-score for model performance Alicja Gosiewska, Katarzyna Woźnica, Przemysław Biecek Nature Machine Intelligence (2022) Elo-based predictive power (EPP) meta-score that is built on other performance measures and allows for interpretable comparisons of models. Differences between this score have a probabilistic interpretation and can be compared directly between data sets. Furthermore, this meta-score allows for an assessment of ranking fitness. We prove the properties of the Elo-based predictive power meta-score and support them with empirical results on a large-scale benchmark of 30 classification data sets. Additionally, we propose a unified benchmark ontology that provides a uniform description of benchmarks. fairmodels: a Flexible Tool for Bias Detection, Visualization, and Mitigation in Binary Classification Models Jakub Wiśniewski, Przemyslaw Biecek The R Journal (2022) This article introduces an R package fairmodels that helps to validate fairness and eliminate bias in binary classification models quickly and flexibly. It offers a model-agnostic approach to bias detection, visualization, and mitigation. The implemented functions and fairness metrics enable model fairness validation from different perspectives. In addition, the package includes a series of methods for bias mitigation that aim to diminish the discrimination in the model. The package is designed to examine a single model and facilitate comparisons between multiple models. A robust framework to investigate the reliability and stability of explainable artificial intelligence markers of Mild Cognitive Impairment and Alzheimer’s Disease Angela Lombardi, Domenico Diacono, Nicola Amoroso, Przemysław Biecek, Alfonso Monaco, Loredana Bellantuono, Ester Pantaleo, Giancarlo Logroscino, Roberto De Blasi, Sabina Tangaro, Roberto Bellotti Brain Informatics (2022) In this work, we present a robust framework to (i) perform a threefold classification between healthy control subjects, individuals with cognitive impairment, and subjects with dementia using different cognitive indexes and (ii) analyze the variability of the explainability SHAP values associated with the decisions taken by the predictive models. We demonstrate that the SHAP values can accurately characterize how each index affects a patient’s cognitive status. Furthermore, we show that a longitudinal analysis of SHAP values can provide effective information on Alzheimer’s disease progression. LIMEcraft: handcrafted superpixel selection and inspection for Visual eXplanations Weronika Hryniewska, Adrianna Grudzień, Przemysław Biecek Machine Learning (2022) LIMEcraft enhances the process of explanation by allowing a user to interactively select semantically consistent areas and thoroughly examine the prediction for the image instance in case of many image features. Experiments on several models show that our tool improves model safety by inspecting model fairness for image pieces that may indicate model bias. The code is available at: this URL. Fooling Partial Dependence via Data Poisoning Hubert Baniecki, Wojciech Kretowicz, Przemyslaw Biecek ECML PKDD (2022) We showcase that PD can be manipulated in an adversarial manner, which is alarming, especially in financial or medical applications where auditability became a must-have trait supporting black-box machine learning. The fooling is performed via poisoning the data to bend and shift explanations in the desired direction using genetic and gradient algorithms. Manipulating SHAP via Adversarial Data Perturbations (Student Abstract) Hubert Baniecki, Przemyslaw Biecek AAAI Conference on Artificial Intelligence (2022) We introduce a model-agnostic algorithm for manipulating SHapley Additive exPlanations (SHAP) with perturbation of tabular data. It is evaluated on predictive tasks from healthcare and financial domains to illustrate how crucial is the context of data distribution in interpreting machine learning models. Our method supports checking the stability of the explanations used by various stakeholders apparent in the domain of responsible AI; moreover, the result highlights the explanations’ vulnerability that can be exploited by an adversary. A Signature of 14 Long Non-Coding RNAs (lncRNAs) as a Step towards Precision Diagnosis for NSCLC Anetta Sulewska, Jacek Niklinski, Radoslaw Charkiewicz, Piotr Karabowicz, Przemyslaw Biecek, Hubert Baniecki, Oksana Kowalczuk, Miroslaw Kozlowski, Patrycja Modzelewska, Piotr Majewski et al. Cancers (2022) The aim of the study was the appraisal of the diagnostic value of 14 differentially expressed long non-coding RNAs (lncRNAs) in the early stages of non-small-cell lung cancer (NSCLC). We established two classifiers. The first recognized cancerous from noncancerous tissues, the second successfully discriminated NSCLC subtypes (LUAD vs. LUSC). Our results indicate that the panel of 14 lncRNAs can be a promising tool to support a routine histopathological diagnosis of NSCLC. dalex: Responsible Machine Learning with Interactive Explainability and Fairness in Python Hubert Baniecki, Wojciech Kretowicz, Piotr Piątyszek, Jakub Wiśniewski, Przemyslaw Biecek Journal of Machine Learning Research (2021) We introduce dalex, a Python package which implements a model-agnostic interface for interactive explainability and fairness. It adopts the design crafted through the development of various tools for explainable machine learning; thus, it aims at the unification of existing solutions. This library’s source code and documentation are available under open license at this URL. Checklist for responsible deep learning modeling of medical images based on COVID-19 detection studies Weronika Hryniewska, Przemysław Bombiński, Patryk Szatkowski, Paulina Tomaszewska, Artur Przelaskowski, Przemysław Biecek Pattern Recognition (2021) Our analysis revealed numerous mistakes made at different stages of data acquisition, model development, and explanation construction. In this work, we overview the approaches proposed in the surveyed Machine Learning articles and indicate typical errors emerging from the lack of deep understanding of the radiography domain. The final result is a proposed checklist with the minimum conditions to be met by a reliable COVID-19 diagnostic model. Towards explainable meta-learning Katarzyna Woźnica, Przemyslaw Biecek ECML PKDD Workshop on eXplainable Knowledge Discovery in Data Mining (2021) To build a new generation of meta-models we need a deeper understanding of the importance and effect of meta-features on the model tunability. In this paper, we propose techniques developed for eXplainable Artificial Intelligence (XAI) to examine and extract knowledge from black-box surrogate models. To our knowledge, this is the first paper that shows how post-hoc explainability can be used to improve the meta-learning. Prevention is better than cure: a case study of the abnormalities detection in the chest Weronika Hryniewska, Piotr Czarnecki, Jakub Wiśniewski, Przemysław Bombiński, Przemysław Biecek CVPR Workshop on “Beyond Fairness: Towards a Just, Equitable, and Accountable Computer Vision” (2021) In this paper, we analyze in detail a single use case - a Kaggle competition related to the detection of abnormalities in X-ray lung images. We demonstrate how a series of simple tests for data imbalance exposes faults in the data acquisition and annotation process. Complex models are able to learn such artifacts and it is difficult to remove this bias during or after the training. Simpler is better: Lifting interpretability-performance trade-off via automated feature engineering Alicja Gosiewska, Anna Kozak, Przemysław Biecek Decision Support Systems (2021) We propose a framework that uses elastic black boxes as supervisor models to create simpler, less opaque, yet still accurate and interpretable glass box models. The new models were created using newly engineered features extracted with the help of a supervisor model. We supply the analysis using a large-scale benchmark on several tabular data sets from the OpenML database. The first SARS-CoV-2 genetic variants of concern (VOC) in Poland: The concept of a comprehensive approach to monitoring and surveillance of emerging variants Radosław Charkiewicz, Jacek Nikliński, Przemysław Biecek, Joanna Kiśluk, Sławomir Pancewicz, Anna Moniuszko-Malinowska, Robert Flisiak, Adam Krętowski, Janusz Dzięcioł, Marcin Moniuszko, Rafał Gierczyński, Grzegorz Juszczyk, Joanna Reszeć Advances in Medical Sciences (2021) This study shows the first confirmed case of SARS-CoV-2 in Poland with the lineage B.1.351 (known as 501Y.V2 South African variant), as well as another 18 cases with epidemiologically relevant lineage B.1.1.7, known as British variant. Responsible Prediction Making of COVID-19 Mortality (Student Abstract) Hubert Baniecki, Przemyslaw Biecek AAAI Conference on Artificial Intelligence (2021) During the literature review of COVID-19 related prognosis and diagnosis, we found out that most of the predictive models are not faithful to the RAI principles, which can lead to biassed results and wrong reasoning. To solve this problem, we show how novel XAI techniques boost transparency, reproducibility and quality of models. Models in the Wild: On Corruption Robustness of Neural NLP Systems Barbara Rychalska, Dominika Basaj, Alicja Gosiewska, Przemyslaw Biecek International Conference on Neural Information Processing (2019) In this paper we introduce WildNLP - a framework for testing model stability in a natural setting where text corruptions such as keyboard errors or misspelling occur. We compare robustness of deep learning models from 4 popular NLP tasks: Q&amp;A, NLI, NER and Sentiment Analysis by testing their performance on aspects introduced in the framework. In particular, we focus on a comparison between recent state-of-the-art text representations and non-contextualized word embeddings. In order to improve robustness, we perform adversarial training on selected aspects and check its transferability to the improvement of models with various corruption types. We find that the high performance of models does not ensure sufficient robustness, although modern embedding techniques help to improve it. auditor: an R Package for Model-Agnostic Visual Validation and Diagnostics Alicja Gosiewska, Przemyslaw Biecek The R Journal (2019) This paper describes methodology and tools for model-agnostic auditing. It provides functinos for assessing and comparing the goodness of fit and performance of models. In addition, the package may be used for analysis of the similarity of residuals and for identification of outliers and influential observations. The examination is carried out by diagnostic scores and visual verification. The code presented in this paper are implemented in the auditor package. Its flexible and consistent grammar facilitates the validation models of a large class of models. Explanations of Model Predictions with live and breakDown Packages Mateusz Staniak, Przemyslaw Biecek The R Journal (2018) Complex models are commonly used in predictive modeling. In this paper we present R packages that can be used for explaining predictions from complex black box models and attributing parts of these predictions to input features. We introduce two new approaches and corresponding packages for such attribution, namely live and breakDown. We also compare their results with existing implementations of state-of-the-art solutions, namely, lime (Pedersen and Benesty, 2018) which implements Locally Interpretable Model-agnostic Explanations and iml (Molnar et al., 2018) which implements Shapley values. DALEX: Explainers for Complex Predictive Models in R Przemyslaw Biecek Journal of Machine Learning Research (2018) This paper describes a consistent collection of explainers for predictive models, a.k.a. black boxes. Each explainer is a technique for exploration of a black box model. Presented approaches are model-agnostic, what means that they extract useful information from any predictive method irrespective of its internal structure. Each explainer is linked with a specific aspect of a model. Every explainer presented here works for a single model or for a collection of models. In the latter case, models can be compared against each other. Presented explainers are implemented in the DALEX package for R. They are based on a uniform standardized grammar of model exploration which may be easily extended. archivist: An R Package for Managing, Recording and Restoring Data Analysis Results Przemyslaw Biecek, Marcin Kosiński Journal of Statistical Software (2017) Everything that exists in R is an object (Chambers 2016). This article examines what would be possible if we kept copies of all R objects that have ever been created. Not only objects but also their properties, meta-data, relations with other objects and information about context in which they were created. We introduce archivist, an R package designed to improve the management of results of data analysis. "],["software.html", "Software", " Software DALEX XAI with DALEX for R and Python survex Explainable machine learning in survival analysis Arena Interactive tool for the exploration and comparison of models’ explanations fairmodels Fairness with fairmodels COVID-19 COVID-19 Risk Score archivist Model governance with R (MLOps) "],["books.html", "Books", " Books Explanatory Model Analysis Explore, Explain, and Examine Predictive Models. With examples in R and Python Przemysław Biecek, Tomasz Burzykowski Chapman and Hall/CRC, New York (2021) Odkrywać! Ujawniać! Objaśniać! Zbiór esejów o sztuce prezentowania danych Essays on the art of data visualisation Przemysław Biecek University of Warsaw Press (2016) Analiza danych z programem R. Modele liniowe z efektami stałymi, losowymi i mieszanymi Data analysis in R. Linear models with fixed, random and mixed effects Przemysław Biecek Polish Scientific Publishers PWN (2013) "],["seminars.html", "Seminars", " Seminars We meet every Monday, at 10 am online or in MI2DataLab (room 044, Faculty of Mathematics and Information Science, Warsaw University of Technology). Join us at https://meet.google.com/nno-okiz-bxy (or http://meet.drwhy.ai/) List of topics and materials from past seminars: https://github.com/MI2DataLab/MI2DataLab_Seminarium "],["research-grants.html", "Research grants", " Research grants ARES 2022-2026 ARES: Attack-resistant Explanations toward Secure and trustworthy AI Machine learning explainability, fairness, robustness, and security are key elements of trustworthy Artificial Intelligence, an area of strategic importance. In this context, the main goals of the ARES project are: Develop adversarial attacks on state-of-the-art explanations to investigate vulnerabilities and limitations of the existing explainability and fairness approaches in machine learning. Introduce novel robust explanations that are stable against manipulation and intuitive to evaluate. Achieving the first goal primarily impacts various domains of research, which currently use (and explain) black-box models for knowledge discovery and decision-making, by highlighting vulnerabilities and limitations of their explanations. Achieving the second goal impacts more the broad machine learning domain as it aims at improving state-of-the-art by introducing robust explanations toward secure and trustworthy AI. Work on this project is financially supported by the Polish National Science Centre PRELUDIUM BIS grant 2021/43/O/ST6/00347. DARLING 2022-2024 DARLING: Deep Analysis of Regulations with Language Inference, Network analysis and institutional Grammar Aim of the project Developing the tools for automated analysis of content of legal documents leveraging Natural Language Processing, that will help understand the dynamic of change in public policies and variables influencing those changes. Those tools will be firstly used to analyse the case of development of policy subsystem regulating usage of AI in the European Union. Specific goals of the project Developing and evaluating multilingual models for issue classification for legal and public policy documents. Developing embedding-based topic modeling methods for legal and public policy documents suited for analysis of change of the topics between documents. Institutional grammar based analysis of changes in topics between different public policy documents, regulations and public consultation documents. Agent-based models predicting diffusion of issues in public policy documents. Methodology The core of the DARLING project is the issues and topic analysis in documents connected with regulations development using NLP tools. Issues analysis shall allow tracking how different options of AI operationalisation, ways the AI-connected threats are perceived as well as ideas regarding AI regulations are shared among three different types of texts: scientific, expert and legal ones. The extracted issues will then be subject to complex networks analysis and institutional grammar approach. The network analysis, backed by agent-based modeling, will be used to examine the flow of issues among the documents based on their vector-formed characteristics. On the other hand, the Institutional Grammar (IG) will be used to analyze the modality of issues, e.g., the tendency to regulate a specific aspect of AI in a given issue, its deontic character or its conditionality. In result the DARLING project will effect in the development of new methods to analyze legal documents connected to regulation based on deep text processing and links among the documents. An inter-institutional and interdisciplinary team of computer, political sciences and physics of complex systems scientists will elaborate new machine learning approaches to examine the regulation corpora, issues recognition, issues analysis by the means of IG as well as propose new methods of modeling the flow/changes of regulations based on complex networks tools. X-LUNGS 2021-2024 X-LUNGS: Responsible Artificial Intelligence for Lung Diseases The aim of the project is to support the process of identification of lesions visible on CT and lung x-rays. We intend to achieve this goal by building an information system based on artificial intelligence (AI) that will support the radiologist’s work by enriching the images with additional information. The unique feature of the proposed system is a trustworthy artificial intelligence module that: will reduce the image analysis time needed to detect lesions, will make the image evaluation process more transparent, will provide image and textual explanations indicating the rationale behind the proposed recommendation, will be verified for effective collaboration with the radiologist. Work on this project is financially supported from the INFOSTRATEG-I/0022/2021-00 grant funded by Polish National Centre for Research and Development (NCBiR). HOMER 2020-2025 HOMER: Human Oriented autoMated machinE leaRning One of the biggest challenges in the state-of-the-art machine learning is dealing with the complexity of predictive models. Recent techniques like deep neural networks, gradient boosting or random forests create models with thousands or even millions of parameters. This makes decisions generated by these black-box models completely opaque. Model obscurity undermines trust in model decisions, hampers model debugging, blocks model auditability, exposes models to problems with concept drift or data drift. Recently, there has been a huge progress in the area of model interpretability, which results in the first generation of model explainers, methods for better understanding of factors that drive model decisions. Despite this progress, we are still far from methods that provide deep explanations, confronted with domain knowledge that satisfies our ,,Right to explanation’’ as listed in the General Data Protection Regulation (GDPR). In this project I am going to significantly advance next generation of explainers for predictive models. This will be a disruptive change in the way how machine learning models are created, deployed, and maintained. Currently to much time is spend on handcrafted models produced in a tedious and laborious try-and-error process. The proposed Human-Oriented Machine Learning will focus on the true bottleneck in development of new algorithms, i.e. on model-human interfaces. The particular directions I consider are (1) developing an uniform grammar for visual model exploration, (2) establishing a methodology for contrastive explanations that describe similarities and differences among different models, (3) advancing a methodology for non-additive model explanations, (4) creating new human-model interfaces for effective communication between models and humans, (5) introducing new techniques for training of interpretable models based on elastic surrogate black-box models, (6) rising new methods for automated auditing of fairness, biases and performance of predictive models. Work on this project is financially supported from the SONATA BIS grant 2019/34/E/ST6/00052 funded by Polish National Science Centre (NCN). DeCoviD 2020-2022 DeCoviD: Detection of Covid-19 related markers of pulmonary changes using Deep Neural Networks models supported by eXplainable Artificial Intelligence and Cognitive Compressed Sensing Covid-19 is an infectious respiratory disease. A coronavirus infection leaves permanent ramifications in the respiratory system and beyond. In this situation, tools supporting diagnosis and assessment of lung damage after infection and during Covid-19 treatment are crucial. Preliminary results of analysis of CT images and lung xrays suggest that they can help to quickly assess even asymptomatic cases and facilitate prognosis of response to treatment. There are also reports of usefulness of ultrasound images. The aim of the DeCoviD project is to develop methods and tools to support radiologists in the assessment of lung imaging data for the occurrence of changes caused by Covid-19 disease. The developed solution will allow to automate the identification of pathological changes and will support the diagnosis of coexisting lung diseases as well as diseases of other organs visible on chest images. It will also allow to quantify the severity of lung damage caused by the disease Responsible decision support for radiologists requires models based on interpretable features. Such features will be stored in a hybrid knowledge base powered by two research teams from WUT, working on the basis of two, seemingly opposite, paradigms of image data analysis. The eXplainable Artificial Intelligence (XAI) team will use trained deep networks to automatically extract features that are essential for effective disease detection. Cognitive Compressed Sensing (CCS) will build a set of interpretable semantic features using sparse cognitive representations agreed with a group of cooperating radiologists. Combining these two approaches will achieve high effectiveness of the constructed models, combined with high transparency, clarity and stability of the solution. The DeCoviD project is a part of a broader strategy of competence development in the area of deep learning + XAI + medical applications at the Warsaw University of Technology. More information: https://github.com/MI2DataLab/DeCoviD. Work on this project is financially supported by the IDUB against COVID PW. DALEX 2018-2022 DALEX: Descriptive and model Agnostic Local EXplanations Research project objectives. Black boxes are complex machine learning models, for example deep neural network, an ensemble of trees of high-dimensional regression model. They are commonly used due to they high performance. But how to understand the structure of a black-box, a model in which decision rules are too cryptic for humans? The aim of the project is to create a methodology for such exploration. To address this issue we will develop methods, that: (1) identify key variables that mostly determine a model response, (2) explain a single model response in a compact visual way through local approximations, (3) enrich model diagnostic plots. Research project methodology. This project is divided into three subprojects - local approximations od complex models (called LIVE), explanations of particular model predictions (called EXPLAIN) and conditional explanations (called CONDA). Expected impact on the development of science. Explanations of black boxes have fundamental implications for the field of predictive and statistical modelling. The advent of big data forces imposes usage of black boxes that are easily able to overperform classical methods. But the high performance itself does not imply that the model is appropriate. Thus, especially in applications to personalized medicine or some regulated fields, one should scrutinize decision rules incorporated in the model. New methods and tools for exploration of black-box models are useful for quick identification of problems with the model structure and increase the interpretability of a black-box Work on this project is financially supported from the OPUS grant 2017/27/B/ST6/01307 funded by Polish National Science Centre (NCN). MLGenSig 2017-2021 MLGenSig: Machine Learning Methods for building of Integrated Genetic Signatures Research project objectives. The main scientific goal of this project is to develop a methodology for integrated genetic signatures based on data from divergent high-throughput techniques used in molecular biology. Integrated signatures base on ensembles of signatures for RNA-seq, DNA-seq, data as well for methylation profiles and protein expression microarrays. The advent of high throughput methods allows to measure dozens of thousands or even millions features on different levels like DNA / RNA / protein. And nowadays in many large scale studies scientists use data from mRNA seq to assess the state of transcriptome, protein microarrays to asses the state of proteome and DNA-seq / bisulfide methylation to assess genome / methylome. Research methodology. Genetic signatures are widely used in different applications, among others: for assessing genes that differentiate cells that are chemo resistant vs. cells that are not, assess the stage of cell pluripotency, define molecular cancer subtypes. For example, in database Molecular Signatures Database v5.0 one can find thousands of gene sets - genetic signatures for various conditions. There are signatures that characterize some cancer cells, pluripotent cells and other groups. But they usually contain relatively small number of genes (around 100), results with them are hard to replicate and they are collection of features that were found significant when independently tested. In most cases signatures are derived from measurements of the same type. Like signatures based of expression of transcripts based on data from microarrays or RNA-seq, or methylation profile or DNA variation. We are proposing a very different approach. First we are going to use machine-learning techniques to create large collections of signatures. Such signatures base on ensembles of small sub-signatures, are more robust and usually have higher precision. Then out of such signatures we are going to develop methodology for meta-signatures, that integrate information from different types of data (transcriptome, proteome, genome). Great examples of such studies are: Progenitor Cell Biology Consortium (PCBC) and The Cancer Genome Atlas (TCGA) studies. For thousands of patients in different cohorts (for PCBC cohorts based on stemness phenotype, for TCGA based on cancer type) measurements of both mRNA, miRNA, DNA and methylation profiles are available. New, large datasets require new methods that take into account high and dense structure of dependencies between features. The task that we are going to solve is to develop methodology that will create genetic signatures that integrate information from different levels of cell functioning. Then we are going to use data from TCGA and PBCB project to assess the quality of proposed methodology. As a baseline we are going to use following methodologies: DESeq, edgeR (for mRNA), casper (for lternative splicing), MethylKit (for RRBS data) and RPPanalyzer for protein arrays. Here is the skeleton for our approach: (1) Use ensembles in order to building a genetic signature. The first step would be to use random forests to train a new signature. Ensembles of sub-signtures are build on bootstrap subsamples and they votes if given sample fit given signature or not. (2) In order to improve signatures we are going to consider various normalization of raw counts. We start with log and rank transformation. (3) In order to improve the process of training an ensemble we are going to use pre-filtering of genes. (4) Another approach is to use Bayesian based methods, that may incorporate the expert knowledge, like belief-based gaussian modelling Research project impact. Genetic profiling is more and more important and has number of application starting from basic classification up to personalized medicine in which patients are profiled against different signatures. Existing tools for genetic signatures have many citations. This we assume that the methodology for integrated genetic profiling will be a very useful for many research groups. It is hard to overestimate the impact of better genetic profiling on medicine. Moreover we build a team of people with knowledge in cancer genetic profiling Work on this project is financially supported from the OPUS grant 2016/21/B/ST6/02176 funded by Polish National Science Centre (NCN). "],["mi²solutions.html", "MI²Solutions", " MI²Solutions Hire a team of experienced researchers. The blue team will help you develop good predictive models, create a responsible solution tailored to your needs. The red team will help you find and analyse any weaknesses in your predictive models. It will help you confront them with domain knowledge and make sure they are resilient to future changes in the data. If you need tailor-made solutions for your individual needs, we are happy to help you too. Contact us, we can develop software for you, deploy it, provide training, discuss your needs, verify the quality of your existing solutions. Below you will find a sample offer for trainings or deployments. Research as a service Our team has experience not only in groundbreaking research, but also in deploying these research into business. There are many ways we can help, for example help in delivery of champion-challenger evaluations in which we look for potential to increase the effectiveness of predictive models in your company. take care of the whole life cycle of the predictive models, from reproducibility of results to constant monitoring and continuous improvement of the model. audit models and analyse the sensitivity and vulnerability of the model to incorrect or unexpected behaviours. We would be happy to discuss how we could help with your organisation! Trainings Based on our experience in the area of Responsible Machine Learning, developed a unique two-day hands-on training. Jump into the topic of eXplainable Artificial Intelligence with our trainers. Responsible Machine Learning images/training_xai The training is conducted once a month in small groups online. Small groups encourage questions and the interactions within the group. Language Depending on the group’s preference, the hands-on part can be carried out in R (mlr3 + DALEX) or Python (scikit-learn + dalex). The methodology part does not depend on the language. Book your training To book a training please contact with trainings(at)solutions42.ai. "],["mi²education.html", "MI²Education", " MI²Education The demand for predictive modelling skills is growing at a furious rate. Part of our mission is to develop human capital so that predictive modelling is applied responsibly and safely. We take social responsibility seriously and as part of our activities we support the development of data analysis skills among pupils, students and senior professionals alike. "],["teaching.html", "Teaching", " Teaching Programming R 22/23 Summer Programming and data analysis advanced in R lectures, labs, projects - Anna Kozak Exploratory Data Analysis 22/23 Summer Introduction to exploratory data analysis for Mathematics and data analysis studies lectures, labs, projects - Anna Kozak labs - Hubert Ruczyński, Bartłomiej Sobieski Data Visualization 22/23 Winter Data Visualization Techniques for Data Science studies lectures, labs, projects - Anna Kozak labs, projects - Mateusz Krzyziński, Hubert Ruczyński, Mikołaj Spytek Exploratory Data Analysis 21/22 Summer Introduction to exploratory data analysis for Mathematics and data analysis studies lectures, labs, projects - Anna Kozak labs - Katatzyna Woźnica Interpretable Machine Learning 21/22 Summer Interpretable Machine Learning lectures, projects - Przemysław Biecek Case Studies 21/22 Summer Case Studies for Data Science studies lectures - Weronika Hryniewska ML-1 - labs, projects - Anna Kozak ML-2 - labs, projects - Bartłomiej Eljasiak XAI-tabular - labs, projects - Mustafa Cavus AutoML - labs, projects - Katarzyna Woźnica XIC - labs, projects - Hubert Baniecki TL - labs, projects - Paulina Tomaszewska Data - labs, projects - Weronika Hryniewska NLP - labs, projects - Stanisław Giziński Data Visualization 21/22 Winter Data Visualization Techniques for Data Science studies lectures, labs, projects - Anna Kozak labs, projects - Hubert Baniecki Exploratory Data Analysis 20/21 Summer Introduction to exploratory data analysis for Mathematics and data analysis studies lectures, labs, projects - Anna Kozak labs - Krzysztof Spaliński Case Studies 20/21 Summer Case Studies for Data Science studies lectures - Katarzyna Woźnica XAI1 - labs, projects - Anna Kozak XAI2 - labs, projects - Szymon Maksymiuk DL1 - labs, projects - Weronika Hryniewska DL2 - labs, projects - Paulina Tomaszewska ML - labs, projects - Hubert Baniecki RashomonML - labs, projects - Katarzyna Woźnica Interpretable Machine Learning 20/21 Summer Interpretable Machine Learning for Data Science studies XAI stories 2 lectures, projects - Przemysław Biecek Data Visualization 20/21 Winter Data Visualization Techniques for Data Science studies lectures, labs - Alicja Gosiewska projects - Hubert Baniecki Case Studies 19/20 Summer Case Studies for Data Science studies lectures - Alicja Gosiewska Imputation - labs, projects - Katarzyna Woźnica Reproducibility of scientific papers - labs, projects - Alicja Gosiewska Interpretability - labs, projects - Katarzyna Kobylińska Interpretable Machine Learning 19/20 Summer Interpretable Machine Learning for Data Science studies lectures, projects - Przemysław Biecek Data Visualization 19/20 Summer Data Visualization for Data Science studies lectures, labs, projects - Michał Burdukiewicz "],["beta-bit.html", "Beta Bit", " Beta Bit Chaos Game EN: Are you curious about fractals? The Chaos Game is the book for you. You will learn the mathematical basis behind these figures, find out what algorithm can be used to code them, write code in your favourite programming language (Python, R, Julia?) and also explore the bibliographies of three mathematicians associated with the development of mathematics around these shapes. This is the next book in the Beta Bit series for anyone interested in computational mathematics and data analysis. PL: Jesteś ciekawy czym są fraktale? Gra w Chaos to książka dla Ciebie. Poznasz matematyczne podstawy tych figur, dowiesz się, jaki algorytm można wykorzystać do ich zaprogramowania, napiszesz kod w swoim ulubionym języku programowania (Python, R, Julia?), a także poznasz bibliografie trzech matematyków związanych z rozwojem matematyki wokół tych kształtów. To kolejna książka z serii Beta Bit dla wszystkich zainteresowanych matematyką obliczeniową i analizą danych. Flipbook online [ENG] Flipbook online [POL] Wykresy od kuchni PL: Jak tworzyć dobre wykresy? Dobre, czyli takie, które z przyjemnością się ogląda, z których można wyciągnąć wiele informacji, które są zrozumiałe dla szerokiego odbiorcy, a jednocześnie docenią je smakosze. Na bazie doświadczeń z prowadzenia tych zajęć powstały Wykresy od kuchni. To zbiór krótkich wykładów omawiających różne wątki przydatne w lepszym zrozumieniu tego, jak działa komunikacja z użyciem wykresów statystycznych. Na kolejnych stronach pojawi się wiele analogii do przyrządzania posiłków, ponieważ zarówno w kuchni, jak i w przygotowaniu wykresów statystycznych potrzebna jest praktyka, znajomość pewnych fundamentalnych prawideł, garść sprawdzonych przepisów i dużo zapału do eksperymentowania. Będąc tak uzbrojonym, każdy adept sztuki kulinarnej jest skazany na sukces. Flipbook online [POL] The Hitchhiker’s Guide to Responsible Machine Learning EN: A one-of-a-kind 52-page story about responsible machine learning. Beta and Bit use decision trees, random forests, and AutoML tools to build a risk model after a covid infection, and then use explainable artificial intelligence tools to analyze the behavior of that model. The description of the data analysis process is intertwined with descriptions of ML tools and code snippets. All examples are fully reproducible! PL: Jedyna w swoim rodzaju 52-stronicowa opowieść o odpowiedzialnym uczeniu maszynowym. Beta i Bit używają drzew decyzyjnych, lasów losowych i narzędzi AutoML do budowy modelu ryzyka po zakażeniu covid, a następnie używają narzędzi wyjaśnialnej sztucznej inteligencji by przeanalizować działanie tego modelu. Opis procesu analizy danych przeplata się na opisem kolejnych narzędzi i przykładami kodu. Wszystkie wyniki są całkowicie odtwarzalne! Flipbook online Przemysław Biecek, Anna Kozak, Aleksander Zawada Fundacja Naukowa SmarterPoland.pl. 2022 W pogoni za nieskończonością. Szeregi EN: What does hiking in the mountains have to do with the convergence of series? Quite a lot! We start with the paradoxes related to infinity, but step by step we learn the techniques of geometric series. In this book, the conditions for convergence are explained, together with numerous examples. The comic ends with a collection of exercises with different levels of difficulty. PL: Co wspólnego ma chodzenie po górach ze zbieżnością szeregów? Otóż całkiem sporo! Zaczynamy od paradoksów związanych z nieskończonością, ale krok po kroku poznajemy techniki szeregów geometrycznych. W tej pozycji wyjaśnione są warunki zbieżności wraz z licznymi przykładami. Komiks kończy zbiór zadań o różnych poziomach trudności. Flipbook online Przemysław Biecek, Łukasz Maciejewski, Aleksander Zawada Fundacja Naukowa SmarterPoland.pl. 2022 Przewodnik po pakiecie R EN: The Guide to the R package was the first published Polish book focused on the R language. The current fourth edition consists of four parts: Basics of using R (+tidyverse, shiny, knitr and other goodies), Programming in R (object-oriented, package development, class system), Statistics with R (statistical tests, models, exploration techniques) and Visualization with R (graphics, lattice and ggplot2 packages). PL: Przewodnik po pakiecie R był pierwszą wydaną polskojęzyczną książką poświęconą językowi R. Aktualne czwarte wydanie składa się z czterech części: Podstaw posługiwania się językiem R (+tidyverse, shiny, knitr i inne smaczki), Programowanie w R (obiektowe, tworzenie pakietów, system klas), Statystyka z R (testy statystyczne, modele, techniki eksploracji) i Wizualizacja z R (pakiety graphics, lattice i ggplot2). Wersja online, Książka w księgarnii. Przemysław Biecek Wydawnictwo GiS. 2008-2021 Analiza danych z programem R EN: An academic textbook describing estimation and testing topics for linear models with fixed effects, random effects and mixed effects. The theoretical introduction is complemented by numerous examples for one-way and multivariate ANOVA, one and multiple random components. The examples focus on biological and medical applications and are based on real analyses of real data. PL: Podręcznik akademicki opisujący zagadnienia estymacji i testowania dla modeli liniowych z efektami stałymi, losowymi i mieszanymi. Wprowadzenie teoretyczne jest uzupełnione o liczne przykłady dla jednokierunkowej i wielokierunkowej ANOVA, jednym i wieloma komponentami losowymi. Przykłady dotyczą głównie zastosowań biologicznych i medycznych i bazują na prawdziwych analizach rzeczywistych danych. Książka w księgarnii. Przemysław Biecek Wydawnictwo Naukowe PWN 2013-2018 Eseje o sztuce wizualizacji danych EN: Discover! Reveal! Explain! These three roles can be fulfilled by good statistical graphics. Good means understandable, faithful to the data, aesthetic. How to create such graphics? A collection of essays on the art of displaying data systematises knowledge useful in designing and producing good data visualisations. It is not easy. On the one hand, we can fall into the trap of a colourful mush full of numbers, which is sometimes proudly called infographics. On the other hand, we can fall into the trap of graphics that perfectly reproduce the complexity of numbers, and thus completely incomprehensible. Somewhere in the middle is a graphic that explains, that informs, that is aesthetically pleasing and informative. PL: Odkrywać! Ujawniać! Objaśniać! Te trzy role może spełniać dobra grafika statystyczna. Dobra czyli zrozumiała, wierna danym, estetyczna. Jak tworzyć taką grafikę? Zbiór esejów o sztuce pokazywania danych systematyzuje wiedzę przydatną do projektowania i wykonania dobrej wizualizacji danych. Nie jest to proste. Z jednej strony możemy wpaść w pułapkę pstrokatej papki najeżonej liczbami, którą czasem dumnie nazywa się infografiką. Z drugiej strony wpaść można w pułapkę grafiki idealnie odwzorowującej złożoność liczb a przez to zupełnie niezrozumiałej. Gdzieś po środku jest grafika, która wyjaśnia, która informuje, która jest estetyczna i informatywna. Książka online, Książka w księgarnii. Przemysław Biecek Wydawnictwo SmarterPoland 2008-2021 Pogromcy Danych EN: Data Crunchers is the first MOOC (Massive Open Online Course) developed in Polish for data scientists. Two modules were developed in 2015: the first one is an introduction to R, with loading data, overview of syntax, basic data types, descriptive statistics and pipelined processing. The second module is dedicated to data visualisation and statistical modelling. More than 8,000 people have registered on the Data Crunchers platform. PL: Pogromcy Danych to pierwszy MOOC (Massive Open Online Course) opracowany w języku polskim do analizy danych. W roku 2015 powstały dwa moduły: pierwszy jest wprowadzeniem do programu R, przez wczytywanie danych, omówienie składni, podstawowych typów danych, statystyk opisowych oraz przetwarzania potokowego. Drugi moduł poświęcony jest wizualizacji danych oraz modelowaniu statystycznemu. W platformie Pogromców Danych zarejestrowało się ponad 8000 osób. Przetwarzanie danych w programie R, Wizualizacja i modelowanie, Strona WWW. Przemysław Biecek ICM UW. 2015 Wykresy unplugged EN: Can you create clear charts without any electricity? An illustrated collection of exercises showing eight of the most popular ways to visualise data, with do-it-yourself challenges. Grab your crayons and start creating fantastic charts. PL: Czy można tworzyć czytelne wykresy bez użycia prądu? Ilustrowany zbiór ćwiczeń przedstawiających osiem najpopularniejszych sposobów wizualizacji danych, wraz z zadaniami do samodzielnego wykonania. Weź kredki i zacznij tworzyć fantastyczne wykresy. Flipbook online, Komiks w księgarnii. Przemysław Biecek, Ewa Baranowska, Piotr Sobczyk Fundacja Naukowa SmarterPoland.pl. 2018 W pogoni za nieskończonością EN: Two mathematicians share stories about infinity. In the first Beta attends a lecture on the properties of prime numbers. In the second, Bit breaks into the Palace of Culture and Science. How should we talk about mathematics? PL: Dwójka matematyków wymienia się opowiadaniami o nieskończoności. W pierwszym Beta bierze udział w wykładzie o właściwościach liczb pierwszych. W drugim Bit włamuje się do Pałacu Kultury i Nauki. Jak opowiadać o matematyce? Flipbook online, Komiks w księgarnii. Przemysław Biecek, Łukasz Maciejewski, Tomasz Samojlik, Sebastian Szpakowski Fundacja Naukowa SmarterPoland.pl. 2018 Jak długo żyją Muffinki? EN: A collection of three stories for children showing statistical relationships in the world around us. Beautifully illustrated stories about the distribution of height according to age, the life span of dogs or measuring the weight of trees. PL: Zbiór trzech opowiadań dla dzieci pokazującym zależności statystyczne w świecie wokół nas. Pięknie ilustrowane opowiadania o rozkładzie wzrostu w zależności od wieku, czasie życia psów czy pomiarze wagi drzew. Online: Jak szybko urosnę, Jak długo żyją Muffinki. Przemysław Biecek Fundacja Naukowa SmarterPoland.pl.2016 Pieczara Pietraszki EN: How linear regression can help in getting home, and why it’s not worth hacking into a mad mathematician’s office. A short story describing the adventures of two teenagers Beta and Bit moving around historic Warsaw. PL: W jaki sposób regresja liniowa może pomóc w powrocie do domu, oraz dlaczego nie warto włamywać się do pokoju szalonego matematyka? Lekkie opowiadanie opisujące przygody dwójki nastolatków Bety i Bita w historycznej Warszawie. Online: W jezyku Polskim, In English, По-Русски. Magda Chudzian, Przemysław Biecek Fundacja Naukowa SmarterPoland.pl. 2015 How to weight a dog with a ruler? EN: Workshop materials for children aged 8-10. Kids measure different parameters of their body, such as arm span or height. Then they create a graph summarizing the collected data and look for relations between the measured features. It just so happens that parts of the human body are proportional to each other and you can use a ruler to find this relationship. Part of the StatTub project. PL: Materiały do warsztaty dla dzieci w wieku 8-10. Dzieci mierzą różne parametry swojego ciała, takie jak rozpiętość ramion lub wzrost. Następnie tworzą wykres podsumowujący zebrane dane i szukają zależności pomiędzy zmierzonymi cechami. Tak się składa, że części ciała ludzkiego są do siebie proporcjonalne i można z użyciem linijki znaleźć tę relację. Część projektu StatTuba. Online: English, Polish, Chinese, Simplified Chinese, Czech, German, Spanish, Spanish (Latin America), French, Dutch, Vietnamese. Przemysław Biecek, Klaudia Korniluk Fundacja Naukowa SmarterPoland.pl.2016-2021 "],["responsibleml-blog.html", "ResponsibleML Blog", " ResponsibleML Blog Read more about the research, solutions and education on our blog: Tools for Explainable, Fair and Responsible ML BASIC XAI with DALEX— Part 1: Introduction Anna Kozak In this post, we will take a closer look at some algorithms used in explainable artificial intelligence. You will find here an introduction to methods of global and local model evaluation. Each description will include a technical introduction, example analysis, and code in R and Python. R packages for eXplainable Artificial Intelligence Przemysław Biecek We have prepared an overview of the most popular R-packages, which can be used to build interpretable models or to explore complex ones. Examples of knitr notebooks for more than 30 packages are available at http://xai-tools.drwhy.ai/. Adversarial attacks on Explainable AI Hubert Baniecki There are various adversarial attacks on machine learning models; hence, ways of defending, e.g. by using Explainable AI methods. Nowadays, attacks on model explanations come to light, so does the defense to such adversary. Here, we introduce fundamental concepts related to the domain. A further reference list is available at https://github.com/hbaniecki/adversarial-explainable-ai. "],["404.html", "Page not found", " Page not found The page you requested cannot be found (perhaps it was moved or renamed). You may want to try searching to find the page's new location, or use the table of contents to find the page you are looking for. "]]
+[["index.html", "MI² MI².AI", " MI² MI².AI MI².AI is a group of mathematicians and computer scientists who love to play with predictive models. We are spread between Warsaw University of Technology and University of Warsaw. Here we have workshops and seminars, here we are forging new ideas, creating tools, solving problems, doing consulting and sharing our positive attitude. Feel free to jump in. Mission Machine learning is like atomic energy. We develop leaders, skills, methods, tools and good practices so that predictive models can be deployed responsibly and sustainably. Vision MI² is a group of experts supporting global initiatives aimed at responsible and sustainable machine learning. We support the development of future leaders of responsible machine learning through internships, PhDs, postdoctoral fellowships and so on. We seek for research grants and business projects to conduct both scientific and applied research. We develop and maintain the software and infrastructure necessary to build responsible and sustainable ML. We develop cooperation with international teams working on similar topics. We support companies to implement best practices related to responsible modelling in their operation. We conduct workshops and training on responsible predictive modelling. "],["the-team.html", "The Team", " The Team Members Przemysław Biecek, PhD, DSc (Team Leader) Hubert Baniecki, PhD student Mustafa Cavus, PhD Maciej Chrabąszcz, MSc student Mateusz Grzyb, MSc student Stanisław Giziński, MSc student Weronika Hryniewska-Guzik, PhD student Piotr Komorowski, MSc student Mateusz Krzyziński, MSc student Tymoteusz Kwieciński, BSc student Stanisław Łaniewski, PhD student Piotr Piątyszek, BSc student Hubert Ruczyński, MSc student Barbara Rychalska, PhD student Nuno Sepúlveda, PhD Bartek Sobieski, MSc student Mikołaj Spytek, MSc student Tomasz Stanisławek, PhD Hoang Thien Ly, BSc student Paulina Tomaszewska, PhD student Piotr Wilczyński, BSc student Emilia Wiśnios, MSc student Paweł Wojciechowski, BSc Katarzyna Woźnica, PhD student Vladimir Zaigrajew, PhD student Artur Żółkowski, BSc student Collaborators Mariusz Adamek, Prof, MD Przemysław Bombiński, PhD, MD André Fonseca, PhD student Katarzyna Kobylińska, PhD student Anna Kozak, MSc Marcin Luckner, PhD João Malato, PhD student Bartek Pieliński, PhD, DSc Hanna Piotrowska, MA Elżbieta Sienkiewicz, PhD Julian Sienkiewicz, PhD Adrian Stańdo, MSc student Patryk Szatkowski, PhD student, MD Jakub Wiśniewski, MSc student Alumni Piotr Czarnecki, MSc Alicja Gosiewska, MSc Adrianna Grudzień, BSc Maria Kałuska, BSc Marcin Kosiński, MSc Adam Kozłowski, MSc Wojciech Kretowicz, BSc Michał Kuźba, MSc Szymon Maksymiuk, BSc Tomasz Mikołajczyk, PhD Katarzyna Pękala, MSc Adam Rydelek, BSc Bartosz Sawicki, BSc Patryk Słowakiewicz, BSc Michał Sokólski, MSc Mateusz Stączek, BSc Szymon Szmajdziński, BSc Zuzanna Trafas, BSc Kinga Ułasik, BSc Anna Wróblewska, PhD Hanna Zdulska, BSc Przemysław Biecek My personal mission is to enhance human capabilities by supporting them through access to data-driven and knowledge-based predictions. I execute it by developing methods and tools for responsible machine learning, trustworthy artificial intelligence and reliable software engineering. I work as an associate professor at Warsaw University of Technology and the University of Warsaw. I graduated in software engineering and mathematical statistics and now work on model visualisation, explanatory model analysis, predictive modelling and data science for healthcare. In 2016, I formed the research group MI² which develops methods and tools for predictive model analysis. Google Scholar: Af0O75cAAAAJ GitHub: pbiecek LinkedIn: pbiecek Mariusz Adamek I work at two Medical Universities (Silesia and Gdańsk) holding a Professorship in Medicine and Health Sciences. My interests are focused on lung cancer prevention and screening, the latter by means of low-dose computed tomography (LDCT) with special emphasis put on molecular biology methods, prediction models and image analysis aimed to enhance the performance of lung screening outcomes. Website: www.mariuszadamek.io Hubert Baniecki I’m a PhD student in Computer Science at the University of Warsaw. Prior, I did my MSc (2022) and BSc (2021) in Data Science at Warsaw University of Technology. My main research interest is explainable machine learning, with particular emphasis on adversarial attacks &amp; explanation evaluation. Website: hbaniecki.com Mustafa Cavus I work as an assistant professor at Warsaw University of Technology and the Eskisehir Technical University. I joined the MI² DataLab as a post-doc researcher in 2021. I work on explainable artificial intelligence and AutoML. Google Scholar: I63d1WIAAAAJ&amp;hl GitHub: mcavus LinkedIn: mcavus Twitter: mcavus Julian Sienkiewicz I work as an assistant professor at Faculty of Physics, WUT. My main research area links with sociophysics, complex networks and agent-based models. In the scope of MI² DataLab I follow my other interest - scientometrics. Google Scholar: mIwu11QAAAAJ LinkedIn: julek-sienkiewicz-873829 Maciej Chrabąszcz Master’s student in mathematical statisctics at Warsaw University of Technology. Interested in deep learning on text and images, explainable and responsible AI. GitHub: maciejchrabaszcz LinkedIn: maciej-chrabaszcz Stanisław Giziński A Research Software Engineer and student of Machine Learning at Faculty of Mathematics Informatics and Mechanic, University of Warsaw. His work in the lab focuses on using natural language processing and network analysis to better understand the spread of AI public policies. Interested also in applying machine learning in bioinformatics. Google Scholar: Stanisław Giziński GitHub: Gizzio LinkedIn: stanislaw-gizinski Mateusz Grzyb MSc student in Data Science at Warsaw University of Technology. Interested in artificial intelligence and scientific computing, but above all simply enjoys programming. GitHub: mgrzyb99 Weronika Hryniewska-Guzik PhD candidate in computer science at Warsaw University of Technology. Interested in deep learning modelling on medical images in the context of explainability and responsible AI. Google Scholar: aJeg3IQAAAAJ GitHub: Hryniewska LinkedIn: weronikahryniewska Piotr Komorowski Master’s student in Machine Learning at the University of Warsaw. Mainly interested in image processing and XAI applied to medical images. GitHub: piotr-komorowski LinkedIn: Piotr-Komorowski Anna Kozak Graduated in mathematical statistics at Warsaw University of Technology. Interested in explainable artificial intelligence and data visualization. Organizes projects related to education. Google Scholar: JIrqf9kAAAAJ GitHub: kozaka93 LinkedIn: kozakanna Mateusz Krzyziński MSc student in Data Science at Warsaw University of Technology. Interested in explainable artificial intelligence, with particular emphasis on XAI methods for survival analysis models and XAI applications in the medical field. Also an enthusiast of data visualization. Google Scholar: i_r7EUgAAAAJ GitHub: krzyzinskim LinkedIn: krzyzinskim Tymoteusz Kwieciński BSc student in Data Science at Warsaw University of Technology. Particularly interested in explainable artificial intelligence, computer vision and NLP. GitHub: Fersoil LinkedIn: Tymoteusz-Kwieciński Stanisław Łaniewski PhD student in Quantitative Psychology and Economics at University of Warsaw, Machine Learning Researcher at MI2 Data Lab, Msc in Actuarial Science and Mathematical Finance at University of Amsterdam, former Quantitative Researcher at Flow Traders His research focuses on enhancing classical methods used in discrete choice and finance with machine learning and how to apply them to explain behavioral phenomena and heuristics. He is also keen on finding balance between best predictive models and their explainability. Avid gamer who applies statistical techniques to deepen the understanding of best strategies LinkedIn: Stanisław-Łaniewski Piotr Piątyszek Undergraduate Data Science student at Warsaw University of Technology. Works as a research software engineer on enhancing accessibility and completeness of explainable AI. During pandemic contributes to a system of monitoring covid variants. Github: piotrpiatyszek Bartosz Pieliński I am an Assistant Professor at the Faculty of Political Science and International Studies at Warsaw University. I am interested in applying quantitative methods to study public policies. I am a founding member of the Institutional Grammar Research Initiative, which is focused on developing a new way of analysing social rules. I have participated in several research projects covering social policy, non-profit organizations, social enterprises, and international organizations. Website: https://pielinski.info/ Google Scholar: hnWiaVEAAAAJ LinkedIn: Bartosz Pieliński Hanna Piotrowska Information designer, focusing mainly on data visualization, branding and book design, with a strong interest in Data Science and perception studies. Winner of numerous awards, including The Kantar Information Is Beautiful Awards, HOW International Design Awards, Polish Graphic Design Awards and KTR. LinkedIn: hanna-piotrowska Twitter: hannapio Behance: hannapio. Hubert Ruczyński I am working towards Masters’s degree in Data Science at Warsaw University of Technology. I am also teaching students about data exploration and visualisation. My major interests are: AutoML | Natural Language Processing | Data Visualization | Fairness. GitHub: HubertR21 LinkedIn: Hubert Ruczyński Barbara Rychalska PhD candidate in computer science at Warsaw University of Technology. Mainly interested in deep learning for natural language processing (NLP), recommender systems and graph-based learning. Google Scholar: Wp0wHJoAAAAJ LinkedIn: Barbara-Rychalska Bartek Sobieski MSc student in Data Science at Warsaw University of Technology. Interested in deep learning and hyperparameter optimization. GitHub: sobieskibj LinkedIn: Bartłomiej-Sobieski Mikołaj Spytek MSc student in Data Science at Warsaw University of Technology. Interested in explainable artificial intelligence, data vizualization and survival analysis. Google Scholar: 1u49AqYAAAAJ GitHub: mikolajsp LinkedIn: Mikołaj-Spytek Tomasz Stanisławek PhD candidate in computer science at Warsaw University of Technology. Mainly interested in deep learning for natural language processing (NLP). Google Scholar: gq8NY_UAAAAJ GitHub: tstanislawek LinkedIn: Tomasz-Stanisławek Paulina Tomaszewska PhD candidate in Computer Science at Warsaw University of Technology. Gained experience in AI at leading universities during: Deep Learning Summer School at Tsinghua University (China), one-semester exchange at Nanyang Technological University (Singapore) and research internships at Gwangju Institute of Science and Technology (South Korea) and Institute of Science and Technology (Austria). Mainly interested in Deep Learning, Computer Vision and Transfer Learning. Recently, focused on digital pathology. Google Scholar: eO245iMAAAAJ LinkedIn: paulina-tomaszewska Hoang Thien Ly Bachelor student in Maths and Data Analysis at Warsaw University of Technology. Interested in working with data, and learning explainable artificial intelligence methods. Google Scholar: JkysewYAAAAJ GitHub: lhthien09 LinkedIn: hthienly Piotr Wilczyński BSc student in Data Science at Warsaw University of Technology. Interested in Large Language Models, AI Deception and Natural Language Processing. Currently working on my thesis, which applies Computer Vision to medicine. GitHub: wi1lku LinkedIn: Piotr-Wilczyński Jakub Wiśniewski Research Software Engineer and third year Data Science student at Warsaw University of Technology. Developer of tools for bias detection and fairness. Currently researching responsible applications of deep learning. President of Data Science Science Club at WUT. Google Scholar: _6eQsXMAAAAJ GitHub: jakwisn LinkedIn: jakwisn Emilia Wiśnios Research Software Engineer and student of Machine Learning at Faculty of Mathematics, Informatics and Mechanics, University of Warsaw. Interested in natural language processing and reinforcement learning. GitHub: emiliawisnios LinkedIn: emilia-wisnios Paweł Wojciechowski Graduated with a bachelor’s degree in Data Science from Warsaw University of Technology. Interested in explainable artificial intelligence, computer vision, and active learning. GitHub: p-wojciechowski LinkedIn: wojciechowski-p Katarzyna Woźnica PhD candidate in computer science at Warsaw University of Technology. Graduated in mathematical statistics. Interested in automated machine learning especially in hyperparameter tuning for tabular data. Carrying statistical analysis and predictive modelling for healthcare. Google Scholar: tAQS1gQAAAAJ GitHub: woznicak LinkedIn: woznicak Vladimir Zaigrajew PhD candidate in computer science at Warsaw University of Technology. Interested in deep learning, primarily on images, with a focus on representation learning. GitHub: WolodjaZ LinkedIn: vladimir-zaigrajew Artur Żółkowski BSc student in Data Science at Warsaw University of Technology. Interested in explainable artificial intelligence, computer vision and NLP. GitHub: arturzolkowski LinkedIn: Artur-Żółkowski "],["open-positions.html", "Open Positions", " Open Positions We have an open call for HOMER and xLungs projects for following positions. If you are interested in any of them please send your CV and Motivation Letter to przemyslaw.biecek at pw.edu.pl. We reserve the right to contact only selected candidates. MI2 policy on benefits and obligations We are seeking highly motivated and talented researchers at different career levels to join MI2.AI research team in the exciting field of explainability and red-teaming of AI models. There is a common recruitment policy for all positions. We are looking for people interested in research who want to create methods and tools that have a real impact (not yet another paper). proactive people with initiative. people with proficiency in programming languages commonly used in AI research, at least one out of Python, R or Julia. solid understanding of AI/ML models, including their architecture and training processes. good communication skills and proficiency in academic writing; excellent problem-solving skills. We offer: the opportunity to learn from experienced and motivated researchers at different career levels as well as the opportunity to work with a rich network of international and interdisciplinary collaborators (for example Harvard Medical University, Vienna University, LMU, NTU, KDD Lab, Bari University and many others). a possibility for a tenure-track position after the completion of the postdoc. possibility to present research results at international conferences. access to our supercomputers, among them Eden with 32x A100 cards and 1.5TB of storage, a perfect place for training large DNN models. Salary Salary depend on experience and type of project (applied projects tend to have higher salaries than typical research projects). Average rates range from 50-60 PLN/hour (equivalent to 8-9.5k FTE) for a student with experience, through to 80-85 PLN/hour (equivalent to 12-13k FTE) for an experienced PhD student. The base salary for a post-doc (no obligatory teaching duties) is 120-140k gross per year (this is a base salary, extras depend on activity). Low commitment projects (1/5 FTE) in 2023 Ex 1: Any object can be embedded with an appropriately trained neural network. But can objects be embedded using tree-based models? This project (an ideal candidate for MSc thesis) is about exploring different embedding construction techniques based on random forest or boosting methods. Ex 2: Analysis of champion-challenger model pairs on subsets of observations obtained using data segmentation. The project consists of an extension of the EPP method (Elo for predictive models) based on segmented data (e.g. by k-means or by random splits). In addition to the performance analysis, a PD profile comparison analysis is also performed on the splits. Mid commitment projects (2-2.5/5 FTE) in 2023 Ex 1: Development of techniques for explaining segmentation models for 1d (feature-producing models, e.g. survival models), 2d (image-producing models) or 3d data. (CT, hyperspectral images) Experience and knowledge of classical explanatory techniques (SHAP, LRP, CRP) are desirable. Ex 2: Development of red-teaming techniques for LLaMA 2-type language models. Development of test suites testing the vulnerability of the model to a set of prompting-based attacks. Full-time positions for post-docs in 2023 In addition to the research tasks listed above, responsibilities include mentoring junior staff (students, PhD students) and leading small research groups. Additional information about costs of living in Warsaw From FENS: Warsaw is the capital of Poland, so in principle it is one of the most expensive cities in the country but living in Warsaw is still much cheaper than in the majority of European cities. In general, a single person in Warsaw is able to cover all necessary expenses (flat, media, transport, food) with ca. 5000-5500 PLN per month. A typical cost of renting a 2-room flat (30-40 square meters) varies between 2000-3000 PLN, depending on the part of the city. Electricity and media (TV, Internet) are usually not included in the rent and have to be paid separately. The cost of the lunch in the vicinity of the main campus of WUT is ca. 25-40 PLN, while a monthly ticket for public transport is 110 or 180 PLN depending on the area of validity. Public health insurance is included in the post (this includes also admission to hospitals), however the medicines are paid separately. Research Software Engineer In addition to research positions, we also offer positions for skilled programmers. Required: Background in Computer Science, Mathematics, Statistics or similar. Experience in Computer Vision or NLP models Scope of work: Training of machine learning models for tabular and image data Interpretable solutions for tabular and image data Responsible ML solutions for the healthcare domain "],["contact.html", "Contact", " Contact Our rooms: 44 (DataLab - separate entrance in front of the main entrance) 316 (xLungs) 317 (HOMER) Faculty of Mathematics and Information Science, Warsaw University of Technology, Koszykowa 75, 00-662 Warszawa VAT: PL 5250005834 "],["mi2redteam.html", "MI²RedTeam", " MI²RedTeam MI²RedTeam analyses machine and deep learning predictive models through the lens of AI explainability, fairness, security and human trust. We develop methods and tools for explanatory model analysis and apply them in practice. MI²RedTeam is a group of researchers experienced in XAI who perform a rigorous evaluation of AI solutions in order to improve their transparency and security. We apply state-of-the-art methods and introduce new ones to tailor our analysis to the specific predictive task. We openly collaborate on various topics related to explainable and interpretable machine learning. Feel free to reach out to us with research ideas and development opportunities. We help organizations to better understand the vulnerabilities of their AI systems, and take steps to mitigate them. Our current core research topics of interest include: [ARES] Attack-Resistant Explanations towards Secure AI, i.e. a critical evaluation of the state-of-the-art analysis techniques [xSurvival] Explanatory analysis of machine learning survival models [Large Model Analysis] Explanatory analysis of large models, e.g. transformers Methods and methodologies introduced by our team: Evaluating explanations of vision transformers InteractiveEMA towards human-model interaction in explainable machine learning for tabular data SurvSHAP(t) for time-dependent analysis of machine learning survival models LIMEcraft for human-guided visual explanations of deep neural networks Fooling PD &amp; Manipulating SHAP for stress-testing widely-applied explanation methods Checklist towards responsible deep learning on medical images SAFE for lifting interpretability-performance trade-off via automated feature engineering WildNLP for stress-testing deep learning models in NLP Explanatory Model Analysis towards comprehensive examination of predictive models Tools developed by our team: DALEX, breakDown, auditor &amp; modelStudio for explainable machine learning in R dalex for explainable and fair machine learning in Python survex dedicated to explaining machine learning survival models fairmodels for fairness analysis of machine learning classification models Applications supported by our team: In medicine, we analyzed hundreds of models predicting among others: survival in uveal melanoma eye cancer, survival in sepsis, type of lung cancer, lung cancer risk in screening, lung cancer mortality, COVID-19 mortality, hospital length of stay, progression of Alzheimer’s disease. In credit scoring, we analyzed the transparency, auditability, and explainability of machine learning models. In football analytics, we analyzed expected goal models for performance analysis. … This initiative is generously supported by the following institutions. "],["mi²research.html", "MI²Research", " MI²Research On a mission to responsibly build machine learning predictive models. Responsible and sustainable predictive modelling is still a new and developing area. We are conducting a number of studies in this domain that examine predictive models applied to tabular data, computer vision or natural language processing models. We investigate the stability and robustness of various methods, work on explainability and transparency for simple and complex models. As part of our this effort, we develop open source software packages (usually in R and Python) for model explanatory analysis, publish scientific articles describing new methods or investigating properties of already known methods, and create educational materials, recommendations and examples of application in specific domains. If you want to find out more about what we are working on, check out our seminar, which is always open to those interested in responsible and sustainable data science. "],["papers.html", "Papers", " Papers Consolidated learning: a domain-specific model-free optimization strategy with validation on metaMIMIC benchmarks Katarzyna Woźnica, Mateusz Grzyb, Zuzanna Trafas, Przemysław Biecek Machine Learning (2023) This paper proposes a new formulation of the tuning problem, called consolidated learning, more suited to practical challenges faced by model developers, in which a large number of predictive models are created on similar datasets. We show that a carefully selected static portfolio of hyperparameter configurations yields good results for anytime optimization, while maintaining the ease of use and implementation. We demonstrate the effectiveness of this approach through an empirical study for the XGBoost algorithm and the newly created metaMIMIC benchmarks of predictive tasks extracted from the MIMIC-IV medical database. Towards Evaluating Explanations of Vision Transformers for Medical Imaging Piotr Komorowski, Hubert Baniecki, Przemysław Biecek CVPR Workshop on Explainable AI for Computer Vision (2023) This paper investigates the performance of various interpretation methods on a Vision Transformer (ViT) applied to classify chest X-ray images. We introduce the notion of evaluating faithfulness, sensitivity, and complexity of ViT explanations. The obtained results indicate that Layerwise relevance propagation for transformers outperforms Local interpretable model-agnostic explanations and Attention visualization, providing a more accurate and reliable representation of what a ViT has actually learned. Hospital Length of Stay Prediction Based on Multi-modal Data towards Trustworthy Human-AI Collaboration in Radiomics Hubert Baniecki, Bartlomiej Sobieski, Przemysław Bombiński, Patryk Szatkowski, Przemysław Biecek International Conference on Artificial Intelligence in Medicine (2023) To what extent can the patient’s length of stay in a hospital be predicted using only an X-ray image? We answer this question by comparing the performance of machine learning survival models on a novel multi-modal dataset created from 1235 images with textual radiology reports annotated by humans. We introduce time-dependent model explanations into the human-AI decision making process. For reproducibility, we open-source code and the TLOS dataset at this URL. SurvSHAP(t): Time-dependent explanations of machine learning survival models Mateusz Krzyziński, Mikołaj Spytek, Hubert Baniecki, Przemysław Biecek Knowledge-Based Systems (2023) In this paper, we introduce SurvSHAP(t), the first time-dependent explanation that allows for interpreting survival black-box models. The proposed methods aim to enhance precision diagnostics and support domain experts in making decisions. SurvSHAP(t) is model-agnostic and can be applied to all models with functional output. We provide an accessible implementation of time-dependent explanations in Python at this URL. The grammar of interactive explanatory model analysis Hubert Baniecki, Dariusz Parzych, Przemyslaw Biecek Data Mining and Knowledge Discovery (2023) This paper proposes how different Explanatory Model Analysis (EMA) methods complement each other and discusses why it is essential to juxtapose them. The introduced process of Interactive EMA (IEMA) derives from the algorithmic side of explainable machine learning and aims to embrace ideas developed in cognitive sciences. We formalize the grammar of IEMA to describe human-model interaction. We conduct a user study to evaluate the usefulness of IEMA, which indicates that an interactive sequential analysis of a model may increase the accuracy and confidence of human decision making. Climate Policy Tracker: Pipeline for automated analysis of public climate policies Artur Żółkowski, Mateusz Krzyziński, Piotr Wilczyński, Stanisław Giziński, Emilia Wiśnios, Bartosz Pieliński, Julian Sienkiewicz, Przemysław Biecek NeurIPS Workshop on Tackling Climate Change with Machine Learning (2022) In this work, we use a Latent Dirichlet Allocation-based pipeline for the automatic summarization and analysis of 10-years of national energy and climate plans (NECPs) for the period from 2021 to 2030, established by 27 Member States of the European Union. We focus on analyzing policy framing, the language used to describe specific issues, to detect essential nuances in the way governments frame their climate policies and achieve climate goals. Explainable expected goal models for performance analysis in football analytics Mustafa Cavus, Przemyslaw Biecek International Conference on Data Science and Advanced Analytics (2022) The expected goal provides a more representative measure of the team and player performance which also suit the low-scoring nature of football instead of the score in modern football. This paper proposes an accurate expected goal model trained on 315,430 shots from seven seasons between 2014-15 and 2020-21 of the top-five European football leagues. Moreover, we demonstrate a practical application of aggregated profiles to explain a group of observations on an accurate expected goal model for monitoring the team and player performance. Multi-omics disease module detection with an explainable Greedy Decision Forest Bastian Pfeifer, Hubert Baniecki, Anna Saranti, Przemyslaw Biecek, Andreas Holzinger Scientific Reports (2022) In this work, we demonstrate subnetwork detection based on multi-modal node features using a novel Greedy Decision Forest (GDF) with inherent interpretability. The latter will be a crucial factor to retain experts and gain their trust in such algorithms. To demonstrate a concrete application example, we focus on bioinformatics, systems biology and particularly biomedicine, but the presented methodology is applicable in many other domains as well. Our proposed explainable approach can help to uncover disease-causing network modules from multi-omics data to better understand complex diseases such as cancer. Interpretable meta-score for model performance Alicja Gosiewska, Katarzyna Woźnica, Przemysław Biecek Nature Machine Intelligence (2022) Elo-based predictive power (EPP) meta-score that is built on other performance measures and allows for interpretable comparisons of models. Differences between this score have a probabilistic interpretation and can be compared directly between data sets. Furthermore, this meta-score allows for an assessment of ranking fitness. We prove the properties of the Elo-based predictive power meta-score and support them with empirical results on a large-scale benchmark of 30 classification data sets. Additionally, we propose a unified benchmark ontology that provides a uniform description of benchmarks. fairmodels: a Flexible Tool for Bias Detection, Visualization, and Mitigation in Binary Classification Models Jakub Wiśniewski, Przemyslaw Biecek The R Journal (2022) This article introduces an R package fairmodels that helps to validate fairness and eliminate bias in binary classification models quickly and flexibly. It offers a model-agnostic approach to bias detection, visualization, and mitigation. The implemented functions and fairness metrics enable model fairness validation from different perspectives. In addition, the package includes a series of methods for bias mitigation that aim to diminish the discrimination in the model. The package is designed to examine a single model and facilitate comparisons between multiple models. A robust framework to investigate the reliability and stability of explainable artificial intelligence markers of Mild Cognitive Impairment and Alzheimer’s Disease Angela Lombardi, Domenico Diacono, Nicola Amoroso, Przemysław Biecek, Alfonso Monaco, Loredana Bellantuono, Ester Pantaleo, Giancarlo Logroscino, Roberto De Blasi, Sabina Tangaro, Roberto Bellotti Brain Informatics (2022) In this work, we present a robust framework to (i) perform a threefold classification between healthy control subjects, individuals with cognitive impairment, and subjects with dementia using different cognitive indexes and (ii) analyze the variability of the explainability SHAP values associated with the decisions taken by the predictive models. We demonstrate that the SHAP values can accurately characterize how each index affects a patient’s cognitive status. Furthermore, we show that a longitudinal analysis of SHAP values can provide effective information on Alzheimer’s disease progression. LIMEcraft: handcrafted superpixel selection and inspection for Visual eXplanations Weronika Hryniewska, Adrianna Grudzień, Przemysław Biecek Machine Learning (2022) LIMEcraft enhances the process of explanation by allowing a user to interactively select semantically consistent areas and thoroughly examine the prediction for the image instance in case of many image features. Experiments on several models show that our tool improves model safety by inspecting model fairness for image pieces that may indicate model bias. The code is available at: this URL. Fooling Partial Dependence via Data Poisoning Hubert Baniecki, Wojciech Kretowicz, Przemyslaw Biecek ECML PKDD (2022) We showcase that PD can be manipulated in an adversarial manner, which is alarming, especially in financial or medical applications where auditability became a must-have trait supporting black-box machine learning. The fooling is performed via poisoning the data to bend and shift explanations in the desired direction using genetic and gradient algorithms. Manipulating SHAP via Adversarial Data Perturbations (Student Abstract) Hubert Baniecki, Przemyslaw Biecek AAAI Conference on Artificial Intelligence (2022) We introduce a model-agnostic algorithm for manipulating SHapley Additive exPlanations (SHAP) with perturbation of tabular data. It is evaluated on predictive tasks from healthcare and financial domains to illustrate how crucial is the context of data distribution in interpreting machine learning models. Our method supports checking the stability of the explanations used by various stakeholders apparent in the domain of responsible AI; moreover, the result highlights the explanations’ vulnerability that can be exploited by an adversary. A Signature of 14 Long Non-Coding RNAs (lncRNAs) as a Step towards Precision Diagnosis for NSCLC Anetta Sulewska, Jacek Niklinski, Radoslaw Charkiewicz, Piotr Karabowicz, Przemyslaw Biecek, Hubert Baniecki, Oksana Kowalczuk, Miroslaw Kozlowski, Patrycja Modzelewska, Piotr Majewski et al. Cancers (2022) The aim of the study was the appraisal of the diagnostic value of 14 differentially expressed long non-coding RNAs (lncRNAs) in the early stages of non-small-cell lung cancer (NSCLC). We established two classifiers. The first recognized cancerous from noncancerous tissues, the second successfully discriminated NSCLC subtypes (LUAD vs. LUSC). Our results indicate that the panel of 14 lncRNAs can be a promising tool to support a routine histopathological diagnosis of NSCLC. dalex: Responsible Machine Learning with Interactive Explainability and Fairness in Python Hubert Baniecki, Wojciech Kretowicz, Piotr Piątyszek, Jakub Wiśniewski, Przemyslaw Biecek Journal of Machine Learning Research (2021) We introduce dalex, a Python package which implements a model-agnostic interface for interactive explainability and fairness. It adopts the design crafted through the development of various tools for explainable machine learning; thus, it aims at the unification of existing solutions. This library’s source code and documentation are available under open license at this URL. Checklist for responsible deep learning modeling of medical images based on COVID-19 detection studies Weronika Hryniewska, Przemysław Bombiński, Patryk Szatkowski, Paulina Tomaszewska, Artur Przelaskowski, Przemysław Biecek Pattern Recognition (2021) Our analysis revealed numerous mistakes made at different stages of data acquisition, model development, and explanation construction. In this work, we overview the approaches proposed in the surveyed Machine Learning articles and indicate typical errors emerging from the lack of deep understanding of the radiography domain. The final result is a proposed checklist with the minimum conditions to be met by a reliable COVID-19 diagnostic model. Towards explainable meta-learning Katarzyna Woźnica, Przemyslaw Biecek ECML PKDD Workshop on eXplainable Knowledge Discovery in Data Mining (2021) To build a new generation of meta-models we need a deeper understanding of the importance and effect of meta-features on the model tunability. In this paper, we propose techniques developed for eXplainable Artificial Intelligence (XAI) to examine and extract knowledge from black-box surrogate models. To our knowledge, this is the first paper that shows how post-hoc explainability can be used to improve the meta-learning. Prevention is better than cure: a case study of the abnormalities detection in the chest Weronika Hryniewska, Piotr Czarnecki, Jakub Wiśniewski, Przemysław Bombiński, Przemysław Biecek CVPR Workshop on “Beyond Fairness: Towards a Just, Equitable, and Accountable Computer Vision” (2021) In this paper, we analyze in detail a single use case - a Kaggle competition related to the detection of abnormalities in X-ray lung images. We demonstrate how a series of simple tests for data imbalance exposes faults in the data acquisition and annotation process. Complex models are able to learn such artifacts and it is difficult to remove this bias during or after the training. Simpler is better: Lifting interpretability-performance trade-off via automated feature engineering Alicja Gosiewska, Anna Kozak, Przemysław Biecek Decision Support Systems (2021) We propose a framework that uses elastic black boxes as supervisor models to create simpler, less opaque, yet still accurate and interpretable glass box models. The new models were created using newly engineered features extracted with the help of a supervisor model. We supply the analysis using a large-scale benchmark on several tabular data sets from the OpenML database. The first SARS-CoV-2 genetic variants of concern (VOC) in Poland: The concept of a comprehensive approach to monitoring and surveillance of emerging variants Radosław Charkiewicz, Jacek Nikliński, Przemysław Biecek, Joanna Kiśluk, Sławomir Pancewicz, Anna Moniuszko-Malinowska, Robert Flisiak, Adam Krętowski, Janusz Dzięcioł, Marcin Moniuszko, Rafał Gierczyński, Grzegorz Juszczyk, Joanna Reszeć Advances in Medical Sciences (2021) This study shows the first confirmed case of SARS-CoV-2 in Poland with the lineage B.1.351 (known as 501Y.V2 South African variant), as well as another 18 cases with epidemiologically relevant lineage B.1.1.7, known as British variant. Responsible Prediction Making of COVID-19 Mortality (Student Abstract) Hubert Baniecki, Przemyslaw Biecek AAAI Conference on Artificial Intelligence (2021) During the literature review of COVID-19 related prognosis and diagnosis, we found out that most of the predictive models are not faithful to the RAI principles, which can lead to biassed results and wrong reasoning. To solve this problem, we show how novel XAI techniques boost transparency, reproducibility and quality of models. Models in the Wild: On Corruption Robustness of Neural NLP Systems Barbara Rychalska, Dominika Basaj, Alicja Gosiewska, Przemyslaw Biecek International Conference on Neural Information Processing (2019) In this paper we introduce WildNLP - a framework for testing model stability in a natural setting where text corruptions such as keyboard errors or misspelling occur. We compare robustness of deep learning models from 4 popular NLP tasks: Q&amp;A, NLI, NER and Sentiment Analysis by testing their performance on aspects introduced in the framework. In particular, we focus on a comparison between recent state-of-the-art text representations and non-contextualized word embeddings. In order to improve robustness, we perform adversarial training on selected aspects and check its transferability to the improvement of models with various corruption types. We find that the high performance of models does not ensure sufficient robustness, although modern embedding techniques help to improve it. auditor: an R Package for Model-Agnostic Visual Validation and Diagnostics Alicja Gosiewska, Przemyslaw Biecek The R Journal (2019) This paper describes methodology and tools for model-agnostic auditing. It provides functinos for assessing and comparing the goodness of fit and performance of models. In addition, the package may be used for analysis of the similarity of residuals and for identification of outliers and influential observations. The examination is carried out by diagnostic scores and visual verification. The code presented in this paper are implemented in the auditor package. Its flexible and consistent grammar facilitates the validation models of a large class of models. Explanations of Model Predictions with live and breakDown Packages Mateusz Staniak, Przemyslaw Biecek The R Journal (2018) Complex models are commonly used in predictive modeling. In this paper we present R packages that can be used for explaining predictions from complex black box models and attributing parts of these predictions to input features. We introduce two new approaches and corresponding packages for such attribution, namely live and breakDown. We also compare their results with existing implementations of state-of-the-art solutions, namely, lime (Pedersen and Benesty, 2018) which implements Locally Interpretable Model-agnostic Explanations and iml (Molnar et al., 2018) which implements Shapley values. DALEX: Explainers for Complex Predictive Models in R Przemyslaw Biecek Journal of Machine Learning Research (2018) This paper describes a consistent collection of explainers for predictive models, a.k.a. black boxes. Each explainer is a technique for exploration of a black box model. Presented approaches are model-agnostic, what means that they extract useful information from any predictive method irrespective of its internal structure. Each explainer is linked with a specific aspect of a model. Every explainer presented here works for a single model or for a collection of models. In the latter case, models can be compared against each other. Presented explainers are implemented in the DALEX package for R. They are based on a uniform standardized grammar of model exploration which may be easily extended. archivist: An R Package for Managing, Recording and Restoring Data Analysis Results Przemyslaw Biecek, Marcin Kosiński Journal of Statistical Software (2017) Everything that exists in R is an object (Chambers 2016). This article examines what would be possible if we kept copies of all R objects that have ever been created. Not only objects but also their properties, meta-data, relations with other objects and information about context in which they were created. We introduce archivist, an R package designed to improve the management of results of data analysis. "],["software.html", "Software", " Software DALEX XAI with DALEX for R and Python survex Explainable machine learning in survival analysis Arena Interactive tool for the exploration and comparison of models’ explanations fairmodels Fairness with fairmodels COVID-19 COVID-19 Risk Score archivist Model governance with R (MLOps) "],["books.html", "Books", " Books Explanatory Model Analysis Explore, Explain, and Examine Predictive Models. With examples in R and Python Przemysław Biecek, Tomasz Burzykowski Chapman and Hall/CRC, New York (2021) Odkrywać! Ujawniać! Objaśniać! Zbiór esejów o sztuce prezentowania danych Essays on the art of data visualisation Przemysław Biecek University of Warsaw Press (2016) Analiza danych z programem R. Modele liniowe z efektami stałymi, losowymi i mieszanymi Data analysis in R. Linear models with fixed, random and mixed effects Przemysław Biecek Polish Scientific Publishers PWN (2013) "],["seminars.html", "Seminars", " Seminars We meet every Monday, at 10 am online or in MI2DataLab (room 044, Faculty of Mathematics and Information Science, Warsaw University of Technology). Join us at https://meet.google.com/nno-okiz-bxy (or http://meet.drwhy.ai/) List of topics and materials from past seminars: https://github.com/MI2DataLab/MI2DataLab_Seminarium "],["research-grants.html", "Research grants", " Research grants ARES 2022-2026 ARES: Attack-resistant Explanations toward Secure and trustworthy AI Machine learning explainability, fairness, robustness, and security are key elements of trustworthy Artificial Intelligence, an area of strategic importance. In this context, the main goals of the ARES project are: Develop adversarial attacks on state-of-the-art explanations to investigate vulnerabilities and limitations of the existing explainability and fairness approaches in machine learning. Introduce novel robust explanations that are stable against manipulation and intuitive to evaluate. Achieving the first goal primarily impacts various domains of research, which currently use (and explain) black-box models for knowledge discovery and decision-making, by highlighting vulnerabilities and limitations of their explanations. Achieving the second goal impacts more the broad machine learning domain as it aims at improving state-of-the-art by introducing robust explanations toward secure and trustworthy AI. Work on this project is financially supported by the Polish National Science Centre PRELUDIUM BIS grant 2021/43/O/ST6/00347. DARLING 2022-2024 DARLING: Deep Analysis of Regulations with Language Inference, Network analysis and institutional Grammar Aim of the project Developing the tools for automated analysis of content of legal documents leveraging Natural Language Processing, that will help understand the dynamic of change in public policies and variables influencing those changes. Those tools will be firstly used to analyse the case of development of policy subsystem regulating usage of AI in the European Union. Specific goals of the project Developing and evaluating multilingual models for issue classification for legal and public policy documents. Developing embedding-based topic modeling methods for legal and public policy documents suited for analysis of change of the topics between documents. Institutional grammar based analysis of changes in topics between different public policy documents, regulations and public consultation documents. Agent-based models predicting diffusion of issues in public policy documents. Methodology The core of the DARLING project is the issues and topic analysis in documents connected with regulations development using NLP tools. Issues analysis shall allow tracking how different options of AI operationalisation, ways the AI-connected threats are perceived as well as ideas regarding AI regulations are shared among three different types of texts: scientific, expert and legal ones. The extracted issues will then be subject to complex networks analysis and institutional grammar approach. The network analysis, backed by agent-based modeling, will be used to examine the flow of issues among the documents based on their vector-formed characteristics. On the other hand, the Institutional Grammar (IG) will be used to analyze the modality of issues, e.g., the tendency to regulate a specific aspect of AI in a given issue, its deontic character or its conditionality. In result the DARLING project will effect in the development of new methods to analyze legal documents connected to regulation based on deep text processing and links among the documents. An inter-institutional and interdisciplinary team of computer, political sciences and physics of complex systems scientists will elaborate new machine learning approaches to examine the regulation corpora, issues recognition, issues analysis by the means of IG as well as propose new methods of modeling the flow/changes of regulations based on complex networks tools. X-LUNGS 2021-2024 X-LUNGS: Responsible Artificial Intelligence for Lung Diseases The aim of the project is to support the process of identification of lesions visible on CT and lung x-rays. We intend to achieve this goal by building an information system based on artificial intelligence (AI) that will support the radiologist’s work by enriching the images with additional information. The unique feature of the proposed system is a trustworthy artificial intelligence module that: will reduce the image analysis time needed to detect lesions, will make the image evaluation process more transparent, will provide image and textual explanations indicating the rationale behind the proposed recommendation, will be verified for effective collaboration with the radiologist. Work on this project is financially supported from the INFOSTRATEG-I/0022/2021-00 grant funded by Polish National Centre for Research and Development (NCBiR). HOMER 2020-2025 HOMER: Human Oriented autoMated machinE leaRning One of the biggest challenges in the state-of-the-art machine learning is dealing with the complexity of predictive models. Recent techniques like deep neural networks, gradient boosting or random forests create models with thousands or even millions of parameters. This makes decisions generated by these black-box models completely opaque. Model obscurity undermines trust in model decisions, hampers model debugging, blocks model auditability, exposes models to problems with concept drift or data drift. Recently, there has been a huge progress in the area of model interpretability, which results in the first generation of model explainers, methods for better understanding of factors that drive model decisions. Despite this progress, we are still far from methods that provide deep explanations, confronted with domain knowledge that satisfies our ,,Right to explanation’’ as listed in the General Data Protection Regulation (GDPR). In this project I am going to significantly advance next generation of explainers for predictive models. This will be a disruptive change in the way how machine learning models are created, deployed, and maintained. Currently to much time is spend on handcrafted models produced in a tedious and laborious try-and-error process. The proposed Human-Oriented Machine Learning will focus on the true bottleneck in development of new algorithms, i.e. on model-human interfaces. The particular directions I consider are (1) developing an uniform grammar for visual model exploration, (2) establishing a methodology for contrastive explanations that describe similarities and differences among different models, (3) advancing a methodology for non-additive model explanations, (4) creating new human-model interfaces for effective communication between models and humans, (5) introducing new techniques for training of interpretable models based on elastic surrogate black-box models, (6) rising new methods for automated auditing of fairness, biases and performance of predictive models. Work on this project is financially supported from the SONATA BIS grant 2019/34/E/ST6/00052 funded by Polish National Science Centre (NCN). DeCoviD 2020-2022 DeCoviD: Detection of Covid-19 related markers of pulmonary changes using Deep Neural Networks models supported by eXplainable Artificial Intelligence and Cognitive Compressed Sensing Covid-19 is an infectious respiratory disease. A coronavirus infection leaves permanent ramifications in the respiratory system and beyond. In this situation, tools supporting diagnosis and assessment of lung damage after infection and during Covid-19 treatment are crucial. Preliminary results of analysis of CT images and lung xrays suggest that they can help to quickly assess even asymptomatic cases and facilitate prognosis of response to treatment. There are also reports of usefulness of ultrasound images. The aim of the DeCoviD project is to develop methods and tools to support radiologists in the assessment of lung imaging data for the occurrence of changes caused by Covid-19 disease. The developed solution will allow to automate the identification of pathological changes and will support the diagnosis of coexisting lung diseases as well as diseases of other organs visible on chest images. It will also allow to quantify the severity of lung damage caused by the disease Responsible decision support for radiologists requires models based on interpretable features. Such features will be stored in a hybrid knowledge base powered by two research teams from WUT, working on the basis of two, seemingly opposite, paradigms of image data analysis. The eXplainable Artificial Intelligence (XAI) team will use trained deep networks to automatically extract features that are essential for effective disease detection. Cognitive Compressed Sensing (CCS) will build a set of interpretable semantic features using sparse cognitive representations agreed with a group of cooperating radiologists. Combining these two approaches will achieve high effectiveness of the constructed models, combined with high transparency, clarity and stability of the solution. The DeCoviD project is a part of a broader strategy of competence development in the area of deep learning + XAI + medical applications at the Warsaw University of Technology. More information: https://github.com/MI2DataLab/DeCoviD. Work on this project is financially supported by the IDUB against COVID PW. DALEX 2018-2022 DALEX: Descriptive and model Agnostic Local EXplanations Research project objectives. Black boxes are complex machine learning models, for example deep neural network, an ensemble of trees of high-dimensional regression model. They are commonly used due to they high performance. But how to understand the structure of a black-box, a model in which decision rules are too cryptic for humans? The aim of the project is to create a methodology for such exploration. To address this issue we will develop methods, that: (1) identify key variables that mostly determine a model response, (2) explain a single model response in a compact visual way through local approximations, (3) enrich model diagnostic plots. Research project methodology. This project is divided into three subprojects - local approximations od complex models (called LIVE), explanations of particular model predictions (called EXPLAIN) and conditional explanations (called CONDA). Expected impact on the development of science. Explanations of black boxes have fundamental implications for the field of predictive and statistical modelling. The advent of big data forces imposes usage of black boxes that are easily able to overperform classical methods. But the high performance itself does not imply that the model is appropriate. Thus, especially in applications to personalized medicine or some regulated fields, one should scrutinize decision rules incorporated in the model. New methods and tools for exploration of black-box models are useful for quick identification of problems with the model structure and increase the interpretability of a black-box Work on this project is financially supported from the OPUS grant 2017/27/B/ST6/01307 funded by Polish National Science Centre (NCN). MLGenSig 2017-2021 MLGenSig: Machine Learning Methods for building of Integrated Genetic Signatures Research project objectives. The main scientific goal of this project is to develop a methodology for integrated genetic signatures based on data from divergent high-throughput techniques used in molecular biology. Integrated signatures base on ensembles of signatures for RNA-seq, DNA-seq, data as well for methylation profiles and protein expression microarrays. The advent of high throughput methods allows to measure dozens of thousands or even millions features on different levels like DNA / RNA / protein. And nowadays in many large scale studies scientists use data from mRNA seq to assess the state of transcriptome, protein microarrays to asses the state of proteome and DNA-seq / bisulfide methylation to assess genome / methylome. Research methodology. Genetic signatures are widely used in different applications, among others: for assessing genes that differentiate cells that are chemo resistant vs. cells that are not, assess the stage of cell pluripotency, define molecular cancer subtypes. For example, in database Molecular Signatures Database v5.0 one can find thousands of gene sets - genetic signatures for various conditions. There are signatures that characterize some cancer cells, pluripotent cells and other groups. But they usually contain relatively small number of genes (around 100), results with them are hard to replicate and they are collection of features that were found significant when independently tested. In most cases signatures are derived from measurements of the same type. Like signatures based of expression of transcripts based on data from microarrays or RNA-seq, or methylation profile or DNA variation. We are proposing a very different approach. First we are going to use machine-learning techniques to create large collections of signatures. Such signatures base on ensembles of small sub-signatures, are more robust and usually have higher precision. Then out of such signatures we are going to develop methodology for meta-signatures, that integrate information from different types of data (transcriptome, proteome, genome). Great examples of such studies are: Progenitor Cell Biology Consortium (PCBC) and The Cancer Genome Atlas (TCGA) studies. For thousands of patients in different cohorts (for PCBC cohorts based on stemness phenotype, for TCGA based on cancer type) measurements of both mRNA, miRNA, DNA and methylation profiles are available. New, large datasets require new methods that take into account high and dense structure of dependencies between features. The task that we are going to solve is to develop methodology that will create genetic signatures that integrate information from different levels of cell functioning. Then we are going to use data from TCGA and PBCB project to assess the quality of proposed methodology. As a baseline we are going to use following methodologies: DESeq, edgeR (for mRNA), casper (for lternative splicing), MethylKit (for RRBS data) and RPPanalyzer for protein arrays. Here is the skeleton for our approach: (1) Use ensembles in order to building a genetic signature. The first step would be to use random forests to train a new signature. Ensembles of sub-signtures are build on bootstrap subsamples and they votes if given sample fit given signature or not. (2) In order to improve signatures we are going to consider various normalization of raw counts. We start with log and rank transformation. (3) In order to improve the process of training an ensemble we are going to use pre-filtering of genes. (4) Another approach is to use Bayesian based methods, that may incorporate the expert knowledge, like belief-based gaussian modelling Research project impact. Genetic profiling is more and more important and has number of application starting from basic classification up to personalized medicine in which patients are profiled against different signatures. Existing tools for genetic signatures have many citations. This we assume that the methodology for integrated genetic profiling will be a very useful for many research groups. It is hard to overestimate the impact of better genetic profiling on medicine. Moreover we build a team of people with knowledge in cancer genetic profiling Work on this project is financially supported from the OPUS grant 2016/21/B/ST6/02176 funded by Polish National Science Centre (NCN). "],["mi²solutions.html", "MI²Solutions", " MI²Solutions Hire a team of experienced researchers. The blue team will help you develop good predictive models, create a responsible solution tailored to your needs. The red team will help you find and analyse any weaknesses in your predictive models. It will help you confront them with domain knowledge and make sure they are resilient to future changes in the data. If you need tailor-made solutions for your individual needs, we are happy to help you too. Contact us, we can develop software for you, deploy it, provide training, discuss your needs, verify the quality of your existing solutions. Below you will find a sample offer for trainings or deployments. Research as a service Our team has experience not only in groundbreaking research, but also in deploying these research into business. There are many ways we can help, for example help in delivery of champion-challenger evaluations in which we look for potential to increase the effectiveness of predictive models in your company. take care of the whole life cycle of the predictive models, from reproducibility of results to constant monitoring and continuous improvement of the model. audit models and analyse the sensitivity and vulnerability of the model to incorrect or unexpected behaviours. We would be happy to discuss how we could help with your organisation! Trainings Based on our experience in the area of Responsible Machine Learning, developed a unique two-day hands-on training. Jump into the topic of eXplainable Artificial Intelligence with our trainers. Responsible Machine Learning images/training_xai The training is conducted once a month in small groups online. Small groups encourage questions and the interactions within the group. Language Depending on the group’s preference, the hands-on part can be carried out in R (mlr3 + DALEX) or Python (scikit-learn + dalex). The methodology part does not depend on the language. Book your training To book a training please contact with trainings(at)solutions42.ai. "],["mi²education.html", "MI²Education", " MI²Education The demand for predictive modelling skills is growing at a furious rate. Part of our mission is to develop human capital so that predictive modelling is applied responsibly and safely. We take social responsibility seriously and as part of our activities we support the development of data analysis skills among pupils, students and senior professionals alike. "],["teaching.html", "Teaching", " Teaching Programming R 22/23 Summer Programming and data analysis advanced in R lectures, labs, projects - Anna Kozak Exploratory Data Analysis 22/23 Summer Introduction to exploratory data analysis for Mathematics and data analysis studies lectures, labs, projects - Anna Kozak labs - Hubert Ruczyński, Bartłomiej Sobieski Data Visualization 22/23 Winter Data Visualization Techniques for Data Science studies lectures, labs, projects - Anna Kozak labs, projects - Mateusz Krzyziński, Hubert Ruczyński, Mikołaj Spytek Exploratory Data Analysis 21/22 Summer Introduction to exploratory data analysis for Mathematics and data analysis studies lectures, labs, projects - Anna Kozak labs - Katatzyna Woźnica Interpretable Machine Learning 21/22 Summer Interpretable Machine Learning lectures, projects - Przemysław Biecek Case Studies 21/22 Summer Case Studies for Data Science studies lectures - Weronika Hryniewska ML-1 - labs, projects - Anna Kozak ML-2 - labs, projects - Bartłomiej Eljasiak XAI-tabular - labs, projects - Mustafa Cavus AutoML - labs, projects - Katarzyna Woźnica XIC - labs, projects - Hubert Baniecki TL - labs, projects - Paulina Tomaszewska Data - labs, projects - Weronika Hryniewska NLP - labs, projects - Stanisław Giziński Data Visualization 21/22 Winter Data Visualization Techniques for Data Science studies lectures, labs, projects - Anna Kozak labs, projects - Hubert Baniecki Exploratory Data Analysis 20/21 Summer Introduction to exploratory data analysis for Mathematics and data analysis studies lectures, labs, projects - Anna Kozak labs - Krzysztof Spaliński Case Studies 20/21 Summer Case Studies for Data Science studies lectures - Katarzyna Woźnica XAI1 - labs, projects - Anna Kozak XAI2 - labs, projects - Szymon Maksymiuk DL1 - labs, projects - Weronika Hryniewska DL2 - labs, projects - Paulina Tomaszewska ML - labs, projects - Hubert Baniecki RashomonML - labs, projects - Katarzyna Woźnica Interpretable Machine Learning 20/21 Summer Interpretable Machine Learning for Data Science studies XAI stories 2 lectures, projects - Przemysław Biecek Data Visualization 20/21 Winter Data Visualization Techniques for Data Science studies lectures, labs - Alicja Gosiewska projects - Hubert Baniecki Case Studies 19/20 Summer Case Studies for Data Science studies lectures - Alicja Gosiewska Imputation - labs, projects - Katarzyna Woźnica Reproducibility of scientific papers - labs, projects - Alicja Gosiewska Interpretability - labs, projects - Katarzyna Kobylińska Interpretable Machine Learning 19/20 Summer Interpretable Machine Learning for Data Science studies lectures, projects - Przemysław Biecek Data Visualization 19/20 Summer Data Visualization for Data Science studies lectures, labs, projects - Michał Burdukiewicz "],["beta-bit.html", "Beta Bit", " Beta Bit Chaos Game EN: Are you curious about fractals? The Chaos Game is the book for you. You will learn the mathematical basis behind these figures, find out what algorithm can be used to code them, write code in your favourite programming language (Python, R, Julia?) and also explore the bibliographies of three mathematicians associated with the development of mathematics around these shapes. This is the next book in the Beta Bit series for anyone interested in computational mathematics and data analysis. PL: Jesteś ciekawy czym są fraktale? Gra w Chaos to książka dla Ciebie. Poznasz matematyczne podstawy tych figur, dowiesz się, jaki algorytm można wykorzystać do ich zaprogramowania, napiszesz kod w swoim ulubionym języku programowania (Python, R, Julia?), a także poznasz bibliografie trzech matematyków związanych z rozwojem matematyki wokół tych kształtów. To kolejna książka z serii Beta Bit dla wszystkich zainteresowanych matematyką obliczeniową i analizą danych. Flipbook online [ENG] Flipbook online [POL] Wykresy od kuchni PL: Jak tworzyć dobre wykresy? Dobre, czyli takie, które z przyjemnością się ogląda, z których można wyciągnąć wiele informacji, które są zrozumiałe dla szerokiego odbiorcy, a jednocześnie docenią je smakosze. Na bazie doświadczeń z prowadzenia tych zajęć powstały Wykresy od kuchni. To zbiór krótkich wykładów omawiających różne wątki przydatne w lepszym zrozumieniu tego, jak działa komunikacja z użyciem wykresów statystycznych. Na kolejnych stronach pojawi się wiele analogii do przyrządzania posiłków, ponieważ zarówno w kuchni, jak i w przygotowaniu wykresów statystycznych potrzebna jest praktyka, znajomość pewnych fundamentalnych prawideł, garść sprawdzonych przepisów i dużo zapału do eksperymentowania. Będąc tak uzbrojonym, każdy adept sztuki kulinarnej jest skazany na sukces. Flipbook online [POL] The Hitchhiker’s Guide to Responsible Machine Learning EN: A one-of-a-kind 52-page story about responsible machine learning. Beta and Bit use decision trees, random forests, and AutoML tools to build a risk model after a covid infection, and then use explainable artificial intelligence tools to analyze the behavior of that model. The description of the data analysis process is intertwined with descriptions of ML tools and code snippets. All examples are fully reproducible! PL: Jedyna w swoim rodzaju 52-stronicowa opowieść o odpowiedzialnym uczeniu maszynowym. Beta i Bit używają drzew decyzyjnych, lasów losowych i narzędzi AutoML do budowy modelu ryzyka po zakażeniu covid, a następnie używają narzędzi wyjaśnialnej sztucznej inteligencji by przeanalizować działanie tego modelu. Opis procesu analizy danych przeplata się na opisem kolejnych narzędzi i przykładami kodu. Wszystkie wyniki są całkowicie odtwarzalne! Flipbook online Przemysław Biecek, Anna Kozak, Aleksander Zawada Fundacja Naukowa SmarterPoland.pl. 2022 W pogoni za nieskończonością. Szeregi EN: What does hiking in the mountains have to do with the convergence of series? Quite a lot! We start with the paradoxes related to infinity, but step by step we learn the techniques of geometric series. In this book, the conditions for convergence are explained, together with numerous examples. The comic ends with a collection of exercises with different levels of difficulty. PL: Co wspólnego ma chodzenie po górach ze zbieżnością szeregów? Otóż całkiem sporo! Zaczynamy od paradoksów związanych z nieskończonością, ale krok po kroku poznajemy techniki szeregów geometrycznych. W tej pozycji wyjaśnione są warunki zbieżności wraz z licznymi przykładami. Komiks kończy zbiór zadań o różnych poziomach trudności. Flipbook online Przemysław Biecek, Łukasz Maciejewski, Aleksander Zawada Fundacja Naukowa SmarterPoland.pl. 2022 Przewodnik po pakiecie R EN: The Guide to the R package was the first published Polish book focused on the R language. The current fourth edition consists of four parts: Basics of using R (+tidyverse, shiny, knitr and other goodies), Programming in R (object-oriented, package development, class system), Statistics with R (statistical tests, models, exploration techniques) and Visualization with R (graphics, lattice and ggplot2 packages). PL: Przewodnik po pakiecie R był pierwszą wydaną polskojęzyczną książką poświęconą językowi R. Aktualne czwarte wydanie składa się z czterech części: Podstaw posługiwania się językiem R (+tidyverse, shiny, knitr i inne smaczki), Programowanie w R (obiektowe, tworzenie pakietów, system klas), Statystyka z R (testy statystyczne, modele, techniki eksploracji) i Wizualizacja z R (pakiety graphics, lattice i ggplot2). Wersja online, Książka w księgarnii. Przemysław Biecek Wydawnictwo GiS. 2008-2021 Analiza danych z programem R EN: An academic textbook describing estimation and testing topics for linear models with fixed effects, random effects and mixed effects. The theoretical introduction is complemented by numerous examples for one-way and multivariate ANOVA, one and multiple random components. The examples focus on biological and medical applications and are based on real analyses of real data. PL: Podręcznik akademicki opisujący zagadnienia estymacji i testowania dla modeli liniowych z efektami stałymi, losowymi i mieszanymi. Wprowadzenie teoretyczne jest uzupełnione o liczne przykłady dla jednokierunkowej i wielokierunkowej ANOVA, jednym i wieloma komponentami losowymi. Przykłady dotyczą głównie zastosowań biologicznych i medycznych i bazują na prawdziwych analizach rzeczywistych danych. Książka w księgarnii. Przemysław Biecek Wydawnictwo Naukowe PWN 2013-2018 Eseje o sztuce wizualizacji danych EN: Discover! Reveal! Explain! These three roles can be fulfilled by good statistical graphics. Good means understandable, faithful to the data, aesthetic. How to create such graphics? A collection of essays on the art of displaying data systematises knowledge useful in designing and producing good data visualisations. It is not easy. On the one hand, we can fall into the trap of a colourful mush full of numbers, which is sometimes proudly called infographics. On the other hand, we can fall into the trap of graphics that perfectly reproduce the complexity of numbers, and thus completely incomprehensible. Somewhere in the middle is a graphic that explains, that informs, that is aesthetically pleasing and informative. PL: Odkrywać! Ujawniać! Objaśniać! Te trzy role może spełniać dobra grafika statystyczna. Dobra czyli zrozumiała, wierna danym, estetyczna. Jak tworzyć taką grafikę? Zbiór esejów o sztuce pokazywania danych systematyzuje wiedzę przydatną do projektowania i wykonania dobrej wizualizacji danych. Nie jest to proste. Z jednej strony możemy wpaść w pułapkę pstrokatej papki najeżonej liczbami, którą czasem dumnie nazywa się infografiką. Z drugiej strony wpaść można w pułapkę grafiki idealnie odwzorowującej złożoność liczb a przez to zupełnie niezrozumiałej. Gdzieś po środku jest grafika, która wyjaśnia, która informuje, która jest estetyczna i informatywna. Książka online, Książka w księgarnii. Przemysław Biecek Wydawnictwo SmarterPoland 2008-2021 Pogromcy Danych EN: Data Crunchers is the first MOOC (Massive Open Online Course) developed in Polish for data scientists. Two modules were developed in 2015: the first one is an introduction to R, with loading data, overview of syntax, basic data types, descriptive statistics and pipelined processing. The second module is dedicated to data visualisation and statistical modelling. More than 8,000 people have registered on the Data Crunchers platform. PL: Pogromcy Danych to pierwszy MOOC (Massive Open Online Course) opracowany w języku polskim do analizy danych. W roku 2015 powstały dwa moduły: pierwszy jest wprowadzeniem do programu R, przez wczytywanie danych, omówienie składni, podstawowych typów danych, statystyk opisowych oraz przetwarzania potokowego. Drugi moduł poświęcony jest wizualizacji danych oraz modelowaniu statystycznemu. W platformie Pogromców Danych zarejestrowało się ponad 8000 osób. Przetwarzanie danych w programie R, Wizualizacja i modelowanie, Strona WWW. Przemysław Biecek ICM UW. 2015 Wykresy unplugged EN: Can you create clear charts without any electricity? An illustrated collection of exercises showing eight of the most popular ways to visualise data, with do-it-yourself challenges. Grab your crayons and start creating fantastic charts. PL: Czy można tworzyć czytelne wykresy bez użycia prądu? Ilustrowany zbiór ćwiczeń przedstawiających osiem najpopularniejszych sposobów wizualizacji danych, wraz z zadaniami do samodzielnego wykonania. Weź kredki i zacznij tworzyć fantastyczne wykresy. Flipbook online, Komiks w księgarnii. Przemysław Biecek, Ewa Baranowska, Piotr Sobczyk Fundacja Naukowa SmarterPoland.pl. 2018 W pogoni za nieskończonością EN: Two mathematicians share stories about infinity. In the first Beta attends a lecture on the properties of prime numbers. In the second, Bit breaks into the Palace of Culture and Science. How should we talk about mathematics? PL: Dwójka matematyków wymienia się opowiadaniami o nieskończoności. W pierwszym Beta bierze udział w wykładzie o właściwościach liczb pierwszych. W drugim Bit włamuje się do Pałacu Kultury i Nauki. Jak opowiadać o matematyce? Flipbook online, Komiks w księgarnii. Przemysław Biecek, Łukasz Maciejewski, Tomasz Samojlik, Sebastian Szpakowski Fundacja Naukowa SmarterPoland.pl. 2018 Jak długo żyją Muffinki? EN: A collection of three stories for children showing statistical relationships in the world around us. Beautifully illustrated stories about the distribution of height according to age, the life span of dogs or measuring the weight of trees. PL: Zbiór trzech opowiadań dla dzieci pokazującym zależności statystyczne w świecie wokół nas. Pięknie ilustrowane opowiadania o rozkładzie wzrostu w zależności od wieku, czasie życia psów czy pomiarze wagi drzew. Online: Jak szybko urosnę, Jak długo żyją Muffinki. Przemysław Biecek Fundacja Naukowa SmarterPoland.pl.2016 Pieczara Pietraszki EN: How linear regression can help in getting home, and why it’s not worth hacking into a mad mathematician’s office. A short story describing the adventures of two teenagers Beta and Bit moving around historic Warsaw. PL: W jaki sposób regresja liniowa może pomóc w powrocie do domu, oraz dlaczego nie warto włamywać się do pokoju szalonego matematyka? Lekkie opowiadanie opisujące przygody dwójki nastolatków Bety i Bita w historycznej Warszawie. Online: W jezyku Polskim, In English, По-Русски. Magda Chudzian, Przemysław Biecek Fundacja Naukowa SmarterPoland.pl. 2015 How to weight a dog with a ruler? EN: Workshop materials for children aged 8-10. Kids measure different parameters of their body, such as arm span or height. Then they create a graph summarizing the collected data and look for relations between the measured features. It just so happens that parts of the human body are proportional to each other and you can use a ruler to find this relationship. Part of the StatTub project. PL: Materiały do warsztaty dla dzieci w wieku 8-10. Dzieci mierzą różne parametry swojego ciała, takie jak rozpiętość ramion lub wzrost. Następnie tworzą wykres podsumowujący zebrane dane i szukają zależności pomiędzy zmierzonymi cechami. Tak się składa, że części ciała ludzkiego są do siebie proporcjonalne i można z użyciem linijki znaleźć tę relację. Część projektu StatTuba. Online: English, Polish, Chinese, Simplified Chinese, Czech, German, Spanish, Spanish (Latin America), French, Dutch, Vietnamese. Przemysław Biecek, Klaudia Korniluk Fundacja Naukowa SmarterPoland.pl.2016-2021 "],["responsibleml-blog.html", "ResponsibleML Blog", " ResponsibleML Blog Read more about the research, solutions and education on our blog: Tools for Explainable, Fair and Responsible ML BASIC XAI with DALEX— Part 1: Introduction Anna Kozak In this post, we will take a closer look at some algorithms used in explainable artificial intelligence. You will find here an introduction to methods of global and local model evaluation. Each description will include a technical introduction, example analysis, and code in R and Python. R packages for eXplainable Artificial Intelligence Przemysław Biecek We have prepared an overview of the most popular R-packages, which can be used to build interpretable models or to explore complex ones. Examples of knitr notebooks for more than 30 packages are available at http://xai-tools.drwhy.ai/. Adversarial attacks on Explainable AI Hubert Baniecki There are various adversarial attacks on machine learning models; hence, ways of defending, e.g. by using Explainable AI methods. Nowadays, attacks on model explanations come to light, so does the defense to such adversary. Here, we introduce fundamental concepts related to the domain. A further reference list is available at https://github.com/hbaniecki/adversarial-explainable-ai. "],["404.html", "Page not found", " Page not found The page you requested cannot be found (perhaps it was moved or renamed). You may want to try searching to find the page's new location, or use the table of contents to find the page you are looking for. "]]
diff --git a/docs/seminars.html b/docs/seminars.html
index e18fd93..00fb1a1 100644
--- a/docs/seminars.html
+++ b/docs/seminars.html
@@ -92,10 +92,11 @@
 <li class="chapter" data-level="" data-path="the-team.html"><a href="the-team.html"><i class="fa fa-check"></i>The Team</a></li>
 <li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html"><i class="fa fa-check"></i>Open Positions</a>
 <ul>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#phd-positions-in-2023"><i class="fa fa-check"></i>PhD positions in 2023</a></li>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#deep-learning-engineer"><i class="fa fa-check"></i>Deep Learning Engineer</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#mi2-policy-on-benefits-and-obligations"><i class="fa fa-check"></i>MI2 policy on benefits and obligations</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#low-commitment-projects-15-fte-in-2023"><i class="fa fa-check"></i>Low commitment projects (1/5 FTE) in 2023</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#mid-commitment-projects-2-2.55-fte-in-2023"><i class="fa fa-check"></i>Mid commitment projects (2-2.5/5 FTE) in 2023</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#full-time-positions-for-post-docs-in-2023"><i class="fa fa-check"></i>Full-time positions for post-docs in 2023</a></li>
 <li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#research-software-engineer"><i class="fa fa-check"></i>Research Software Engineer</a></li>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#post-doc"><i class="fa fa-check"></i>Post-doc</a></li>
 </ul></li>
 <li class="chapter" data-level="" data-path="contact.html"><a href="contact.html"><i class="fa fa-check"></i>Contact</a></li>
 </ul></li>
diff --git a/docs/software.html b/docs/software.html
index 861a87f..9339d9b 100644
--- a/docs/software.html
+++ b/docs/software.html
@@ -92,10 +92,11 @@
 <li class="chapter" data-level="" data-path="the-team.html"><a href="the-team.html"><i class="fa fa-check"></i>The Team</a></li>
 <li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html"><i class="fa fa-check"></i>Open Positions</a>
 <ul>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#phd-positions-in-2023"><i class="fa fa-check"></i>PhD positions in 2023</a></li>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#deep-learning-engineer"><i class="fa fa-check"></i>Deep Learning Engineer</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#mi2-policy-on-benefits-and-obligations"><i class="fa fa-check"></i>MI2 policy on benefits and obligations</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#low-commitment-projects-15-fte-in-2023"><i class="fa fa-check"></i>Low commitment projects (1/5 FTE) in 2023</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#mid-commitment-projects-2-2.55-fte-in-2023"><i class="fa fa-check"></i>Mid commitment projects (2-2.5/5 FTE) in 2023</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#full-time-positions-for-post-docs-in-2023"><i class="fa fa-check"></i>Full-time positions for post-docs in 2023</a></li>
 <li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#research-software-engineer"><i class="fa fa-check"></i>Research Software Engineer</a></li>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#post-doc"><i class="fa fa-check"></i>Post-doc</a></li>
 </ul></li>
 <li class="chapter" data-level="" data-path="contact.html"><a href="contact.html"><i class="fa fa-check"></i>Contact</a></li>
 </ul></li>
diff --git a/docs/teaching.html b/docs/teaching.html
index 520b3e9..036091d 100644
--- a/docs/teaching.html
+++ b/docs/teaching.html
@@ -92,10 +92,11 @@
 <li class="chapter" data-level="" data-path="the-team.html"><a href="the-team.html"><i class="fa fa-check"></i>The Team</a></li>
 <li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html"><i class="fa fa-check"></i>Open Positions</a>
 <ul>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#phd-positions-in-2023"><i class="fa fa-check"></i>PhD positions in 2023</a></li>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#deep-learning-engineer"><i class="fa fa-check"></i>Deep Learning Engineer</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#mi2-policy-on-benefits-and-obligations"><i class="fa fa-check"></i>MI2 policy on benefits and obligations</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#low-commitment-projects-15-fte-in-2023"><i class="fa fa-check"></i>Low commitment projects (1/5 FTE) in 2023</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#mid-commitment-projects-2-2.55-fte-in-2023"><i class="fa fa-check"></i>Mid commitment projects (2-2.5/5 FTE) in 2023</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#full-time-positions-for-post-docs-in-2023"><i class="fa fa-check"></i>Full-time positions for post-docs in 2023</a></li>
 <li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#research-software-engineer"><i class="fa fa-check"></i>Research Software Engineer</a></li>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#post-doc"><i class="fa fa-check"></i>Post-doc</a></li>
 </ul></li>
 <li class="chapter" data-level="" data-path="contact.html"><a href="contact.html"><i class="fa fa-check"></i>Contact</a></li>
 </ul></li>
diff --git a/docs/the-team.html b/docs/the-team.html
index c54a59e..f8193af 100644
--- a/docs/the-team.html
+++ b/docs/the-team.html
@@ -92,10 +92,11 @@
 <li class="chapter" data-level="" data-path="the-team.html"><a href="the-team.html"><i class="fa fa-check"></i>The Team</a></li>
 <li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html"><i class="fa fa-check"></i>Open Positions</a>
 <ul>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#phd-positions-in-2023"><i class="fa fa-check"></i>PhD positions in 2023</a></li>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#deep-learning-engineer"><i class="fa fa-check"></i>Deep Learning Engineer</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#mi2-policy-on-benefits-and-obligations"><i class="fa fa-check"></i>MI2 policy on benefits and obligations</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#low-commitment-projects-15-fte-in-2023"><i class="fa fa-check"></i>Low commitment projects (1/5 FTE) in 2023</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#mid-commitment-projects-2-2.55-fte-in-2023"><i class="fa fa-check"></i>Mid commitment projects (2-2.5/5 FTE) in 2023</a></li>
+<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#full-time-positions-for-post-docs-in-2023"><i class="fa fa-check"></i>Full-time positions for post-docs in 2023</a></li>
 <li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#research-software-engineer"><i class="fa fa-check"></i>Research Software Engineer</a></li>
-<li class="chapter" data-level="" data-path="open-positions.html"><a href="open-positions.html#post-doc"><i class="fa fa-check"></i>Post-doc</a></li>
 </ul></li>
 <li class="chapter" data-level="" data-path="contact.html"><a href="contact.html"><i class="fa fa-check"></i>Contact</a></li>
 </ul></li>

From b458087cf2125ee6e2bc36a1f9497c1d7b21bce9 Mon Sep 17 00:00:00 2001
From: Kaczyniec <100872633+Kaczyniec@users.noreply.github.com>
Date: Fri, 13 Oct 2023 17:59:32 +0200
Subject: [PATCH 3/7] =?UTF-8?q?Adding=20Paulina=20Kaczy=C5=84ska=20to=20me?=
 =?UTF-8?q?mbers=20(#102)?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

* add member

adding Paulina Kaczyńska

* add photo

adding photo of Paulina Kaczyńska
---
 "images/paulina_kaczy\305\204ska.jpg" | Bin 0 -> 168530 bytes
 index.Rmd                             |  10 ++++++++++
 2 files changed, 10 insertions(+)
 create mode 100644 "images/paulina_kaczy\305\204ska.jpg"

diff --git "a/images/paulina_kaczy\305\204ska.jpg" "b/images/paulina_kaczy\305\204ska.jpg"
new file mode 100644
index 0000000000000000000000000000000000000000..3d4966642e837a0cef36147917561f68fdacc9fd
GIT binary patch
literal 168530
zcmb@t2Ut^Ex<4EQL_kE4UR97@q<2se5HKJup`-K`TBL=hsGw8<>C$@%NC}}MUAjPM
zfq?X0Lht_L%$%8X&dizT-tW5`*0XnZU2j?MddqM99RK+Pa8pfLRT+SXhX;6$`vd%(
z1SkUVuUz@{kJ|`v|Af~F2?+=YNr;H9Ub{|m{W>WLDJj_v%9~_2C~lCF-lVxnaf^zY
zn)*69EgcON9VHbt)vuS};p3hmAS5OvB&H%GC8PRZZ$G~RD6SC#2m$zbcL7%@@bD?{
zezpNv0RTKgTyKAM_#YeI6<i;$5?v!EA;sN*xCyv|hmU`S0RLC7ad-RSt^)`t2r2If
zJ-m8L>jlwWS1RGxQ5o0R9+fpyYY*<Qi<r9w5|hx-($O>A<KX1t<`ESWmync_R(!0a
ztnx%vO-EM`1lBh&w6L_Ywz0LdhrV?8@bvPAy?OgCDER${km#6Cv2pPUiAkAR**Up+
z`2~gL6_to8WOYq#OKV$uN9VV$?xA7S$mrPk1bTLEeqnKGd1ZACySul4aCmfla{7xd
zJRF_BZGR{1Pjpe>=(<8cfKNd5i!Qt?p16ijK|pv%=qlwyEut5$x9$qRzDD&ZDx<8K
zm`y}`huYk2kc5U^boL(h7ioW^?0-gB;QuGe{z=%s=t2Xo<Ky9khfe`e09?prhL|8M
zZ)ApW0e)*ald78OS3<ac*SMz*((!vlPk(<7cN`D7C*uy_r_y>r_~$0DrpATmF*K+I
zpwjw7!zba&eEqxIweggww14|Hot6^SHLlDM0IprdJ$MCo^Q%9$oXH2oN&;v#+%54C
zeCDTl!(6|fBAH*w3?cdBt$%*>|KD#~2&0{+_V+7Jx&I_ez3LU^ANce~{#@_&Sc3?l
z%ZYIP1N_!>RzIEoBJZD#$1o)4cdY*VD+ni*(IzOTbi&_+@~1r>>W^Ig#pOS9_Loqe
z;N+N%<M6ZlxEI{S_2Jd;KmGe-?#Y76XZT1RJC8veRcXLU_1af7i5eJSdM@tW1d+Js
z`F?-@|MfBVKBjUFU`3_9nNEPy$x6VdO%xiZlsFyxvkLw5xAiY(mg7kLC;td1HGaq1
zUw!n9v)lhq`iWNaSOXaTn|l3SN0^m3)(rnCq5ndvB2MEH0{_9|Uwr<DHgk_(3E`F?
zH+;ZPm0V!=>K}~1Nr)pa`^r|9I|ojIJ=TDgI9+f;tN95h{Y~@za02^+WAtys;s1J%
zx1XGk6X+G3bT!hi>L9ftSrEWIsyf@r2c)&h>Q&Ky8f*U=Fd`nq`I*J9e(PHfPCecK
zDF&Q=0AzYZ(Q%23NdO!T|85O%=g5D1_D_49kr;CNZ2)jfJi9XZyTS6?p!vuCI!+P<
zzrKcR95_zv05Zc+5NSaeAUee6b86)bLgiu&5HE^(Gn%;<e}+0w0{Ej1ofEj%_g;zu
zK54%)3`ek0;*=}#F!y#IuT!p5PQ=F>7$DB3Xrq29T(TWE0H<~)m7l)<GEulbbJP6e
zdw-zz|K^;cZzVuW*|rZL<IY%6lO>A#v{UX<*f8ECPD*_O!%qltUF&S1JHm~Kc$R&O
zX`Hn&_8I!6W4~{8w)oN^Tt&BkQ`0|6K&QUqFM0n<zwq||-IDyBR-Enn`v|$^eop``
z{vEG2E*WPO$WWEZcD8?^n%uwa+*ht_jM|jX1s9-?Gl7R#fJUL-ff0n&D@nUyX#4%B
zi%F+~da!bj&kv}VdP;|?n0;_is!(^FlEc^5<+j%~K@A&W9Jp!b_4=t=P(RITTGOM)
zQ09iC(8@C_qF(aU!SIf<1?<B06aDPY+}5Q2VBi@)ytX5RvB1$hjeXFdAlh!^!5Z{y
zZu)d{f84UtixS4K@9skDMqCXc1$+T8B{R79jWVpzo#YLet|!B%b<9=Ty`uQCwWQdF
zlCuh1IxFl^k*4)&DB43#@eCvq_B2>N2(B2w6ml)`+Gd{%jJ1_iD%Q!VwA3jf0KY48
z(d1rZFyY$}bCWN5Zfz*HB;EsQuSN43rXlYH$6Fx-$VPM>wDKna@f0Pt?k6qR+sc`%
zVi-!Bkvrr_FULbuhpv%4Z-;(a*DRzYegEBknd*cr>V!R^cV=75TD9SD#jDjuPBO3c
zKJ;~03Czf4=N1*#XnyPDL)yB2ZBUN)^O<TWTY$G_w}SLl7`arZ8tlvFz)*b$E6W<i
z)O?p+)`3>iYO$;Zb<2|C;%S&056NN~YS?2$-<_b^<eQtEy*p%j#t=rSS{og2*I_44
za+6rSD!g)y?$`X0{-@#k7jpk!nyeW2>&><|MF{gvUp$R7LBxA{BcMs6LLzRWWJ8g2
z`HkZWjl2nsV{z2x1}?m*b=b{pK1Gc}l>+Kg?$AP*!X(krno_;;%E%(;oL<7ZTUv@b
zZCj}FPe9!INr+P-%LGf#Vf*EQAUghRWGNI&G8nI5tJ$j)KSy08;5%~Ou35oaCC9m`
zez6uX8Gxo<`U%)cHA;=nIJZ~Aj${S&=$)?|TAe8l%8xfzHrbmQ?u}sE$bJI&<I;Wt
zh}(#=LjoA&5*o_o@HcrcW>wJOvzBUVY)YMg@8bDt!0i3tOIYc$N(C!Irm0F0F0;vf
zu{KNF@vMBg&#eEm><j#Wo`<-rl`?6Sc}MwYY5}n7Sga*K?)~7s3}5iAfF7mn)lhUw
zQ)*+S_xifA+4;S}p8!9*nVmS5i-W8QB7}^BR6v8u=cefrAzz(a0dzS(0dJEOwnI$2
zf)O&Tm7GU~nw5e_MPYwG4HrDWhmX?QNbu3g%?#lcaruWx@g@uG>g-Ve^h~z-hE4(_
z?`wNozheEPh54hk!9!{P(_sH==DqTdS@>_x4sM!J#aU_GWS35)rCJ+~0#yD-E2CQb
z@#>nyjb*A1mj(y&wp4+#oy}xT*}`iF+ZuiEa5jS7-r&6+b5{v%+YI%~0P@A_X?CC8
zb(iXmI?`AL+zoP(Wsg2`PCJ3A_XfiK`GHoVnf%Q%{R1(_dexIq$)3a&Kel=DNY8Rb
zHWhY|I3^qV#?)+fCS><`tc}%jhz1B8I9*-%7VgkiFEz`0%hzIT-Y1a*F(~0|*TuyY
zP|my7M11U^cnr<#yw1&>Bsk*+usKbt9*zYnad#xEsHWSRe>;h+^YL4Pe)i+_0Gg=}
zQZFPu-;%z*R3F;T%0MX-j2N~sn5ez!#cG&gbBIwjuSMnB&bfZ14GbXJ)OAEiS{a#;
zKTqzEfGrl=XzDBy?M~E^AwYYYG2j=BPdl9->l`)o0e=+fPb&F~2c3+-7)m+cvkfot
z(ktFRQD?yoT;KZI+Zt;dBjtm)6vLl7EE^-IeSd&WvQPM}u2)2gy&uo4KLcX<8gExq
zC|>50ACw1pqHB9hG&iutCF%EBYa&r7w<^QfDw%5sT6vGD(mwAS=3T(~F&!gPW^(~M
zB^nU@I<;4oNZbI3xZrl&E@7*Rb)z2JX4*d;jmmQO;568HDYDLO=W>>O2NrSgy2Pu+
z!06j++CGJrugpzq%kDsXx(pZV30lv!j{IW~Xyh|I-In$^_aieZMKUu>8))xtXy4=a
z@{X{`@;2dIonBc)|8+iPRR#3A?zu#!N}fUSf<N}Otl%KvmOw-ilRfg*Hy4jxhd2il
zvblX!G(^r>kWz{bBOF#{uCl@t2me9gNAODXw%|$|9$54OI0l6X8-MeMu5b&;sxn6=
zZHudgRWtLlDEeDMYXe$k0?3gm@>!e{XOD$NPDjD-;gfwLN=uy#7Z~{1x`}@>U+SHY
zrp7Uc0Vg($t|z02!(d4M<c1Ym4wO<$CZE42$J%krsJoEET>sUrsl$??o=#1I8JmQ_
zp8yOS{P~O*XrPrxzVDW7MrhDS=VgQ&Lq~!0>_Hsx5h!}s>5V6^L3O-8BobPN6@#z+
zP(3dKNzKP^-FMFMolW<uF18&sne5&ix)+dqFn>_0{N2~F?vXLgyPp8GHqo*{tOHAO
zXMXko!hHNVdV20b$$HET0-|#FNsTrcL8uaQ#UDbi{~H(PmkZe^gzt+WZE9eT)>AKd
zzvA$ut?ygN+&uGo@<?e_hYL)vPk9gj_89FPnYy_=L#=t3kyIqj@tbZC$#|bv{0Fgu
zg{YqZFO~0O7IPO`X_A|!;NmGEn|no;(0vhSvGx3)fW->(j%ad96GzNpaOdb3chZd8
z3)uB%&C;9OAqb~j-#jbU_gjba0|EWTopTPDne|Cl!v(f#x&9frNq8sRu7{GZzeIuB
zOc0aOBpvH(<=?bBd{6?CBC5_S67eipd<ZQuW3VTw*$Xg?lJyQ?RaG!tkex*`4e>fT
zYIVQ$=)qn&0TWyzB+mKNw+&~96mNf91@1A9M(7fGHBkx#&U>zdXy>fZW2wAXH@F*L
z9EiGQOzpcLXSu*5llOlD7*Z8NDPc*whvr3NNuX+ZFi~2Lz~g3dSmZf>#l<{`?YO$=
zD7(+==<_7BZa-&l(kUQ^DL=W==kB>lxQ6V^cT^8HkOnuS#m@P=Vm>36wM0|HPqn%O
zH;&SdjqYy_e6B9YI4*NWaxA)<DO{v)cKiekB@ByiG0nx_E`{!Qhj5kh9R~g~%D4zh
z{*R7*TxQ-$;49|8GV-{fFY^^G1m{%#)ohRFsZ|>t0^s%-(%wqPcmEYa0B|9JgcdGl
zjB#ho$HfzWe0g#{0Q1zEq%L8TWnZS)-XH={DsY^JH5_ZDffGP^jcZzNFC^u+cSUh?
zZ5bWMU_qVD9=AlE<~|%$7BOQxZkn?$n&($4aaxX((VPu8o~sh;2~DH9S~tRZ&vWbc
zDHw0ACw~(p3BT2AEvf$<tZrPS7%T7TS=M`;Y=A=8F?66ktKK!e7dI)&*%546DM6#>
z!HV}a4_xH^1K^Iwv|n8(V@Jf4L@*Se-tMby=8XkZz)Ltku<bp&c77tMXR&Ic>#{X9
zBm{qttb4lRB<3SXtY6qp1#zP#NnqvoCJ&Rf@+`6Mx*j7Mfw>)-G-{+*_wgbZo9%hF
znd%^*zV0P>4(C8~jVO#wE3jXXy{uzOYB0$7adCBoUBoO8L$9w1I!)R)BR3>n|1oT>
z%6?XKE>t=^S{*=kfz+?%AkrZvD~p&hJ_Y*;6xq%4v)D0MMX!X{ElAQIiY3=%sw)<c
z)IUg4iZMjXx@oc-QN@HwRPpxCRy6KH9A$q3q}?tZcbyiK&v_b?^Hp4`hT9d;!HKPB
zK4T9p9IKElzOJ;j@r5=yx>~|f!?vl)u?}S61^nDH*%MvRZ4-tm9^c_Yp3-R>U&m80
zvz;GHl57cK>-9}j1;gG^u_x?7IzB>PI>l<^L~9P2qT6Mmg&ry$6eg;7SDP)W%$nRU
z=2wW<P3`6eLcgrS%2##<YL^}hd<%v?ht1U^pB*Q9&o?m|t}r}16Hg{6WbHXoCITIm
zbV#<3A_~n~?I6i6{QL_-uZCc0IY4;kc#gZE*FG!Ranu2jMJ9PtZ<n1u((`(N^!`9F
z%>IPjUAXc^kebuC&Qi6#0D@Jus<=(}TDw?RHZ^&(<7*{#AZ+uy_@_fIdsOHLtS@0n
z?ad8j7a8caOCE}E7U7M0m((k~62NMs4}`5RzDy7n1)6AC#BX4GmU~`VCs+5`3N$fA
zZNx3YH{!QzOY|><v^IvO{D+p^{pRS3H;fCotZ;+&P3f%hVKRHEvuM0_B0F?9o&IZ~
zbZ^Jn^u#A`icJiYO+nGAaj5e(=UPpt+^kqeD05J}OPu(#4-2!(51a@ghU;4Q*}Kk4
zajqwOEP^z$5-1eWN6ydLEd&<*37~8(Fx^0CTLi}?gpub!`*{n7iT73V(TFc^dw5C$
z(92>=W;g5cbG*{)k)^WwE29hery=Y635zJp@I@JOv)MkI;|lFa?P_VO$_c)NsiJ;|
zXcc2kg2o;Bl*}1I6IJ+~jUj;m!!-?rLzaOO;pT?)yTr3{n<dR$a<SL*r8m|kqT~=8
z){?_pT_Jli%*^P`Z>ZZ<`aV;X-E!4eYvgImP?^~!6WgcY$UBJLxf=Gz%SSp~-{en9
zhSjmdM#g*^{p!~vc$aU@Q}1A+0~{U1k#QyB#U5_H)koJ0Rc1|;I>YZoDt<x1pAMmE
zt_gdi{Ty_-RdB;_Q?OlY20*T}b`AawV`*H_=X>#E<^u>x63VeNUF^m?vafmbVE_Z&
z7M0Vg+KL80GcLOlLH|ZT^@8Ua`B35&t^bzf_@8GFvoQ}yf4Q}zCw4I|&T*OiRUa6N
z)CJIkcNsT??E>A4i+bP<Ky`0dWW)0rQ$Gg>x~YK?!72gb5~4QnWR^eB*@4z36$U0Q
zHFXL<fe%Le_1|ifFtw;%PTGUpwVZq&vw<^DzgQMqK3xcQ3`#iB>X{L7S|dW*htOX-
z)$2;Xoe?+MFRX&H94OnUTu%-_pN@<#$k)S0<tnFZim+XEO$r$JE%rfW3=8A7ZV$)Q
zjimxyPOF?wG{h$)lVQU^53_lYC)hpF_c+nfNzQH^5#*g7ZC3a7EH!1d&IqNw#q2X&
zv1k68f@kySR3}kw=q>M1&~p?XzK}f-j$~dlskcr(^|QUgpJ&Z2v}>L{`E>HZ&4pq5
zN2rdway%R+DB`G1mdZZ(8qdHd9=7%?3P9V;5bS{MuOx*@<r)x$OMUz<(X-7rxR~Vk
zuPo{Bul?=#m)!nKRPwI~MI(1;E1dwB!^Wj5C{utrI7$6InU{@G|65{;7?<fXU|}1*
zr81nCn0#GYM{TRc>D!V&n-}IKqX^Yg28+p|M(eSXG(%MC>UfvHMuafYP^~kqSKz!c
z>K<iUy-3#rn5^J9gj9?M6x!$_yrn`S*XOC^2ERcwm6~bqnYQ9gr3QDs*cu7Ww*6wz
zD^RlF5Ssd$Br@T2<QgKzHD86gh<;kLzWWB@OR&VVV^1LD(YEc&ui=)LcO77^b#=<_
zR#qA0AAC(%-Z&L3IM>FyQWyKOULq`~{7UvThG(?ys-{tr`add=6Nx~WsaTF~C2c(>
zVVAVayOVIi>!JR2jIxJGfjxlTo9VMB%gBfnR}ovFkeeshdOVFO^Y>{#Wvf@7ddr;O
zL08lA100Q&u1r>Xysrz>Z$72u<<vvOw4qy6#5zBJ7k3@G%ZeI<*yJ0nz-;4WT>9uv
zRCVI*yOZ2dXi!Ah3cowi&P8BWCjnSkU?faHobz~$XpZ0g+@Odf(RS&{re>l{%v++N
z3Nb%*?L?@+i5O`sE-<A*$STZM<h78Nj71@3gj&`0Y&fDH*Kt)%OnR9%XAI+i^XM?*
z0(3Z}o0#*96El$~A3szWttkp!EJI#37*S;eQ7gEB?sdM8vWWK9=<#X_(KwhSvFF*C
z|8z1PbTC~i^)4^=Jm+j28YWmE_SFm7QDrtRAss$~DxZv6Y-LjO=$<j$8(=tp!opV{
z*E(SP(pJOi-qRB7#dklR_`q|Z>l{$=OlTu1fV|`Dy_yY8cd>Vl+UjvcdQ7}cqW#N1
zHp3f3v!}e0wO*f3#=J+>aITupWI70X6%OgQURC~)AsukAo)+Kgk>uC#3eVyLz!vUj
zgrxcjxapu{w^D_D{5VI^%C$@WQ-M+!g{g1xy1OP5ys*>8?y8enX3_qUn3@6=M_kY}
zHvuwC!r1seJ<HdzYvroBt?UsKwJ$zcb6z~dIhY+)4rvs|%;Ig8Aoboc=5ME?TU_0g
z5DZIB_VEVL^>zqXk}L$Q|62G7jAYGv#5zdwc1mW1;war%-e~>2rhetM7K7SFDm*Fr
z4mpL6-0HEAR`x!gbmQ&B1WOBSiNUs|hmwaJGh(~S1fo-uMYosyN<OfP|A4*r<K0F!
z({h&BMN@k0UHi~bi$&!on9WJ3A8*|dTk^AT1G8?Voagmh+d2-s<uQ<C^k<}WcGTo#
z)dKTZLh!mKa)hs5cJ$O-?VcUuQG#jS47yC#tZ>Rs{Mq<Q)sLwuCebke?a?xbd7S$D
zM6<qVp|h^!@a%(iV>jbYOTS^vK1PEG@noLfMv@U|@)q6d!=}de0sP(aLc~-tYaRAB
zhF0b|#d%b7%rGv16No>o9>FYt0nQdFQ)4o(hx?#|92I3F5KJn@2|zFNy503CfZSKA
zRPk;}lrx0I%_BS~dAb7e6ToPsZ>!UNq^(2%337}b7#8dIl9;RYeK2%6m~%S>O<Gz3
zWZYGuA6PV~)*n=@cz^tQN1|m^{4~7nexGN)T7ss~YhC$PvF76*H7rZa&UPzWK9OcU
z5id@^r?86Xsb$#MXrK(&!kKK<XoZZxW?LW-nnodeE7JkAlia`~MRzAcDx;h3>*aQi
z)y8OUSwR_gIrw1HLOtPOexEzl>c~ApAZ#G$@}2{ztmRA1e1T-U>!sjMJ&lANgYlF`
z))wO{3BkQStJR~bsNOr#P-0e}rBW#*y*Ep==^#}9G_9qHR!q-|R=h`=;#_$DaH+ww
zb(%X8^j75S=$79Th0(?bJ0m{<A3;C`?I~!S!RPyAC*bU3&vu4-;j2k_GJ@duA8}K)
zmS<Tzm;UJseer8jR1h4sHf_VkEy=NiO{?kc9yc*vw@+T3^D0s<p~Z(OXW3crO!>*6
zlJoH@e<g>0OB(&}_P8p+ZCr?wyhKuT&18frT+Nst6s=ZS_i^TV9I(OuUE5pFys4*h
z>*8wxun{(ySiwAbME_d+<CkKv(;uf}LWL&FHSi5fOZzdSqvr=Dj0Op{Ux?r@lTtd+
z3+@KoA7lF^^b^hxzOF|L&_laQ#`G<Mj!N*2@m-}>pH@VmvX|+^n3JzABa{lB`cCP3
zXq(?B2Yk1DNwYy_`xP8ff3&TCSUQ}F@Sg@w`W42oqu%VD>Q&EQZw#u)i0N4c-$~$i
zcu9!RuJb0M=Ck+Th6U10)rg6@kxewWb8PS$N9mR?rObFfRIL}Dwevp{hmlz3OzFwg
zF*twnE6@uWu5W(on{VBUq`qZ&@gsmF&0Q)%!v1>qna0$xcoL%0>42Jb_KAljS&w-J
z7pNFJOyn;11(p0IWJXN6+Ym1NZXO}gb411IQg&*fL<Nra`BhH1vYGeiq#iMD!v4z=
z*xzbbn<#)&POi?%Uqy-k@qEtTGJ}7wEdWtY+0E%!MPE~ut?;%r@_3c2ufMUd_k2C|
zvh?xs>(X#q<pf^psK!m-=6TIqVpS{kn7t;~(ONrY^yd*yc-Vy@W8xNBrCVXOf^0FS
zsNeU=;c3e0N|i+aGvul9_kym&KJ__%7(@d2vCU9@J?8zGY!dPZoZ43*53!U^oEYyL
zv79n}Fi5q~Z=n7ORfQZnXR}%UViRj|F>bvW*<fzE7C!d`GP%)><qNW7WS{?zXW-c<
z7>2M}jKh%Yt+Z4vR|hKv_k{6jwbQeCK9?B-3+KAE@5L26-3?kBpX#KdDT5MbmI_NW
zAzpaRbi~KS<)T|1)^x5!b<*`ujVO?<6%;#=6+JZo(d^Dzgjt}j51%XKX4J-mV>Nqt
z`D;~C^Ept=Nc5AH&~Pzt26?86+V3wz9SpW4OunT;rizIchl-5t!3_0XJTx>PZR`uq
zdKGf?QMr3fZtNKLtR<<1(u~5*0+kn6I!Owt@JY9)8?Lo08L;@o#MDtvWB8YRx)$vO
z%oYzHZi1RB{d;F@Khk<A)P+ZCx*7~_+ZxVUR4L34*yni5KE>~1qnx+j+>k3Oq!?P}
zTlUaT*7mbsGwFM4R_{Gme>SHQ#uh<aMk)H>GoO&#<}p~c)GG`4eZM^Yni*rns=Now
zTs^NA_GN8JrbRhYQ8w5!!QZB{tW<{f!!zR#YcMZkzCxc(TUDB5SGZ|2V<)>%Wg34m
zFdB?HEX|a^<s~o2fCg$Tg6J&J6q;_lz5aB^%G<_udY%(B+cV$th*(0Dd=`cb5XCOk
zcU!()eD(4?66*=h<B3={y<3>K6)&q);5%qfD|{`chnJ2sedEDF3HZ!a&+;RsrTLR7
zw`U&3X1BaPlyZbUA8=)Rq36GLNevR~#iNUhJ&uYmKlO8%4wrwj=5-^lGG)-3x*B_L
zsGUBV`Hse_6DjmgN%{1Oio+qZ426MmlZXIbfbW=8zM0V~`l*AYk?pmv=4OS5rjU8(
z_d`=g*Br4{5M&QUhj)qe$S1vt)n~cW!8!J((y9oq{RVNpsZJLAl+LM3YJwQ`IS5RS
zWJvwP?GK7i!L08&M{CI6%2U)L15E;S&3NpB;~iL@@T5(5NGBbn2<$Sm5WP{a0w-Bd
z?dG2(V@sYuHK%QWU^)Ta{`J->ae1cU%n)5t&Yj0n9JNZ=`|nt60FGyQ>E}g(L;MTX
z5mHV$t@3UuYV$+2biEREZ^Js8rZXDfD$8kk4DHTq)nAcPwQ;Nnc<0<Y%UD+=87PEQ
zigy)eu{>SM8ayC{$WuFbv4yK(4>)2BvHmHuDKzuOW2KD)HrHB;!;dDJ0VO^uBuiU0
zbNsYlPf4fN#%>nh<XEmv*lhQ5kK7c?@KwX&+sZRG)|J`Cle)6sgYGXiY8_*-Sw_c$
zrNyqY1sTl(D}0#*cknYD{pL4^3eF{m4w~W3MVh2&Qt{~`-nZIB567fR>_DS~G8P&=
zNl%Xw>pdFjvBN7Ycd*Q(aO!4TmcEQTo@uph%dgFV=hVY7nmbeFa-U$jM{Y7Fc1I=;
z(D3!o;NEHWV@<1Kufo<2bo-qCO3bybk~=fe)~1CTUaX3y>$5)r@8b-E!(^ont@mad
z-*iXX&0Y!H5gAQhTi{*YN*ztJ4Ay=7!BRN(L_I1DtZ8EdDKOmjR(wT_j}%+-=Nons
zle17mR(S+n8?674mkWcMq@-q%eHL2%vJB-OH=>`5K|ZqJ^wVJNOr<D8IohcESb4nI
zdrCBQNWHiGUY-@+J{HeJ$v?DHVX>X2cA6>U*?^=8@wldc627Jqkn{XFa9Z%j2~@?t
zV#-)v$7MyKqq<OgX=*!9uU-FS54a=O@v6YJ$bQwvI<)>F$yQ2aZNyBtyF9~jN}i&h
zgH1N^0ev<kP*v61Vv8rT&H2ij!p<jjpQU9@v|H|@4ZH_FS9pzD{HZV7E62q29djd=
zx+Pv8Sd&oC&`&TdN6UOBc@KvWNNKEtp`-N*Sh>sZ5P^4fU5pT`>-hMqA^RBe4hFx4
z45Cxf&SNGDGg2>`xz3<{##f_7K2Seit8M2w0ybFTE9miDLsoMdMasOAJSc<ce4k6N
z@FZSQ9)z$xts<s;zyxu`r9wu+bL(fC8%3&6;_T29=ZhAVgObbX1CD;W`^--YZzF+|
z%G>7LC;+AslBf0H`hSI!{CibjwKmX+u>gnQs6HyZS@-U-KELRq56Vr<(4@=x%aajp
zRq^44===eGI}4$hZaIg$yUP&$>NJSC8F1T(-FhM|-<W#axFZK@nDnlHZApx~=#izR
z888*mZ=dM6H&i9ajv4xLAkIT~iki}r+<Xyidaw*z+D=Q_Cwo!Uf3PaP-S_OtEQ0&$
zRy_*jNy;Xm0mw&)lZs27qG*O;03Ih|Pib8JKxd3P&zWPixtLe#N&#Llb%6uNXb%(=
zHt{7K6#X!4i#mnsb^JCN!kW0Ma}VJEsm)%IWAw0yqR(Kkso`lfDCxdHb}Tz&_@Gd2
z5?2gvXxGRmJ~|s(7<s{A51c-Du3Ej>Z*O*A57JZD$nCx1>Bl9+L59qcfNDEY)f8_-
zUj#j36zyvrjaH`WEBXl-E1ZE+uv<KF$&$(w;nlsK5l^47`h_+3la`XV$`>kJ3I>NS
zdYbqB_b>*6N*&*3jlc8<f5D;Lsc=GE`}cz5|HyaDtghajKWi1cBUz%;k;)#bIJ}gk
zLXwryTo_VUChS|doaAedHTLki>#0*|_0D$?nlE4K!@m)HS@Xt_mEs&C@g5Pw^wc0b
znc&Swnu_^y{V`AXm8=FBQ<X{zx5-rR&ej`_qB*ugQ;Dq?tan*L?8hSNAC0-?ra5_B
zbtKSGCx?xq1fTD01)U9SV?2c;-b}6=2c?;JH=#XmpOlMlCl~sFRhk(;4@}2%myvz_
z{&MFL&(j%q6&+LPJWzRe7xlIHC%~cJ3S_kw0(-@q8GW=lHsI1f)VLY)W-Dl)8q|$(
za#v0yp-X3%ad#H^Lft9_mAjG9u0&g}kRETud(!|iOUfSBh;b;vzUvjg-x+X)pwgT`
zeKNp>m(2hbQ1<e15Jmacitk2)uFoh(_#DBAEJpq!v2tI=oU`I50M?);iqad9Mre4u
zeMEOHWXJTe4frVVa~=^P)NSZupO&*uMb5(<<C<Q0sw^J4s63J@%q@G#jy#j!=BDyp
zlkou?o4a+bzHmCot#)<3VG3~3v_3io-DsFo_hXQ{Kf9%gWqGrz>pnoCm_0$y{_>6a
zOP%4#byH{W!h8K`0`Qj$+yil?Gt?2hHw!xBWRidEY-NASwmj}JPAs7?1Rfndkj;~#
zcylh6wbR**WH@XftGw2{z}gUB>9BfG5}-D?<*Qm`)Xiqco{&#*dX{E?TG4Ds82?)I
zAp_oF7f~OFl=XgTYKF4@2h4^azy0@?q4KHmp-M()vhntuiF%iJ^hqu?7k&<LShG+X
zoGQtvlQGx&W}s|*V5#<+v|cI25FO3XrJ5QxAHnAi(Mc~d+Sp$4CBa)Q8~ycz3rdS~
zfZE|`d2u*c8>F=E4c7c**w*3)>|e%wi=3246o~CJ0K?q6r(d3VvChDj_iN&YsL_p<
z8#AKe=PF}zt_Ent6!N+@8Zu^8broO6Z#}MxEn^}$3t~VNis%`*Io;0Wc7pB5eo_{3
zw_T#;k`4he5GWn%4JS|1YFs%TJqoIn6m24NfZR>Yy;&U2y{2C^vkvx^$ral#dyGxH
zD~b`3@Yok??_L;ORW>dwL!tWw;OxQ_i$<)m7EHd4BSrY3pljE7(pnHPMGntwt9lEc
z$%N-;<+F@jeeKHkv2u8UKFERy4omIRc{jj~a_7<C6zzMo_QaGfR?D)(tia3|tdh4C
zBq%7QL~qAX6Uw+!JS*o(u2wWBNI&W2F|Y}KXW>^Jf-#Ca4cS>6h+Xs1&2qQ&l#UPO
z^X0X!Fne0~6VQCsTWn9WGxEcb?_<<HMnF;#t-{4m&wlk|YOJOyR&VURJF!#A(vzAt
z2R0`19CN!`czBRa2%a1|hTW|nkBrkI<VWnwLpdH@$F0#^P|c6CnuAWHD$6JuLnZx{
zL_npHcSv-bbYfK=KSukt|AcT)BD-TOPZodJQvE5!ZRvE?raD#qjR@f`{&OZ=0W;*3
zeD+4Ne9r7-1IOw8a*2f18$?rVOlK(Vc#-eu%yQJ8iF!w8I}2gj6BNBR-NG%SM`<F|
zoJ?bIU&3;SaHAGwNeMD(8;sOOj#sHz<m`0fEXIC_Oanw1r%zZJ;9c{r6n1RnQue6g
zA?1W&$C<V$v$D|1P~E4E-bLG%^@2@?cPGil^+Ghd@4nH`yWmUVj08cCK#r?xK0ST2
z&13#ST_#JA;0G}S3IU26T>?}=(Ei>qfA>i*R!wzHGW~}PANO*Vx7IYECiQs?_KOz`
zdPd7B6S4Z04>eW9+}68MbI8<}#%-6z-dleDu`0R<{(`%c7vn+SzOES6fi9Ens|F{H
zicP)VJ56CkRNDqclE!3Zm{TG@zqr757+GslFELb@HMI3Gba_fesC)5g_E2X^wn|;@
zAp_KoHlaQG&no?&TX)brYl<AIA8Wr4@fLt|X9PsOKz7lcj*??W82L3L-AP%m9xLPQ
z1V~k|^yo?`Y1GYug8l%t>7Ac|HaS+sL;}j60GkYYMw3MH?@NhiioSQG#j~zvIVL}^
zf2F#3&>E>@DwN|D2qXhtMIx);MT|vQ(@$=iLewHe5S4QJIy!kC14v&|EdHZ0YT8W@
z5{QX=6ed@+LUvWTw*391ni;+%yLFQY+ZY1F;Q-N8s||m7Xg%H=S}V>R3HLE;H49@A
zKtxCZj1r=%$~!q){9cKDFO-@1Jd_E4IZ#*ZPE7Ug(@k;@xRuL#)VET~9feL(6GJhe
zs%f_tTL1lp!<cA;>9Ec#IYgTq=TX!<Td3&hQVLaKPWnR}PBZfr{XgK&e;tkElIq0O
z+D=*S;t4k}Pch1CPHX1OwORaCk)l2*N-<%h4N5U^G_bN<Og9NQf_P7<8aV2^%W4?y
z&GUATWRUlxUhSvTQRe;21K65<OC#<$+i=M=<BsV;@7gUVxE~IpwCv12Gn7EZVUFQt
z(}i?>AI(DcBz2*(<qZz=IXLOz;v<|x_1kUu{^za-rAniTh7zTsSen@;0v%$e8}PP7
zrqXr2I|CNlYtfM8h*%K4_(8S-!hw7K-4UG5Dcr(P2+n06l?)PxFUYvQ24y6>S!5g)
zj~m%9i$TnwH?F-H9x9-;?YvVYgcMU^rxtbdZQ2-SuV=^5HmDUBk1rMSy#`aWLv|0A
zQo02=<dhi0AEygQ_4{!df4Tf*+7>6F-7$VP6;P))WO5;yGwreT`NQJlvjS2H;{_jG
zzDcjO%2aC<x`9VoqtpG7<z$Dyp~M-APpfXUL-B~EVYPm<-_BFlc(dKgC+tE72cqhg
zbIEhc#(5<GH9rZ6^q<S?1vq>pE<N=}i0p6f<u9}+4r;yf_ku+7uY3Qx(*AGPaw`b)
zbhV!mctQz7BOA=b{J*ovhG4+zss=gG@VX+5uSro!Opd0MUoirk3KxYPeGot-ckJB-
z$w>HsomfccY;N2Qu<RY$FJhE@yVug9dP8|{U=#sae3xymOO`UA*1yN7+u+@}?{Z#N
z>Zl{evAfykT1|fEq@iKGiq}L#f|pGKjLK}R=OkTd^Az_UC@ZmO_h*Zv1F^HN3SWG0
z_z57X$hWV$R8I&z(JJdnn|raosV~p^0g<A;XHd-;EjoGSkZe<bBz9l<OC(Pn$57>o
z=%m-Ef$RjzVoBCJxS*s+X(%&f`u^H=*EC^eStnR4qRp`|+jpRqC4gak4DopAT;WR!
z4%{Ym(5J*@`WThQr{UagIDi%-=O-3^GD%c^IwB*gZ~p9-Id@m5r2Ij6&Xiy6uFnKM
z4Vz`aJUF|Fmo@(Tjmsv(LyJP$MAIYdzR~;`^tLD7+=<Q<m^*(h)Y;xD7U%S6C~rmJ
z?4CA6&#8o+<m=-Ga{IY+s{(cL9vSy&qtM7WG5qv{qE91|Q@V3&sKxtYq(!Cl4&*iz
zKLH>7$}oJdF6N|3y0KJ|1@;b(G01TU1-M76*<uU}_mgj7X3?=?vD}^Pt+A5xD@s2$
z`%zO5EcR=h4Vcq$TvDc+Q`wMPUosPe#clbCdf(q^vKVT-bu7(%qeD((^^tKGLoC@j
z>AHyqqJ&SV%j>-93EPI>DspUa>%Jf{L*V>b^(Dgs0{BcJ#r>shwXXR+x%kn>NHR6i
z!9A~#q2jI6DEp*xyRx2qli29h*Y`_HbD(I)XydrrR1RUGIoX4sfOU{>d|&Cb_mLvW
z4Yn{3Gc5euPk@^X<59$YmFIFq4bZtVUQ6fTmQ@d+|FHhV<w9_MqDPQvD7dnRPSj7d
zvM)KZVac6yRa1Wsq2=yEQgar9S)+qkjtXcdkz|LzyIf%zS?bu0?O!$I=zJ4Cos+!=
zF3+2otBrk*4BH>C-#W({37c_5S)Vc5r=Luob)NIIMU#3#ATOSkwbF%gtu*G=q$rEb
z0aYC{T;dLK9X67c7;Wg46JxgY)?2G84xXque)CKMPBVU2GAuRR1ejddxjgWe(UsxG
zj9O;7k_8FG<m2r&*ND8CRpZW(b1>V}s;bc12Y7b0RX!wMg@a+vcU}~9YY#)SMtmJb
zt4$eo=;P^SYVOv@w6WZd*RJ(RPE49}-hSf0JPR&mKO-v;xiVJlJs_jxiXP2z&TS2C
zVGo!^nrfbtG>Y8P8?e~e`BbA_i$XG{JQLg4BN(2gH=v7JLY}czpU$N_w4<o~yR!w|
zx28LX%2B3f-SJ4l*smO7bh-oSCB+7vnSSq2%nh7ws)aOEteg58#mGspm;f;>Ta&6g
zCk{X{{w?Nfw71j-D!#w+85>lnC<YyE!cC*thu?!=4!PsXJTl<Ea67&jBgCjhkg9Cu
zk$l2IX@UGTyY2LI1uXTPvMx~Q33JnXJ|C}hs0qrs*A--^q7#j^j&MROH}T|>BYeXo
zc}P4;)pz)Aj;|gLHeocUtDW9Voq@UiYG?W9BzKd5o${^pr(n(50|^f|nIayk-PnK+
z!N=enI^I1NJvC&=hoNH_u~?tF?@FJe#=<Yqs|f0Xw+VGna2o}$>=C+5bEPxs8oH(u
zU|}Hbl+2Nsr>wSFAOAWcD)ONb3)7JH5N4bvT9q9mI^?(PT>4_uP_8$2!yN`|X0F$%
zX1|SQM8QTM`*THaVbepG3|MVnd@ZVNi<+=2+6&dUU>?{x9s4xl)o)ajSF_t*74D>F
z_=Qdw3Ej`Ix9hWpi6d)wwJ4uNi;K_a62WQFTn)&;aW#;b49s$QOg{GY5AH=&<5<3C
z9GKBU2SpBB+fNhQB938H`vLt-uA|?85ZA8iB*sEplT5b4A$DRFePkhDI{dLmV!Nre
zwH$BqyVy<XE3JP5QlmLjkn43~hhy*evBidK_hqf_lF)UJ15W1}eX*UdAx~X<sDSx2
z9c)Ne?ty|*`G&cZs1>c~Pu`YZa${}l4X_g`2QY(WRNx02i1h~&?zQ`$W41#MHAyFo
z*F9h_M|;irQjN>o6DB>60`F5V$=?kmCnydV5lC@=eHm#%Syb{sQozmU6fBhIm5IKT
zZRnc_H{#{1DB~P=66@Q93wmuFda8dq<5fY7?gQl|ztpC@rx!7SaN~f;*VBd?Xj@@z
z#ETaxn~vEEFv%mTbIH>1b0REkTPfv=SuaFn&#3=hJj=+IVL;o0QlPH*5V}`{>CHlC
z*-=R~N?V0l4I;8eCebD|Z@I_C94U*dT60HRT@R!4VZfRZ9q$h7g^2F4xnit&9-K0b
z<*vHrpJZyrHtrDJ2Bike6~7eKeHO<XhfA9z7hE%R`jr~!#$nO^1p@dT`0RG(FF@MO
zU(QjCyWU%Vjx`A#PaPZ@jgKa#mzpUtJ?@NTh94AFsLD|IWF%ABT=kw#RKbv{o@F$m
zg5vrv$h|EYh8pfN7=w}wva71?8iN)56ii{D*!{fGn!5GKhHa2)O8yh0h8I8%hex~G
z7&5DhtJ$1R+5CF~yqwxPh+w#bM2~@CjPJWTlIUm^YB)rHhNFJC4OpXpI+nd&7Rmnw
znx?iVc#|6>8%9DLwTvzjOz`gKxfbuKVGq=6>F+Ki7Bi*kJd<BHm~Q6}SGT)z&}aK~
zhGT0a%K5glS&r?4zF{KjbU0K^q{g++R*{YRsJQx`vs%sdWc%K5u4GSG8B@>VzF67~
z+6IHL^Xhh)4{@WwRx#l(dIMsH3N<=53++O2vU7E==$r_RpMZ8f*=Eh~M#pCy8a4fR
z?Xu#Fov>1-w03l?ettjk>9e~n%QaV{vPswM1x3QXH+k|ufV*i5CoYALPd*h~$!Yuq
zIZ?@91*Jh{gdsEGN-?T+&X`<F6xmF{={sD-DGAn!)E?%Z%*Xq4g$cJ5E^UbfpvgD9
z+W0&VCr*i8s<r=EgZkTQwSN|^{){I1H}uHwkR~iV#o@0sbM&L$S=%F39rAia8b`FE
z?K-usrgm7-f}N2L&KI$^Cvi^M;CIH#kc}5e>$YNi5g0O{1&PBouu^+9?p~Vm-)47`
zliV)>jarWiB+c&Lozhk~Qh^<0u;^KQD-f_V9u9UC_j-tsQ*v$O!`?!-45qj3krnz8
zx})EQI$LW=<3&4ki7`KPI^gE)`t;Z(v(o9CmT_+{_TF0Q_)7<m7ouB@PI5TZmQ%JS
z*~FwBmOS28(@hgLFH7vq(J)sd#`yl|UX8GmoL`yI<?TjLV{M|RjN-@Fb*Urmb<(-}
z6#}bGjrAhjy%Z@~!|u<$CXJ-4^5cp)dp<L<5IXcpugEiFQc8XT>TjQpk>y6+YeQ;B
zCPr-efb;0MI(v6>I2IS2=xJIcXSJ`6^!u%o>Wwgn>3=%!M5q?p^G6#K%8IaB;8IxV
zDbmmnTc1C^$=w>MH;MWbiwWH>xL401S138vx5E3mel#0V8KYxqRw?7{u@g}!XW!mF
z7Smv9{6XFtGY$iZC&B0X9Aus#kgE}Cm$kb>VxQA^<ihyrnxP_K#hrzts>a7gNv9{T
zx-M_dco$95$Y_$<#w9_b#y{qw^jN)cG5Fc2|7$Dq;FnQ8(uKggOj2u&uRx9xP-Eb|
zk(Pz(ZoG4`k$0KtHL1%tc#xHo<{KiigR$mdPT&5SZ)5l+YZYE^bd4NtPd^3*MR}9$
z9~St!&T(7n4U3+rMnza!?jIG1`dr&j7Nron7jG+vgI#PENYJgg4CZ-r^w8A8^zS5;
z7^rlmE!ZEXFu6`54yHl(PClL{<!>H-c;&%CcoHFlOfOi7Fac%-YL!n5wcH3s*Bkf@
zm<*)9`=qkE4xKYm_vUA=z1ZH$@$>+dt$r*n#%Fif#b4hxIF%~~9cDZu(46bU>M<J`
z+C?l*-F*vD!p$C~>-(if<SMbo*fQ$UGK;pP{awfM3a!J$APs5MDX<=<ky_*Wtv8wl
zuAk(W%-6C6-m~&I)JP^mSbDu_i?Npoe_<~12p%|}PTgRS%`AcHWbh1?b7EOTI7h6`
z#G=?)<J}HfP*tPzs#QKc9r~kUdcxoQ^&2|jR^kQO$20?*hQ&X+b-cp8p_`G0dv$P!
zXUE|MU82eJ!{+_tH$OH`9F)elnOn(O`iaw$_Sj~cG59_;>M5<TeI4uR7>$i*PqXjY
zxl9_PsNWMBNNjE5YenaoT$#o^yN3Mou@TcrN8Lacs={=;qdoB<Ndv9ekA6@2)D$rS
zLj<@!eE#)A6+L|U5%h|*?3iIyN2rj#QDMam7e9GK@{b0g66-iUa~h#oV;6@B*;}(d
zCW4xAhvyq~K6R;H=!S3yCFAbO6AfU!!bQ&f!1Wa_lAK~Q^sUVWBBv0)c0WcB^TBfw
zyPOAwQP+KIE^!6<o=ZyGrBk1u0IF$T6&$Ld^>znsI%}-caZ`IB#($W8kMq+B%`ZIr
zPe8z`X}e_IhL>KEorm=9&Xj!z8c=-pn53a>YAnY~qeV`hx~9hUeaHCg6{E3LQ_mj*
z<i#HAo-VzTGotX33Y<r=&&d0B@(@>1b_LviyH9XBmdBo1>|GWwU%Z)7V^X)cIGL-d
zU^qGG4!G2}CJB$1S^*rC%9M><&2YUd`bvrQbi#-A1-JyEdZ51;*zOfiv1JM{T`Dg_
zR&x33VvUXicjXv{X0yITQmXCj?#7(6q2r%?pI&Hng>5`la2o`ZWKHSD9o{b~zFmH8
zID^H#lgL_z95#YT-3INY4;?8wrPVqFoAj}<<gKb*kXA@k`WcdPS;gCWy)b#pHSg|W
z%2FLAT9a2DD7h5tR#ag}y5hI~DF{8>D(-SpMx2Mk#)8h7kVT-9m^qr-1fbLv*%uW<
z<x#cc==$MJ1gXU<-ETq*Qr!~Y4=#ASWx=F&8;9cg1+I=k%HQri=y>pL+cbF2`Ibsv
zN8mlXg_sL1e1H$hUK*8|>3SDm*&A6`u6yZEy+f4m`>ltR*K448r&aoz81~aDy^_So
zuY*7GPsPLxx?BxV!^)}6k!fS$8}L~NTMblXk4$5>bz}2!#*f)K{;M5$qM&|kC5O+u
z6Frp&uKfzv$HEnNK&_90eERI?TKcDjc1_kQ=$;Rp3FY<TR*&(4Q#OnjVD^P?HQ1EP
zQ2U4POWyIcb9GaLmrusZj^&>FP(b{g#nTv>*Vs+E;Ts}D^^vAn)0EtjK7~0*K{)Y*
zgH0CDj)o}v-S}l}x4y4syb@%u{wP0>#lF|f^M@2`1n=WmPRlK|(EOUz;PL5OJKdAg
zz#J3tk$#%%6rB|e`xwD?hNUP<1iW`No;NM-RD5@mB|GFK^v6Iu(5(IX8_gZ3viu^8
zTP<@!+ee2XXA)ma3&c}#t5h@`Bqv#GYZC8umOS7*8PnxRs%Zt?5H4$)RrnBXrhI#3
zlcr<e-RgGRPrw%Pc&?6F!nkZC#vR@Zg}r<J<TPR={AzHn$zvT$I)$(4gy$7{96L{w
z9D|M&X%6F?b_dmbW6shPw|E2GH=yEM9+imR<AAmb%>63VMzUtTPGDZFf})AQLwcrB
ztI+?L?uv0In8GbANLCWUt)6zu#ltN;0HVH1==wj;>ZR%?cK#yKqxA$;ZZlf+Z6xzC
zLAV}!baqtXr*K2|CtVB$T^#5*<w6~V{X>FH15f8|FlTgGDLe0U;a8g(H9v=>`p;qx
zW}4f^n+wT)t1)cyi=m!{K84=pFb23b<<*lQ2R5A(y#{ipgW^5}F7x&Lo4()dldKP8
zbcG*fz=>W_k$p4rv8%9U>{#M4ZY5~X00QqGv=5=o%K<KJP-J9D0p*M+jKXwrOF29_
zb1a$3xK#SSw(5-=Ib&JjHi|+W0TCrCsflw4d^P%zk>dq3W00;ev9zV1{glI!yO=V^
zCpRbDtW#M*T|6#=yhs~`KTvniB|Kg@an<?YDPSM{gnON{C!$Hs2M@~W!99rV+148c
z4lXy4a!hw5-WxdqGy8w~TsW)0Je_GEbR=&NtG4d2y#i@<;+AkSmY@chV^*e&-~hF@
z`~7oGY98CLU{WMiayUSJ0M1y*-;uIJaxLmK@^5Zi1`dRq_8%8_6Dsslr5faZ$J0jX
z@&ENQqf;ySSBzCDs<m-{Oql(}Dtr>$I^qXpQ4T%+l3uZu0mMB_W_&S#ES7`SA^Gnb
zFmOLBGorl?MOb68_~?e82)dT5V^?ZjZBSN17kVR_KXV}E<<(W0<py;E;=IWYUB3~M
zqVqlNYrgBR*&?I-x9o`$jO|&p(_*FHm?p_)-Rh8zUQII2Nw(#Ev%Oz>H|E(Hf92bE
z(W)xlGE23ZAxRkyudg(Ee1^!8xiJ9W*u!G$$-ad$Mvi9Y9x6w9$S{*Jwq7zPxnwNG
zN)m+(L`XE&z9Z@l$ysnde5a2k!%|j0`*<fvB&`+kY*^slu&>;*>bs=p&@>&56a}yD
z9QDl7Fa&EQ&!h6ICgYR(tHONtZ#R?~6C+I!w4ZC=UcP)o^5!#a1fqepKt$`BG&gy?
z+HZFkiI%aqwl^%Dbb4!aY*BNWm!8{BgA3of9LReoE;)`4bA~oIH(#77mQD)NN2u&9
zBWLv+g~mD(ovPle8&8Q(OJBI^g9<k0+`s1yun9gJ$wig2>o-ICUWXhAzMH01_i;~p
zd6lb+Jyn7N7w6;uFuTzRFF^DB1QcQ7o#q-Uo+Zh#DyvHx=ziwzH1;gHF3Yso-;X|!
z`)o|1sa1bDjtYrcKr2t;><k$+l@H5#Xc8)G*1r3g$gFfI#XJ8R8->WG9zE1J<&#Z2
zdigLkT6A68qx5;!Xv0d!x093p)4L|*Y7RbH`DV;Vo~$jHh9YyRr!u6enh$uwM~)7|
zTc7$r5qUqBYiIGXBY)So7bXbnK(@Tp@{AVf6knC-p7?New?4ule+QFauj4Jm%aGD7
z2ZTzyr*t4*#+5O>>4ites$?9M#5`*-;ti;nM4rkLBiW9V(KRmn#*4lE#YzEIEV{P#
z3>hGFhQ&{S<x5<my#93_lnqc=t)%c3x3ryS@7Z&;j!VaEC+LJ<5o4PvSLG1jXHB69
zStoBvjNlntYV68LetyStit^{8{?3Pm_F%)IhVVDV;%Wf_9vW2S7RCZdEKlphgjIhQ
zR!F^8aLtRa^3#Z~$8S2(o$pEX?Yc)nMCCi0KX{$L`rk?v>+u%Id`J?*4O$!Da#$#%
zix>AZxh{AwEV0rLtlwx9uI$aC`rey8K7HW1oWqPP_mAV3X83HX<Z)vJ-z1l9=@oLc
zcgzCE+D-4CLAGh`*AO$husgMej2|OPlg$qqeCCe=tW}A2J})VsS}c}IYeQJ-Ew*`?
z6oNN~by1J<Pv|JyYp+)Y#M;`wPx2AXZ*>dI!omd!?NsxQVGeb%LUjw#K3v;cz6v|I
zdK6VXm|eV!-Gf5-B-&z;M8;3kzFFHwx9DEc3XHr(ldz}A&ZQUJMN81bx83or^LWg?
z(1;AE9Bl7;J=(d}YK@K6oqV(s@qyo$KP!~f-7J>Chw8@{W<s~9^$!{2y_;MH%t&<!
z=uf~}sbM%Hd*szdwj}V12doipm%U2rKw2y=UJ$m=9)ajwS-mL3C716ygDx<Ya#<{h
zPJtTFdD1X;2ulh*5z&%>PqZ9w$?$<~<+}Gr=IoO1w8!$X+d4VjB}cMSx=XQfou2M<
zTpQMrnw7mrz-Hqs_&#TPJs4x7@mAI(P7a-qH~gJOmS}3*;;O~lLEb8S-p~CJC@-Il
z!`hee62dX;z@<<AHT}Fi<yxNNd9>`X8}U5w&x0C0R8DwStX@l%K^bxjUna%{KZc11
znKH5<x{d`HH(_?Axo=POd0yB0fU{Pm`KyxYyzl)##@;ik$^QNF#m9nzh|;C1ASeh(
zmo6e*dJh4VPD1FtivrRG1f=(pKtM_e9hDM#2_ZlrAfdO=d-3F%IdjhW&&+Sld2_FI
zzsXA0O1}5bwfFv9j%?jFANqw3dRF;{p~m_n;j)xCl}JN47L5~GT=Cq4F6$G)H4%fN
zaYB&!y#>nT6j_u9@L(dz6c<%I;HL6>y7P_DS(1|L)w7cyRxS0_@ylqOc*0U0@`+l!
z_6@{Sopw9_r_0ddRe$N)eh!>Xu!Q(!JLbwT)86G0>R^M!I462Uq-wY^$rB?U<^FDz
zRqWPZJM%T+=l2GDvlL1a-Xu*9@qEY~R?D4K#iRZ>xw8@2B2eU_BIw&+vcobD(M5hl
zcXIJ0f6C^VW!q*^+Fjwui84aXLfx+X(PtWzs3UWax!(A6pABcB+r$n$<V4{BY!)Rb
zP@F!Qz3n9vTE?E~jkW(Gqkn?|!jOvA;snx^^~g*8w%Bd)v}qF^Eo>ayx|+2@w7wat
zy?Kviy;Wq}f^xwnT#7M@GwJHwq@fKN8`v$~^I&3<uiRK|K^AmL<iDE+xi`+~yoC=i
zCij`|r{$bAr-EkpGYbj?BUTLI_F#56v(|L3+)491A{+~UAw`ECbSfgx)FE>^?A2~P
zrPyv`xp=WN5&lF|nTX%s7TW7|{SN#~R@1&<<|EMyb~S$=Jv4Q1r=00Dask*4UC@YG
zWA}S%8z;B@YIq)~Ued&dF2c#ePzy{W@W2>Khhy)=Hj|0fJP#wrqwlGM`)&eL2}X9T
z!K}uEK6Cd4AK5&cpS+wLSC(fU&eVZe&2u#g6^1a-CKT(H^>|uC5~?qx&kb#Va3w9w
z*>m)@n*9!`vj&|TZXK19!t?Ebi#c<E>VlkR$F(a|?<I(syugR&Hdzc9<wzf?O-Akv
z>XmaHi=;AB=btSVoWQ1m)zMxtiy_~jz(3fFdvP2+zQ<L=b78rxf{)s+hA<bpE-v)w
z3b#{e0)PAG!;6je+<z|AG6K5M+uIc`g7c7?7kZmtgKt<LR=uGYs2t`k7EmVm6W0L|
zk~hkjJY{)7Ym#p=>$8=}gh`N=Z~r!gXt>uQbcVM{N<ex2Jqz?-MR039VSvsrEM0;B
z=s_k#*(BOCg|-3CZ*?OPRuQND@5M+pNzGspl%x!+5p#U-pl4j!kJrx(Vn5Ukr!j6!
zhJ2V)4FanZbxj!29DqFGsuHnCne+|gZ9A$Bll|(vl;9$P?t)UJXrB^QqSlQ6RZ1Mx
zjm16|JNEKJt_30nCXvn|_Q*|YA!yrwUyufwR1~kejcXMFTBeCDvva*4IK61r5%<Hw
z_ccCYVG7v|+xqK``?VQi4D;G>o4c*FR+XI01u2d;6phQUCxYFRYeoyn_JG2NN9Ff>
zsC5BPm^U5FdiNGev2eeER^I4~(S;M(z&q)E0|h@;w<|Y29^%9km`bkhFyQ;YvpKR9
zuG3bSmph8p(D0Ey_MN!OfGo5?Lukd>B=Yikf4GG|MyRSk7L)V-xa3B2iqzuea~}h>
zVSxVy>TKjXlGz(4L9Gs|29fz_wFEgfhpxLF7J4(jdLg?eVV#{H3C)Y*fYy=1Tct3%
ze-+&St|rH^Okqp6{bJ-Asm1jlhj#yVzn&kEpjc!Oxy<^s|3D7=pZP&T@wgx*3A$_9
z<#<Y2IaA;tULtp!(#gjl_G&0_=3%ID8Vl2Y^cgK-x&1OXe~>~AvBXsMjkpg5-K_o`
zyO@k_rFc-&xMGwsSp~Nc&n-PIdLJJ{8jFlZs<dKCUO5IYctbBV%fyShjCk(={~&Bn
z_I&HA?oCfkXZO=s+q;92`M>cAJXpcF<!UZ;Rk{m<<%hM8+r}5ySwd_rz<T*NJ0Kid
zVKsK)MM#_LZn+Q6x3<j_tIZ|e)>bQ4lilm655NXFqk`^=h7P*jrLk%0lxdqqXmm^D
z0N`~mZYSiy8&pk88ab)QgTmdUx=#Eo-UxTqlMczNlu;`BA;)d2+aAo6HcN}79&eoW
zN!X4B=)u&qeP3jliU&aa_{}VKL)SX|?cRbluq!{&uo$(?<x~ad;Z!Zyw^S&ant8Xq
zeB1c2l2Q%PysKR2!o94P3%Z`gdp4eMAEa5nzO(j&`oJtC!_IQSMg1To(t;IiDAlq9
z!HIV;I{8oJ@{Q;6O&)v@Q%;lDui44Tqz*sG8wdr;&mL8n2LOfDxSBt{aLpQTrvvBO
z)=u7AaV9jV2`n~!?d_I&Bzxzu$|8%-DuKCTiO2o2GY-m!i+Gyuy6APmXQcK7eK2`&
zqNBjUZZ}6bq`wtq^{6_@j^pL^CAq7^+%k*nB8;wY4tet6S&fEM-O0%{FC+~j2VJ4}
zIbN4Q9VUc*qj!_W7ER(A7O>UJ;C7~2C2QL6n~{21498-tKQ3c_o{73ncfy~s1!?-N
zHSLXAkJY*EY7+a88SncVYesaB@4G$OBH6q#>)>RJn7yhIlg%*H`YfXpBbK($)#xXo
z`j>20Yz4Ox&yC@d@=m2}W|fX7DV=rWcy?~joxRSX6-|QWmzS&Fa!PgOTwSgv9!)KK
zXg1^Sw4O!jH`M$?WJmPX`4`fyyY!nOLc+PNoCn%@Ro3V_S&^Sd&pua-AyulrcM?i<
z@wuiEzkkOkD$9pokh_#RZkJhvYm)w<h#Y@BhF4LoCa)*U>_`<!yW326>cp3B=mSjD
z2;ma6JtL0M(;3RAZ*%6<iFHS{W1a>XN(Oa3dt*!@67DL;;Nf=hiUZI1?yicCjw_b9
z`M3}13)44(2nkpI66YlMA>Hvv&$reT*Eg-`ZdP+oo6y}9(+>?)5T34!P%DxAE=cxh
zH4JgH1@MB;<%Tji-Fs14cwlaZX<I+o-pb~b{w!-Hh6-EDUlLwaVw0g17xD6+oK|V$
zhSj4cmtS+_D)vtI*J<;(j(Hm8IuIT>?gaatVeMQCY_tX(n1FvRIh_+45WPP~>)W(%
zkfRY(*tV=68oK!8jN<6!i4VUi!OD%bJcR`?5tRWZb)Y$gt29x`+rTq(pfAkknkj@_
zbH?RJEU|d19|jCQXfC*Da3*6vv^zFRB`DYyh317+G+6v|PH{LcM%$gDJU9*a5{)pI
zP-j8hF+QH4Jho=f1+<rToAtd>vn9@T9aiMtT{ZJ&(k=3Vpx@0cm;=@zKEh@ll2IO!
zdl4)0V~6mkqV?Q->3L>f_-#!1F*6pl5z{2#gumq{nP+=9&)=y8b~w@mc2s&Cz;Y;e
zfKq{ubVlba*GYJo-b3!DOKTKBup^t>(TB3R3n=qBjQpHe1lvNwSp*oYZF-J$95KyD
z%-nuKkOutOO}6ma6+Ft>*$%}nXrY%hHT01QR7=+oEU$h^l7_!r6K$ek3sEe53#}6U
z+&p!*TTVESHtp9!sS{SdgT(!LIoUx9q#7G~%LV4xzml#E3&_|j78x(SXJyV|d=j;5
zW|XXtsUnJ=3p`X5dhj-~UW(Qt(*<0-CUN_Uu_nQQ6p4FakOiw|Gf3{KWZytBUJ198
zUe@C`&S$Ewp-s8Ln?p_-QWbC#uybT)_KLf~&M2O$>e=tlXKr13(EdifqDpID?fHvi
z#+UtYwacpD$dJT3c#E@tQoux2eNj_TMk%hPQ8#%YS=w56juhsQn)jMxH3#;2hEST_
zG8vi@<Iyuf76|SJzp6MDWqGx!u*+S|oj1^h?&h9+Dx30Uy$utf8^B4DNv6>CG;QcK
z=+&s98&&1V+XYOLlray%B}OXlbKH$;Q@r3s4I<(ty==c#Ly$(dv@_Qi%YNtbPTkwP
z%+#+7i;dACbuyZP5CBek)ojTx<53<pkWiE2d0>`(Kfk|RP_fLy7ki>YpVyv+Jd5!?
zqMEZ*udVWR^L&P$T|NxPsth`{zf71GLr>yyClYEosE+B3J=@F6Wuo>uZ;<RpR#?}z
zF1K`Sg^#4ONv{8&EO^`I0dzYkajA{3T0%nM2}_|N$(tu9@X0^B{aIt_*)SUzkKn&i
z0cn&VZs1z>JJ}y{tL@$C`x9(h!D-{YoxkWdU6EO-{yuMrH>W7qW}hdey0OjBV&#Ib
zA^Gj)R2-+|u)O-uUAg3=ROIdORH1HJM<~shSyo9d)llBni_$yHi~FIn39YYAf?twi
z#otBLvL12mlPWvlf?lfVpEGw@+)(<5@a}5gmQ05Ng(flg;mMqi*$kV84JhhYK#`+|
z;?cQf-C)+2#CUgEZO~c&d`NU%dlc>+b4pRNV=yQBkLps?j0mtZ`{^N<PgN#@N<h_J
zGpT>0?(UUUl`sExpag+3$Wk?az3+;lXQnxs?POT;pfXzQ#n#Ag#64TD#xv`ka+ES*
zh)<(D{CncJG|3zO%dLjorgXwtpU`bqC#GDC8I%?(PM()qTQa?MSr^gTvq=vwcte`a
zk2~<z#v{&czuX7?iX_gU-!Dcd@*nQCb9QFS+2LfPqvIu0O*)``8A(koLkbck;^lvL
zjx`>tHKnuxASPiXbfeC-ro#Q3?C;BtbyP&CFhfBO_;*fK{%9;0*D(CQxy9vbxqt#^
zt~f&*P~g9p0%g^4BjwVbalpGXYyoFF(z9YZU3|RJ3}S2Et9k9s-ka8qf6iR|(JfyQ
zA1r#CzCT228VK+PL`itUvT_POJ|LD2d<<ar2&}_)=zZ|y*)d`-nzn{_pXyJnnKV_(
zG%0a@nGjhztP<Fj`B>zdmk>O@hFcD7A}P>k6n}OV1dX99fC!lxc2I@<Gu^e>VodD{
zMo4@<VZI{5LYW;QV^K;Re|2;|-Y`k`$X%t_;5CwUmH_|VtPJ|>(qQjUp1Or_of-e=
zEq>~eks~gqE;iFJ2#AZA8GnedBFqP1pt)6OSnQ7>kjXONv11aD*7=IQP?8kATr*s-
zUJXup=Gvd(@WLdnO%o42-JUH+<qArkf@~r9I3a2y&MLZ-$paKUyJFBX-Qw57by<f8
z65pSjw(|0l^kB$$?}b*CVc-=T)Q76E49R@Qdp)>ZSBZW(Cr=-#?~q46`v0+T|A)P+
zfAQ-b2~ipH{rNYY|3iU*e9Rk7!fujSwf}$ecCa+P+atB?WP<?f`i79G94>Qeh4jzg
z6VoWj0cEK4$*6vk^TrV?#~Y+*lFH?_r0gnLC4l%EBl%R`uZdy`J6c8A%emrVFZQ*M
zX6$<6mPMUmN&>qUarvz)56;jyNo&~}LqTZqDC1uuDEzLyJK@Hv4Yu}jdwy;5Pe}8X
z*fJoN`xxk$Fi>5|XMIPO!btr2)1NPz-&x1{@W4S@@3sWAl!pmrki8)eCQ3El_Nn80
zl-cD(ggA|emt(BZOAFzKD#5l!a7+gcTbCQt1-Ne42zr@TIs<gZbK1X`eO1ZQ;pSr0
zoSdDWSbItQ4L^tdMEJZoNyWGIJC?rvX`jw=S!G%5(UgLonriOOFIJkmPkvJFezXhV
zY4*l*vV~Hs|Ea3t!ib4QJhd`jdfnEB_*?{GngThTaCAT_)}!oiEhlCUHZ@&sBeAUs
zx($)q;UDUMg}piwd{sdbG}R{}3h}~}mga1G$dK;FIXW;}{!El}AW@4NG#e+{y8Ho9
zDpE5^$|c&5S0!;-$L`0^9$Kr{il|oH(i7T?pMv3nQy(WB7x6XVPR=LGJ=Qg;+NZX<
z@pF3vp6rVg?`%g_z6CIBi*?ogC3_v-^0UuJ;@rHQVo~f)C3iYMooMV<^&>5fAdp#a
z&ckRx_n2jW988jR!X@tnBuE9rYsp+XFZt9;$Z4Y+i;4~RvR^J0H8v&rhN#E<Zu*eA
z0k&)+PeiHiUx@XC0i$i%MUK>|?Vil-1b#RDA?0L>yt+72Txf-MopwV+Mpt0G9I7YE
zX>-41mu5MUy<mc!)WoOtdr3{u4+~Bib6|f^Q;lF$4`XuMBW^q{w_x09>aOltb+)})
z+d70duYLAQ=wA%ECz9~fc*v-1dAO)h^9=(Fd!x&ItDM=lSmA11b3A44Yu?Hjx~nH`
zVP<<{YCKOyJQdbwQxcxcfTe0~BKIwXVmz-%2m*%$z`^jH)oKS)`KUB_;VRi<*78Ut
zufkdhBc9@=EN53CE;IwEKn^1{J})XBza8A#GVr2u+$Mgm2o7m<`R8LoC2#u|uf<04
z0NBq^KUYEMK{@+_(WRDlHJdlbq+vQ^;{5~mgB6tmp*WYiq0EjQv7%rDUGkQ%d$XU`
zEIxwo<^C~$is0KJwL?8!C2o3*Tsj{4Ln1#O)RtM$OBDMkC*WFqx^sAwfo(~|^`J>T
zdih9P>s@h5rRS_n6#>W>YJqBmnBfQO+5x_c8p%tYEjP%|uvD;Y_0`~=-&souvji9*
zN?>0Uy%}>-<Sb+cfQ&Xgnb1ZE-{SNd{JGSsTBV;ts=!Jmr&a{i_By}V%y>&8N1951
z%2*X0<VJVnbw8#wU39PvO5sKEFaM~}rD=|vq+TJ$PrRRtNv3R8U}f$i7C*UiDCj;f
zZmf9TD0>4fBT}|;H;M~FRdYFQ;ya9)ct1B4OpC-nHd#XTvj*03fJ)0wQu+T8xj{bE
zvWah3pU+rd&~2OgOU9A5$I66Ra5*tf_|oNGY1l~o@T18?^?)*gO_MEWNde4o$K;75
zel6L;nn3kNQ@1E-0i5@!Ho04S=i>~vF10GaJGKAu;{MB1lKpIBtj6&f4!4UqT1wpH
zFJxD${2A;vb4F5ry1nd!yt>-@BJ(bx6?KmVqYxTi2ruQmi;lin*01u+zjkdh(mJ>F
z%&5fGoJ4N~rcZo&c63;obZ%m<x4YQ(!syjBBrC*?(M56FT)V^a4&JxtCrZ5D2>$hp
zcvubzVf54+%*Gby@@bCJ8m&KQV+PjndU!7{aox1NLu}SJp;J*0!3;_IQ8V}B*(t#8
zP%h@m&EP^ah(00%EB<2Cbo!z}e_;D+0^y#)Dpl_+NU-WzETg&mDSiRncv;w1`qlwY
zyPol_Nfy1fqsXPfd@E`B872;OqR+-B^_NUBvr~&lBMs8-#Sw-;uQeIvp)+GrDbk-N
z2AmlctnJ_`&!w&Jw$CLmUy?>npLCImv<f>)t!-SGq9mfg(OS66`8r~I%eo0kB2ZDT
zVf$W|Hz&ysI!xizgyR;=*_Z1#4?=|56>jCXlJ{=|7Y`Ds>L_kR512K=-;MyVO1An5
z1p0Bn<f)mvk5yHSXmDeNWs5Tsvk%c%k~8~yhYH=uZ{5+HZ_FEXa?}+Nh;y>pd@qhw
zI*$Bc>xJuN|4!GMUMj6JxaaEuc2?@Jze{}p)BNL^F+b-lG#X%BWA3XD>3F34%ZZ2*
zqw)2jk?8ZQIQ3`VFY(Rv)swG_V!i(<znp6Rdr=}V$M@%UqD7a)XQTCr=M0u(2>nmL
zHfyBoxU5U*11s7c5`<iC<2sAeEB^_ha}C#C1t*QSJ(@XT&=yo0ohMBlN+#~vh4jQ4
zk1{gf)uiMQyc-1W3k=%1n4J_-t=SlFo_H68kN0tOwm8w)MK7=D#cLd1{v~_ybhVG%
z<q8p$@`pq@G4W0MW!pW)as_z2c4EB36YlA_S@pfvN>vqguGj|NXb)>@R%CrdTdxJf
zlOrHnT1WPo03kew&Z>tRgr?&$KAwOa|76TrJo}fd@U_}$Mw5dL&9g^~%*4{kz0#T5
zccr%m#p^{#bEq(7i9yX$iGI!U`<XELm%5}W)D7h1^KV<@EUJto*We5D07C&O&1dK*
zvSIL;3slARtwzi^Sq||wiqp_hWJ4<v*4zsK{I<~(D&vCGopP}s8hMvIY3<AeKUvo`
zuFRdS79df3uDx0Ruwi2CZg>u5oUIYa6b&uqv&-ApR6o3CUiT%D^&8G35QDqTyh-oM
z$7pa`2#@09)cZ?j6jUbqZXPk0;zs)M+U-j$w@P$U?R~QNKzMB&aw{xB67&6Z1xE@3
zc`-o|+?V?tfGXzGZ1AHOsC~ollj1ByWT2r#Fd%~?jl2=|g`x12kpvOczMBbaabn!|
z(5tHcr-JfFAH7Eo?||RlS#&8&V8hSy2UWzgk#{j*f$=Wwd(^?Qpf9eV0X}5b6&q}z
zKf23r!AsJFmjI*cwC7|Rd{<c^GDv0utYU+CC4-h>&e9%wcWV}8#>}pLP?WGORsQ8?
zky|YM0`UMza{LfNPr7%T>G&d!{slDV1dPmLZU-Mu#r;lmKVyz@|4Zf_hk3wxXh%X1
z%2uLaq`rgdf1-Z>TWh#BYg#%M_)F#cv;Q5<{{L~21#n*fXBc}1@qxR)&J9M|+uv&0
zZw8+4dH^BX279;Ep+9vkP=q*LCnqi|Jfqu_HoNxrjB|q=Z{me>CGNqxt8wPZyEt{R
zN$cV3&`NJ33H4dyei~0WmEQbf)1E%pU$Tc4t?GugNL;Cda|g5%6VE+>2kr3ZnVRrI
zyZnf7uNxBgX!z&5O;+d;gz}OIELwv5ISJR<>dKt5n=zkg_Bivp%Lj7?HO_RRlgVsz
zCD~sxvz*}Ub(Vg3ROn&^i?aM@L*?}RxDQ{#yuY9h#k6(XUUn~BcUT2IL16A(M*%(~
zli(T2v7sva<-yvM2Z+y9#Y4w1BZd`&UC{)Ahn%cIpou8N?ai%aPsyjQ>4ygeg~rtg
zv64~|0-Mr{O$&us9!er;;&;`ZUjCVk;`J=$MnhRDYbtKb2Oz27oV+F|joP8uhP#<!
z{+gR<m%tOHU(VYH6;M<*K2a~Q>0!pwMG{a$B0anOW5&Y5sVVW!J`aB)ve;|_ah<Ch
z`;S{}Eqej9o8i^RA@UtUa}^}s9M$FaBJEnjz#v9nz+?XfhYV_I7<n=VKUR53A<@_I
z=-wP)O_YsZv?uOz?=+1ousY|{#<p(N8zPeN_bhWlhqv9{=4nRq%6n(GwP;eCByp`;
zT0>yUR69M#J5&UlKaOzOtaDuaa1`bDR_1dRS39=y0#c=EYqR83sJvgvn7{CF&|WG!
zV_&pbbKF0^sj+d_xdb~R@M1yIF{DLy(6fmnqO*&CGkE}iyeQvsSXQTD!Z2~p-S3X;
z8V@}!IvVtD<7oP%TI)f*O+<bfHU7#nVqHltM!<Y4I+V?D(-~!v=D(0FNZ_RQjHxzl
zk80frZ_6MkQ-q2QwBrd)=2oYimI5wWZyLO1J|tLnBU|YIWb&CY>?$KU%>W_&EyAvV
z0y@?>-iak+kqe%A(J{|c7vhY<orjK@VORYsUoMoB-pC;K3U%wBHD6l~T8H=m8g{t|
ze)knhl|x41X`$rh6e}Ol_SnPQ1Oox-XWN^6m^gNmDfV|8`~7+=k)D}w_buTPm1H?U
z;uyO1SOM+%mn<?vp|z1>!hYp?J`WD&YuVw=aj(t`BdB@~zFXw(1=wq6zR+nHM?8mb
z8gB^`yAKy1!Ph>ZkBjUrX%1pFLhj2zyeiC_QulKdLg_Bb4=SamHH|*w&CO5mR~b9Q
z59k5JRX|768uDGpQfwgPZ2S&qu7Ph0zDqvh=GqAso!Cn^yWCn?%!c`F){O>y>Uby8
zK1bhbbqq*Lqr2V<gxIm#1k<xe&Uf1>rOL3iC`$_>{JlTTn?bm{oL$V{*KJn#xfHu;
z!Q*kBnKM=MAvcO%miIsO<OqJZawyguMEC;XFGkk#fg#KUwMdQ>kEV&Z3_A^tSLVjA
z;+cNU@nF0!K)FXw-Ze@-gooli?dx?tLm~2jJwEK1BEDm!xtXKnvbJ+yc#^TcHenmz
zQ)IuEF^-6z{=MTDCos+$&LJ{1OG;3fo<&AgiZP*Ye>i_iYNeKkJ<&j7n>1-VMyEP6
zBB~KK-B!rqv-YZ1KbGcNoZ7Rp_?@=QrD@G_kjY0wpJ$tGbAVw?i;b5OCHTCdcCSs&
zN-hUM9@V0X`z@-`=#>0rLHpEUZeDFZ_ujCd*K+^9h&7AKb-A!Q-Q`KNceE78^;KaI
zGV_h3@en6vDp#y=BW|kn(?NsNlb>HN#*nYcOVPC#)cK0@q%7q3Yi_ao7xZaCW!ipJ
z1L#^2EmE0tvlLqCkDz0YBDtA`&0Ltp`f)+VURy7_Q+kisb8K8Ex_%mLl5Q3L=Hk;Q
z5`|g5<U8m7w|OfIVaY+~wWB>Kmn*d;V0pQ%G=CFq8z6nBxO=BdI!x2q#3YwbrfS2u
zVq~i$kG|N2jP+U<OXRcuAbQ&f(}#sRt+%u^wW%76&QbJ^?-v}0W+$Q6NI-caog%B4
z_p9+{J(WB7Si@=GU(lzk`T>uh>_{HNk{p+$SSM6i`C<v*v4stAClz5yNzu6BhWtUD
zhMlYK9)0to)Hwh%MnVTvNTZLL*X!tAASQq!2%THx%g0?_Y4y@+(zh@8gy7v5G^+Um
zOU~|klU${aFZiS&cL(pT)5;CbHb39gQTsi{4^h-?ScXY7OxMUc@A7<}P{YPgii>97
z()3U32+*~E;%u+#k^ACoLTRDuwn7N2{!0=<{(iXy&FH8R=ab9fRzSDhZik*aRyGMA
z3(zb|-^!7SUh`!6OC~H4o15iXsiBcJ6v;!Z`6CD#({oymcNp#)pQ7zcIm_-8zH<%b
zjJZT)md>iUqQHTdChUL7)W`5<594G=poL}gar2t?dCl8N>UQhyKelz-iPI!Gpz|ex
zG}5q0&p=~^X}uJmN)OOe+CoF8nDY~%Hp6h^L|88icMzy0PEU2*aB_Sv|88R3n=L&x
z!=kSN<1{KybsLf)CMoq;2<F>Jw2!_KKc*C0h3_@-(c=JEKQmV|^$~`Y`#=mHBE6vh
zNHQ7F*cB(0*8@ID1m}~2ctYE+pSnx(ar`TG@cV$hmzJ5W)*zZ|9ApXB3f|RSC!tgg
zq<IxD!(4>e8t%wc!^Z7Gc4MGLdG19i`eTj<w9J$j_G7Ed4AbV_1Jl^6=X+!ZGEEq3
zFMby{(O#^?ICv$ywtG?dNasdx+8PR5fkMufS0OZ5u1zGsLBa!0paFMhmVj~9N#`z2
z%9~CjPYC?)Y9PZ1!+7<y0S*ZZP;5p2<LHtZ8u5kb6q4@AM}9q8{-#PL#L(gF90`#V
z7o&4$YlPCqw*(ko6YkwxB$=z|Jj;|;m_f!zpr1?TxZ~CPGMS$_g&SHk;>#at1Xx-{
zdl;+B7P{T^#F-Om<c&!sh6Pz25^z!rZ~7lF_x~@2o0O9yIoDcygJg35H!uIc;wzqe
zk6imd2jpnI3>%Z30(bR^6&t91lQ3U;U1J*!>%m}NP}sDZ@8|E1%6wgdPp^h+liU?=
zvU2zzXh%&H>Wrt-KU5{jdFAQxJfky`q64%7{FT^k)jXurnPq5Z%m#h<SbI(U>wXiK
zw7H)DMES4VeVcY5i2dXuG)f;a*>O5|4Jo;>4Vh+XCln<ff(;d_Ne9Mq%hfZS#u}G(
zlb8h@CDypWw;+3@??Im+>u)5G@W{lyW8m(b&)fx{nzxWZ8~^;Q-?w<O!%A=`;FAid
zx4FuHs!cLWrLn$Bf3}hD5TaocJ?_3S#|Cxy+FM8IjYKas$Y=;HV9qBbDtnK%^!VOi
zN8O%1IXtM`HZa-T`mJ)zzkm}0nuo`jgYHly%xcN>0z;_YAKDv5vB`g|%iTQJht9y#
z@n3ZO7Z66f-0;QUU2Tri&sPy6cun>+EBbr?Bz*H)ZrLy(k#-}K-gpaI?o>9Mmn#k@
zuO#Y;zo)D2?+Nxkpa*fZR2WO1$4u|SMxP5ytBaS0fL1T$i$%KxNL7_pgals-<AroV
zoRB7FqsdO*tC=EAXS~p9R->yzb1!#0cc<(S^3K^8_nyvuv$Pc9w;j`>H~Xf8QSkVK
zA&Iwl0Q-iyzxHcz-Ojh!vT=et2ZxUq{VZ^}0wPV`>qOa}$Z~h$6uwwLV^AYEVp}iC
zv}TVfEjspP6lk;dc_=GS5u<@A-Q_Mi`r&4k)$2B5jc$Qzq^}3uFG(7iZfL*yIrccZ
zTV9y3Rn203$UcsBICI$9YML2(*z#5Hnf3V~U(hmpDH2`V!ZUw*2nX@o{i!p08JAkE
z-Z&)n-WX0IGd3}+RDkl$e-jb)Tz8Sj-~El;N*0MUWrg`s^V_$l*Sv*%)4CeNFz<1W
zt%1J-kp4MFt-DKo>;9RDl9UIX;(tEK^J39;)7$LkY#wI0n`dIs#}a|ftrXF4Qh#7h
zM1^UuZgptVTK9Okj6G#Tw{T|9O$?y?^uSuD@ugY!oSo$<s4(s5pH`0FZESTrTB#YS
z1<}V9B2+D(uW?MwHiBirT02g+{*vi9EMnZg=Px)}TKex*(VQ%p?BY)o20Rh=9jZ+6
zvcV}+eTgWBV|(-6>!@VkOueKTiwGlm`6T!I0md9If|CV%@G+W~t1U@EgNpX0N}vMV
z<LrhRR9ldyC$o5H&5~}X(DTzuOHR3kCk4fCQ=-EqcszvpxgEKp>Q9d$$LVeBL14tr
zcEkcA8GJ#MqMgmQw`QfG3}CuDnDvmQ9osxdePT0(YdN-jF??M7HI(z>gnp&3Br;zs
zWi~upP%Baq`Qpv+bKs5OPrDTx(|#^8({R1R7mzgK;InHg=P@}UtVUkP2~K*8Q>)nR
zidpJY$NSS1K_(e1UNK>{<?F|MW$W6<o*QnB`jgRn>zH~}m2zrRhK}#|F83W@KyM<6
zi!xr>W^cWvT|HulsVvSWay6%&l=4vI{4S~<+qSEfkQx%*u=`79d_(cxTqqnW7Q;ra
zed|of8dU1ZK%tPNcs0ew*Y^#WQ;N8i*NTBSHks!}eqf9+M?XlG4t;%g9hllqFcsVA
z4jbdY?c_q96=109EGEb<-7k60o}2sapErc2LFx|I>4M_c=G(jBNeT_fhNBXw<mr>v
zg3`*(;sOUPD2o(>x6|jVtpJrU8QvEQjK+MiDSHcSTGE(cww>a(oEsB&+gy&#5H9AR
zYw3@pUbFbEeB)S9a#a2Xp;b3nu>j(4>@rjxg-7&F`wYwpv8f)z;1BufWFT?kRoIwk
zB&?c%$^AL6;GDX1cXr6i$LyK2iOA<3(zUCO9jUy}rjT*)`Dk`Jp_yRB=wx@eD@Ddi
z#j$E1Fydg)Wvzmd80zT3@W;uFpjA*eA1YXEk4wxl`eq73a9fQ>&hO^%`lflQIm0$S
z>9crirRa*JBv-TAFAoBM-h_<$W{~REphIjC(Jb;_1HA|@LdoUc5W>|g0=`h))Az19
zlc_9|&v@Xb^{{_gr-ZeEXrB+m(gfL+SF%`TM?*Yco-LID(VANh19xwXeVE2Kmch?I
zG1M(tjy5|t(5Xu6Wj!~1<h&ZxoslMf4*Un47C!_L-WqjR^cwWP@$wUcw%G87Q5*1A
zZR|1pjFo0)Y6m~G@t%6;uri-c(@s`JeMv@|8AbVN7~Rw=`7xICgc;Q~ax)kSLm7x0
z>-HuzO|yp7n$PIn`6|3e<(C7!7O@)Y%I0x0VLdP1Ar@Vjn+}F{?>6=h+FOTsy&Xwi
z;q;PAdUr&q5#znU3w>JC7ntH4)>m7;bv_etTowT-`n6`HiakveGtY#WxTL<8W=}D4
zNks^AS{b*#9YnU$h`reaeW6i?p~d7#sV*V-lYi8nzS*Pp_Yen>!)RH!Qf_=ZCLI=e
zZ$%I`_+I1Eq-tUj?1>T|6il<O-s+9QlJ4=Z$DR+MP8wkdJD1F2BN@`UM>T#6mo`Ap
zzW%l+_M&yIQFQq=e(f6y{6O=+l{X>%!?Q<lJs%_Xl5=A&^g|RlGuUf|W420om*%2E
zUg`kikZQEDB(~i4C}{e0`Y?}LuFWNr3HDLV(!Uw{1D^tOAk;AP#36%b<q*87hF!`<
z-q7o}^4>;IYw;?*Zw+>VzmapvM5x{(9+pknao=t6f~Hg(&a(ty8jtdnzAsX3=<(e;
zI)aV-0OW47R#7PkTv9%$R_cW&<)QkcQ&^q5bkoHbnE&kS$M*KV0##%kLVcW__)TaR
zG38yxG`FCV(^rFhNopMZ$89cfNhWExpT}i>heb0g>hDc0Q`LRG^ETBl{UrmXaVI}f
z*$|`C1sXN3htuhrhHjCTk&Tj8hb4Rc=R`j}zmoR;O){zI=WXvvn|}T$kM6%YcmI9a
z+<#R${{!On@47cMshO(`q#%p7lWnMgIB#qMfU$C<Wm#a&M>y?bS9Tc^;Vo;G)voC|
z%8W@1z^+zh>6fP0=k77vVAg<}&v?i1w0>n4DNtF|G|Nal%RpHTx+mHcsrw+$$JYQo
z`xJoK54r}q->EyqP`(ALrfape_vv`V#@ExX4|Y@e33gMx$F12t>%ClCVNa=`ZS2?0
z3Ggcms^O^fi)L1c7ttQYbTC=T833=u-pqM=>(cmMfAyrQS7NH}#~s`rN7@>oR(i7d
z>6k?3^R){@^6<xMVwCBN^Q{3C<xuS-Gn2%438>EnFERIF?U+uAt>1msg$8mYwl>XL
z^;r1A(02@ovS)EhNapkTxn<PCQVehXd{%MZ(rd=sz8N2CNItfL<u-Wx;+Xbsd%<yK
zp;3~wiFHlPm-p}7X2wccnxiW!D&jplGcxN0kuC$S<+F`n_%k5Y&T@dZV;5}@>1mK-
zJ5aMV;1<hsu7ZON!0n5sr^=H<0v<h8F)xWyANY>a$T$<oy{z{A+DR562fy5gdonK<
z-%4<<KU#b!?OaqFA%Az(p;Dpo`KRey4Q$p%wu!~^Ey*(eL0}1K=I-pC3x)FH76G<j
z+ASyR+WD1&8h!$#W@#O_j*VTXTh7NIP(7kN1O%d_QwbCbM871z8^pU5*F3Ais?liL
z&GhlNAl%&`Gc^+O%Kg!J<;u9%EhNIuVWCs0sqCLt+F!vOks5njv3BcMF5hw2M*PL4
zn9E7ExyxofGg)3t^R2n3zpjIj3l{zJgEXk$59%0{NS)egPb8$Dqu5YNKx+|GkyjDp
zFPmk-QSInLjT2LmPMUc(#V&PiTN|-+kT#)NzOnxqY4Mh6XxmDf0?CzX!5&`P_DzkC
zYJGb=;YXF+u0%qC2>{$H#CTCttN)3oQMP>4V1XXG+?13o<-*|m)8ZM;%Si2*OQ6sW
z$5?{;zNS6u4Z|P_R%q+p%}DGb%}LHXaS2ffviokwdSd4=KCt>u`@pq_4X6Qft^j>6
zhS8_`&BqZ+RI^119%X&O)}5w2oU=$7>B4lE$7=E(Ao%`5(;3AGVgxI$V<wRMTe32G
zXh_;H^qa86PG{kZ5Q=U?t?zo~%BbS0pJDEKZyTp6y<Fa%8Kou{;!nbq8$u2OZl^6J
zvtGOEx#Bdgo4ecV=2<(J;^U1!lKJyimP@W~?!m>Iu|Rni#T7nn=J|i9zi}wBm8WX;
zZDo@RZd-KDy)c}w7XCx@^ibo6PKWDij}K*b_!4)Y4o*a^IEs<p{4Y6Tio=R<NQ(+!
zr9uoI(8YMe(L0Xpo0Uc*>QRXL9AaeC$?Q^UXAW3(Q48EZbq7ZqX8LoEmZ~Wlap#I{
zZE{a*)knTT|0Ro7ZS&J^wEhW~mDd>yd^L1r<kl;F7jZb>>vF837aW$6ij>evdSlux
zb%*TX*(07qIG;TA6i(Thhvk^vzCq^NCbjnO#^p*+K5iRSImka8h2w|BsIO!;$0nd$
zKqKf2YOb_dTO7#?1q#lldOv6WOQwHnl${FgP0&O-7P-{{Aw8UH<~a#}&eP}3vMuZo
zpS{BGJ!~tebyHo7aStqT@TW@Yc>S5>`@$7HQ_TLh`BqD6YRK+}a}2k51Sc0oz9XIY
znY$9JBgDDgD<C?DVfc35`EjNCWH^Txd&*rdSinFwk<A)+?^ssD;K+N=&zE!AV_$Of
z`4P^~dihm(VrNI_3_F4_R4OftGr-V)9c%->((yf2L1%+qn&E0J3@mQR0CDf;W+MrS
z6CWN#tT;X|)*m~b!5cJ(g-zp9R{6BBqO3j#Ma6Du{0jICY<L(JC|nVwaKVf5p}=ow
zeD7QCD!(=nJr;)yAoW>t?DYZl+k4IyN?_VlN1_=8m3m%Pu?poKDduczyNCZE0a<8o
zE+fNjBhL4D$%^edwjr6qy+c9;pfBJS$F|Y(&H4@r#pq&QL1$hxU;7(!3`Rn{2GYZ4
z@U&6$N2_Yta)$t~c5Pj*FItUxYfw!&Fw1~qzS$rxzu8dKdd7M~liu3>*QdMeDfb#n
zqqcgc4@w+ss3ofCoBFsvm4FMRS@n$U1xuREX1MrzWWmK_do{u<aF41O=`TwtvC*k{
z#sQvX_*%KMtAP(KK$lB>rOYXqzI5JH>gB5&gxbTm+~b!h4uIu7AjJ4O&)wS>;5Odf
zSGI-HS5x_Gse4hJdeURl8B0=vo<>PRPgaUWd(^kS+T$G(V_iOt;nJ5O(k#UeDFam>
zZ07YFExyCjsEkIo)&}*P^jg7h)zl`(*uw9P-%US+>k-vh(+|sviW6ZEnZ~2y1a2C8
zi*bsB?l9>*k+3$jfgq%Rz(f$DH#O45v9dEM!CRLb%<Jm;fMqXCL`{j+xc6GEDUnbd
zUZKFr9Jj7)-XTWop=v|_&`JN;KwE+8=pIP;k3o7FOUAA29svQ0gGwKxuctv^_YLB5
zBV13jPo?-Kt6i(B_a8S4w@S^p3*y;C-bSgj@0x{uWiKMj2smJVPeliUqE|BmTa~t0
zp$+AHXeSM3Y<rgJ4zmQ|dPca(Qa^-}$w<5<{!FiEPTn&4uwto!#TB(zs>HoC<hC^n
zdu(kX*}867PpxaRA!KpLgNw|AHF1r{J5Q~kE8ytZL&A^L#j1yY$vzt7I$rjy<Q$k^
z8+khRs^`@iCUHh!bGh*^Sv>@J3(Bxs<u9PkE2X-^UbjlTHRr#Qx&vGBEky<GTZ;iF
z7$Q|z6BlGKv;JG%;B1Yp@-R1z%lwJuwRf3p)LG4>d`hX|ij1-B(XV6V?wYoyqNhwU
z^00J(J3d~L5t~@TXOg(izMO>|#N^02EOVx=Y(6CkoXCF{8ov!CrP04qkQOx4k)1u4
z(EBem(*H^(9scio5p}MYm7IoU|1bT=d!%N<e|(XD*C?lkGNB-E1{*!J?WlA$f4g;w
zR%!&rUcOoAk89kp;NP`0@YJXHBW+x^o@V+r-j(mSjGr$qqTy%#_G`Y;LiW;J0KjBL
z;Nh0%$<L)sr2lYwTx)I3S0Y}mc$aNImz&a~F;d@v$mQVV!=8e5o9Rwk%1pcQ@}cCG
z1@$ha&J}Lrs?Y3L{q<?k##D<TtTJK#JgPR`)Rk#s$mo~`Z#e^GMGmO)98_k26_W}D
zRSVt;zBTF&V1~@oyeqqhiM6rLGO&96PUzF#r`AhQOO1#kA4IyySvLHMpgX0s0e1V#
zHv-ptA8#C=`nq`2k|mXZ0_%>n{s9Y}rtI?Ca{XZY<{2h!jLz7NyCX~BrlPT3^(ggd
zh&x)n{h5xgJOv6{Ea@7T`t_nzhQlA7WPLOK*RME*zhtz)`dZH)ib%J8mL-nO1i<V}
zwGdXX`lFj?x!f9EP|}Qz#BoN&i8v!hkY~aL6JLJUJSBj6C5FwzM1x)$azAj<^UT<^
z3`7aBs~-(a-MuJfa|80@XNa+oeEm-5{Tb1C^!5Ae-7V#v!Oy_ZSRFZL#cJ;}e{5v5
zjA|Mcmr#ADWefRUg$tFh-7&D~9_)joHqbTZN^DTl!y!J}b!j|ljbWd6>9He*Z3-W%
z-Re*MsfDv_U1f{$-lOn{l`Oa1_q^DY`E0?CozCO!74=+i*26Pw>rpFzR|Tkwtr(Wu
ziX`VyN8i7)0%>l>H!C$FMj;W|CEc7HogR&uyX91beh5{DX5TTdFcgyj_fyVan5>&*
zXJOcz8j|S<8$PJ$m#p#<6wSpe>!;^J&x|eWbOEfvUA_2hjxTN8qR3m0nOlG;c`nKx
zg>78pmx@EG;J;);O<#L9#kVY!Wnn{U1szQaHze7XKBT(9{Tbp4rdN+jOeeW&E-|v2
zwu8@E8p_~z@r)f{D=3|imB;a-EaNVGn#CO*V((}lxU!rS{O2-=*CxG^JO0K^ox?c)
zT)1QJ?Xh_c-v!&b<NVmby#TvU(xx-cx#Uwi9Zr3l?HfEYA5SLC@*Ed5krLWGY23mS
z!U;<zI%3Gkh3I1wTSr<B`|8*~caL6en6y3GwJ0TsG3-8yo*KU0;%2H<5VkM<m|R@C
zi(Q~~LZBFZ@cEoO3x1X8iK2df)a?snw*0(lH2<`L#`0aM`|rWTT)2~?9PQrL+p*py
zzJzpYy<i0Ribj*OW3E|%S@-zEXJ@y=A-=v8jlMnm(%WK1y1~)`QxBe`r{btjS03UD
zccJ{A-3Z(B=WksI$!Mqw!KAVv)z7J(nzRe@<?ue-oWr1foX7|%GCGQuj24^GuCB0+
z{v1qO|H`zVFoGTn6y$rEWLETA`q_)YlL-y*pL26P!t*LhVA(nJ@o$Rm>gvXfjV<Qv
ziDZhuWc{t}z0s@r@hb+QWL*$OVnUO=6d_fJge&!>FP?(4b;W<Y&EUusa=KRcKIo?D
znAn}F?}L2RC}kJI8O6{i?Ff$ac7GM6Ei-_B_4F6awF}uubOWzd@ytQld_FyxVaqyW
zPvNn$B+>Vh&j^d5Zj-qB25|P$nG@m&cQ(b9k+#@QyZhMZ6%<rNyM_)>@{pf+beV-Q
zxFw>na77TTPY7Mf|HJTSm#w!M;tT(aqu83I4Xs)FtuC$?3^@XALFF8x4q8i-y;^l6
zgYE7)jxmoxNn+E-s->%9`Hm*->a{oGn%T`&T-boB$AW#TVn%k5%Hndv--`!uX>r?m
zZTf8Lfd;XBkas<mr-Fk=t=^Pj05(g$8F<%7lnMT9#gbEUxyd@eOU&EsXD_6My{;P7
zjLS&&`$}Y!T{j`Om^_3JSbmrVRH<t*%Q<cq9V{2v;TnFyC3Ta0_<+j#UthsqC=b{e
zHQN^A{Ef`<QN}ZJTS{^2?z}k#<!a7czy*4cl61AkM<gAZ_*_<$cWOFUgpAOn@sy8Y
z_Atu%vu2|C(~1xQ;`k{2<N}mW<lDrPmvx_-Vq<-vo@{SH%plF=lvf{geb<%8oDa@_
z%X<%TCaN53>Wt2^$?XL=^zLO5>8F@1BxG<Yvgd3C1j6vEo{CFyFGOcdY^bU{x^&FW
zfnc){P5;4MQK)~Zw%*k8qq**wVg_K{oB7@J{A;jQbzMVIIHUj8^^=A!r=vj+t%(sd
za{*=XO<}3=TaeLIBH!!4jWDp>yY7L{?4SZ3NO#m-zONo?Tww%G(x<FQ2eLaRiExDn
zNT{Pbfe}i4IGEZ~z`64Fw|L&H7(jW_n@)tVGF|b{GCcGXlNtSu*ynpM%h<yId7Q?F
ztzAiiYd^IEisRnvG=DwAD$S($6V*|T$zD2iD=z-|+#tWA!Pi?lHv-~k40!lvi_5t5
z*PlIZ>-6ki&eh$%Ud4{NFG97UTML28{L<-Fa1$Yt@c$l7Y07elk(qtqQSnv9xsGWE
zcwn>A7qwUQY9jQ-qH)oNgL=c&G2o9^z=vhwoRS{_H=9^*F21NSq;bv5kp%~zO`hiD
z0=hD2mo7T?igb<L@EO`7vFJKm5a(VcZ>WX%iYTnZg?zcp`(hSTo|1RxV(IGWIm7$@
z98j%VO8+xDERXZGV%KVqA?Nf{$1A^$EPlJwxSaMlp-L8K7M?YwPA~#HK6F!h%i=JO
z?U&t3v0&D09nf36ZPI(;pM-ikixw=O1>|t+3|)*RmXxO__PJdCsFz@`gZf%Bk@;iE
zr|+TNgUTer86Rax##E17Hn+0`Ju}-Xs?@saF$XzqJEr4bUSi#!%@v|z6ZJV=RXZvb
z&ogeH$2BgsvAG+zPZfh}=DPHXC*zHR7e`6_gj#2mD=8i=(QuQrqx;|C%1x5ch{I$$
z{KqJbbib&dUE04qE&Ei*gSmgRQe<BLbJy<Qdr~#;e-$}Y|LT|mftzH)&~^PU*5yvq
zU|Qp^Kc;8vW4(l&dwyf=?ynu|tNMFZ<#=ffpL`#79qbDloo%%zh}Su7-fR+Dw5_(3
zTQ}bJa(&Gcn6j=HADIR)506%q25R}D@X-2GtbK@rfk27)o{9@2*X4Q@LDSSzkXn-~
z=~;S&!F98}!j#&9D*@lnAf~ZLkAz7yjbguY2>ReJnP5@eY}I5f$iE0V#_&Wdyj-v|
z?Kd7T?()_1pt2}Q(yuXl)M#QkUadCiry#-6&XlXg#2PXw{p#SbT->{WMcPy5rruWW
zwm0;k*@IKI8(;iUNqzw9KGINLYbbC-oYSguoaO$#!(?yNkSZOYaODxz<vOM%;uZF#
zS-%PGfc&=6)0SB)*$x{9nd*(`@1VY9bdTUN!%h+3o=xD^1Zld)yzjyS9&{dlMArLw
zia9}jORPT|=7Z!(<@`}4m?~5Wo}T8W{IF-=SWvf>@A&~UyMu?VHfd)lJ3nQa_5n@^
zF38g|F>d~lvkhsvX|b#Fg%iENdtWruHHP%PU-C4U<w`u1{tb-JEFa8j;vD<zpSUW=
zA{F<$-!BDWnf~oB*@H?vVpa6uTsFA^e0g<<!h`l_diouh-UN6S)uV}(JA3_~W&d#p
zb2_pJ`=&qs+MAl{jF#uw8?+sUfr2@+pVY^z;Kv*Yb?`u*8!)a$J5tzmCF29wS=MS;
z3330|((m@ouc{YdjoP$<1jQXsW>D1O&3&R<C$A5^7!u+968Pd-=@%4P^h%m(l6f1@
z#xL7Di@CjI%?B4&4y38}*73anTiq+v^o$TamwDfV=aQxJWbk*YRpm;)v_DsQ3>F`{
zEt=_-WN%=3lyIHIiI?Y4eqD!N5LQKzyKtAME14$tEqdz?I~H{3Gl{l$aaxuAstPLj
zld15yL4I!}^i9BFg*MfRpx*dLj94!-69*Acn=l-`$r(1LRDaES5<Xs?8sZcZZYz~)
zEcJZM-PS}#>MZ3LM=SzE3!f%>gi<?UlS1s*GJE3ea|ByN!s*r3oQZ)A@f*^P(vZhR
z$7?#ERQ-2rG^j6W<<ev<)1Q)0jT6isKoXqenb&~DP~VzmmA#LBY9l00Y3O*i#IGNN
zpfvFzw0SQnJvkGIsSjoglcZTEI0_b7l$F<uUVx#s*tbX3(kg7Qkl|-nI~U;IY@^wV
zBS-WS;4x0SL$Zof(I+sqXU@;Sw!P!9vhUTwAk82IQY`ed)I*Bw;V`!G0aSzH+xy)J
z5_E}cag-cKw0*rx7744aRKqMyZj~yZXZxS}P$bF7nYZuov0HAKX?VzDy9_D+2^wF$
zY3-Pb20q|)w%=Ymp;$l2=ATGiJ2RG@_L03wPbaNi>mnkYklGX{H5yR)Jk`5E=AY}&
z58(l?CQc%k^z?`(h>y|&a)|sJ0d2?hR86@A)VvWv|3Y!Ej8uI{x5l*{5FJ%S?Mf_O
zM?HysIjB{>v?r7ijgOeI`&RvL>>*XpJZ6**lqz9K9M=dMoQy`YY63b3%Y_ZIoPU<<
z06BCG;qU;7*~9XeI9NTk7>uY?X@P81el=(rrih%ag)bO7rO+4OW2les)|-2?Rm-qS
z1y*8_U{U7d&`d`|Duh$o!Q59VO_;?p_>Uzu?=3^10nWoO@khqovOHeQe~cQVfAO{5
zlh9;fvBFx~^k{miQc3agnNiH^Vprax)IUZgZZ<4x7Lsz`U`3S=J+zdB1{)~q7-hpn
z&61V5>aUr-DhuwB<tqqq6V%3*Y2TWiELw1vOphP^KJ7i009rk$;X+o7<R?3_8ul{?
z-Svj&PcB1eI!@B<j}oY~d~eD*oms>Eh9coXux`o{6fqafe8j<ELZ$eoIi~Il$!QsC
zBmWwsX*3dHYZ{NoS6d9R^Xk}#f63)n@|Zq251*oa+}AK@IL>{4_U#O#e2T$n$%D0{
zycHrLodE1689$y-x23NOe1+{hm$R#p=A|bd8EePE#0F~enV^Cj9l>6+Kc@>T{rmMn
z+HanVmk8y6xS!ZbE=3lr2@VX9NXWyf)|BKHh1(QgK3X`$(l~Rtr@I}~s3W|8p)&Hm
z4;#Wtzfaw{8DNCJ{cPZu*%K0-NsKggn8X%+Ib=ad7TwBE3?~y-rZ)ms0n#T8)uv}d
z`Lwj*w)l6vm_6oRlBO`Bt1PNW$5ooEHP!MsyQxy3UzVA}l$Z0JL9aE(0m@>}U3nQ!
zTifa$B0fkM=mpLoXOBvWWkoL5ub;YAowv1Il9uQNId0CjU<CIL^4y*cjmMM;EPdE8
zOD5E~gmyc;ARNX7;mf~ZAt<1S#V6Ps7I`5;171^EScN_-QRP~I%d)s?1vtjFJl4b9
zIsZ2P|Dx+F<C=UJcI`k=L68zqKx%+UcZjq!Y%oe1HbQDNs0hf2Nq09Z5Jrb|cQd*h
z>E=26Kks>d=lsrj-w*fhdmpw>i|fAf>wClN<%e6lfQP%=x?$|5?U16eg>MpkW98QL
zqaPLox$o-^kc$_6s|U7kflh{*e-NshtWtLwLl@=Oa^(ki>~yiryRCw9wCYkkd5r1y
zb-*O_^z~-y_)3@4ah`p<v1Hj-RP!+W6N9Gq@126$OzFWFKJ60cglc0gj-A4jOU*J3
zMtXPuQjHD+2ac(Qc;NzCT4#@o0Zp3H3R9UA0=*I%*tQThe9ufRFLvO;DL!R*AW`HY
z$(}N@s@ey~n7N&!(?WfTRhv=G#jE@F8eKfif^$3|@FD+Yx7`_;V!M6xpJ)Hi`G1Q&
ztZ_C6z3&p({;m7`N6Gm==Z6G%5<YUP?ElQk{YT{y{7P4y;Xv@y<k7P+$}(eIjFY<!
zkA!<26%`FvYxml{)+V@CNEmnV-{6~hSf;cb6oT~(jo~P}F-){H1jXp}tf;mG@cMU)
z1MBOGBnv%7d82DG1n6)H`b=6pMST5)Ti8LvqV`CFh8v645`%_duvF|}7MGx3XQeG5
zdX91Mlz_QIFxC#Fv`^b63|AHHI7fv4z-sw!CE<qJf^&l-OPp8@$90%2VP8fH-7Q?)
zyQ-bDzJeU(Fh0%Kxu`d$)qLZ0MYE+}#<f)WswPDJ1NDaO+fsASG4BDyhZ|d}5gaWp
zvTL7AKk(JurH@6^D6Gld(|g~N4HFHR6Hrl3k$qyU7~t|Zdp0#+X!73gU;6Dmb2~Jx
zTUB)^t^H4P@53Lz<4^e9FoT-UwZCekH8ZJ}fI#zvrZVHRICd6G#bt^f*@tkFk?$z*
zekVhSmx)NxK<2D*aS|=6A{#q~DMEuANuJJ)w~YFFuj+B?x?@jEi<HFps74lq7Yhvw
zt96wgETpz7Nc#f%Q`M}zVm5@46sCbV=&;L|$)jiBbZa5_Ao~NpE5J~8_>wWR5prF0
ztQEae6!tODU_+12_~Xl2`U5FX(Gv9&FWsMDf-%)jyfXT%<{8>3=mgXiMf7a%*L_}l
zXMlL~N#(D1e+&EeFl^Nkt6Exdttz2Ul3O_#VGpswg<L-DWGSg7e7@UZi<u@bJf}Z~
zb>Fm#U4mi3%fYJ>?k{AK_eYe&CI|H=HZBZruS!CnCg#-cYqdw(J66^3JUj;ptQvLI
zr18#epX76e^|BlFEW<###|1R)j|IEQ_^az_PIkT$nlcqt-Hhpf2djkNud7BbWvrCy
zP0`1hnRAVFI|O5PRo!0RNDjF^CZC+jo%uufpTyo!v~wD2i)9By52ntBXKV_M0lxeJ
zf1G=h<iI2G4gL6Ct=IxihOQm&rhb0*%L03uoq>H{d4?!FE6mYfv(es@&)y9>MRoWM
z^8bits@lA)!q6nNzqRmkEmjt+TBoT9OV+sVEi@~+-kA96_Z9fcE45qj^a(SE`-Ar0
zq<-8`IcCGM{}Ny@=H;EW`~G=$hH$IrP;%TFmd458b>+xAH^n8HH|1r<Ayk*6nU=(H
zRSVe!qedK}+fUbDWk{uMxKblExnUjv?pM{1rLqv-%#YR=#Mo0zQf#}{B*yn}yGwz@
zqm9&}v-3s|c@>WZpbEpXu`=nL{*;g{*IshQlF3!Gs@XO9)cb>F`^3Jq2bie75zpk9
z6n|K{wbh4t7JxFXV4!;eZ5{j3K$ar9YiHdF3@iz>Bh|#qwW%87aUNS+nr~XyC9;L*
zm!?qiZOG*{i4A-xHwfO<gNz5=EzFeP71Q~WpZKhI3@u;PZ|5~cHNmtT_6j}}H2p3U
zwQlN`n9OR6?Apd@hMztLU(h`Nh(48O04E6t=V}F+PoYsOP#a7TnZ*`Ecl(3fFp|V}
z0sUNBAtno*{nFT0C{=WiUWgOOoc;`TWDXp6aWg#w7CI-^b^k5FSGu9SsR4H$OZvdj
z2Gksj+dcUuLla$iAy}c$MW!_q_FkAy#l3vx?TGetwoN1bX8&*nW;niTLnk}t=!V8^
zj6#MXoXDsZO2&U_AYf{(tKqr&o94sJ`B5QynY-xm#)n7;wu!KBdq)MkZ^*)uu&?lb
zJ8C9g^KNKuo~ZIxQu?;WEv+36UQunUb6S?TS?TTwoD)x4>!`Vtt*DR`H=xl~MHua)
zjclxdl=Vg$?>RD}e!OVhGB#LLCEvK=jas05EJ%9=ooMU^JzWb74;X@G{x-jqObfoO
zkAWuC4a=TN^r|w4UdPi4-96hBROeHOlx2Tn&?h?Glv2hoPJ}m=>Ge`YGH};6{8`@p
zWgRGtRVzKu#wf4P{xq~+b~+6P_a~r|J1%=ifC0WsX~Af7qlXohPen|On@U2EkGKUt
zXx8t387Y-5mwGRvIh44P+xaWo?{9?pX@UK&g|ueSqnj6CGJ=?`^Xj;Ws~2MQ@wY)K
zmIFW?V&b0*Q5^eR*2&InN+@l-O^gB-_P^DyG}Rx*X}}3qQ&eb(;3&RO2q@Q+T&;a4
zwk6qjOVn0K5HcP0U@Ix>_%0J6gl)CyQoJ#ho$)9+$$Iv3E?4}jt}vA_k;By93lEvN
zW}guA-H7s;v3V30h3rf!p5t9b_Dp0T^^jjiBYo4~VF19z-+T3LjTzQLh3w4?;1xH;
z!HDhf*;FKEggmy@Mf8VhLofq4gxr~O;A#>DKsqB25b<+pH_E{=t|1}Tui0Z&%Qond
z)wQ2qE*m;|!KVZP4~_i!J$2$7DY@|FQc<LkG!BL}*HFyEi0g0k4`upzY1e%pXXmlh
z9A?RglB(MSlf2xiA2t3+kIQ63uUyS$Wluf4Gy#{EVP`fpcw;Xvv~YjE6?dm-dHQqv
zD<{T7mI_ZYGc8R{3V&kLArbkG@i-yJC27*%;guAW@rV8%-lO5+*TXUzY@8^N!{BL*
z*T5ha5?egt)NTxYLlQ2lJqkPJpSV%Oy^He&sOW`AJe%J1<xQ<ltl)bIXG4V&(o`5}
z0uWFJ5wbb#w{}L#x_oaD3O95Ahfy~7e_+7>;9@19fpkKb=YPF2Q9W-b<6ah=YA54S
zAzdKH)$FK#vTLit425C`vhb6s*4ht>1%?-H+Tu}Zi*@d8-~a=>D3{OiVrWHOy~nvQ
z!N#G1(C%wdnrdgbpAu#4+3HNSMA-ClUt|q^p-|xsg)y3?*P2F1APe_jmQn^EjmH!l
zVmz15$HQ%>eA~msWlll^s%QM1)%^S+N4A@LOx=at2J&LXe-Xj-n@R$JGGaPEJ-oPg
zQqz7AW_YF!#cG2%i%;4kEUB;r=OueQ(uZvmNdVg29w)stbLYnnT#TLCHAl*K_@(D#
za-Htbn=^<9l-nkauHm^qy}wf+6YeriR6ZmakcqOsk$k(7;C6TDw`<Cb$-~bdTD}GC
zjSIw13BYL!4TT88wAp9#3PVC5V4$M%Np`Q;-)ZwXf&TOFiz<m4^BtF<xecRNW8EaR
zU#=Gp_K%KH7<+TU*_oRbBYC5U`=08p$s4YhuBF@*tJEA+7NC+@$TLgg?`%P&&JE!C
zE2h5@{y)<=;&n&O$yl}-KQB=%u2OXjAZmSCwB!28uSYze2AdZdZNrD<tJ7W!TZnX`
zhqE99$lU=)#-DV(On<KZIWGGfrHEa3UiR?^4T9*13Mht!zG~Sl>&N%Vuh$`0=pDt@
z5vBSRY;8kv`KzO~&Ybv{V;`e$l2{&YsYbd{SzS;5tSDKm?>~#Yl-TfNtVlRFR*49w
z0$hdrv@ZGwC&how^VT=B<!T6ouN?{u6OSjp9a-`OL{wC@Gz%iVz#=T)`6c$#*MOkL
ziXmxcoev7L9wWf9#8)&q^xB+OOP4iH20*aguVQaoFu#<Spt!G%M?81_yHRrbc5253
zdo2piAKYTBEV=Npi@{LOupe%;E=KLCo>nw+q|?@=SgC}fQWJ2;Z}DOt^AU$^d?%q9
z8>hg^5JEHRr@`##q|1G!U&P;|`9qp8Hu>3KdAz2)%<<gTiBv*{kUljnDRYdJ@?<E?
z?VDJ7s-0$locudHtJw0dTTcOwDmYIOxy};`7@g#-2t^ZesRUhVtw;4Ui}WjB92Eu2
zPpjm-?v?bJ+whIDIy$F%Tb~jF+7du2F}>R(V&8(#4cmd&-b|L<jH-xcVIX?##QvP@
zHPZ)Bf7RiZbTRJZn_URqLq*~yw!sMRG~9$-?hXR-HF&BdCgLXmNsdDG9y*%-W)}Wv
z98_&D^^t_p-q=5+p(&hCwm&X%H$=zR6Vt3d6gMk@FV#2P3YQ>)l{?uMdxkZzRw!?l
z#Zm$zdqQ!v_#~{O<ExT$=hvniQ|0J4iVH;9W)W~vohqYs5@uDqoa=dx`S2vQ(Fa55
zZ@}>tgv9zpL{4OT)t6eGup0CBeXn8r39n&`=-KF9O^&OP<7qUW;vRG7FarI_O3u#e
zS7%H9XQuBEy%npr#fNCi{x6hJ>qin0K&tU0^OoqMim@z?75?6dy?Q&h!YJS`GM@Ka
zhRn0?f@T69P)Y3T8Z7EKt@yOkRtrHUnM922yj1Htt=goKb3!;z-m6dFMIP?ySe2u$
z{Oz46XEpq%n1Lm5o7TC)bzA4+)WR}^SB)bKZAh0MipD&+_bsV{^SF{))?xl=<3?bR
zj<lbhPE?Wa8sf1czdjdeCRfmgCl}NW&b%rbGH&b)kEz_Ok1hHr8q{Q&vjiUwtgR20
zQeOL9FBTJbD>B6$ZJ^Y~OM95rGPCU(H*9Y>TF6~N%uu5@Fq1dH*SP~s(f~0f4T@1s
z?QYkQhs`|7jgncpYAZ4*b_Bb&DCW?Q3O&HRu=cqc<+PLeQ=nU+1IHV%{Lmi2y~>7h
z$W=b+^F$?L$=Et)4xeU~V6QCg^O}aHJ4)KJnUe-3S4m7{3u`Kka!dc9_EB%Emn8um
zxVi8P3yruJuS2{J3H{?Hi;fXOyrf*wC1FR#t7TdxVTCmeK&cQ??I4XgaND?Bfg4N6
zH{uSzQ=HGvf(f0`mNwa8hvJS=lt&0ZvRmj7thg%dCa|$!QlBJ_`sCY|xEVk6a_G@k
zT4ljeRq*6n1Z+EOuu?DrB}$qG75n7R%`5L`u~=7#%wAh??cKZ6W4!rA)Df#y61~#u
zDfQLBe!A+O<<_WO-?Vj|Cl2u;(rRsY8TfW)7(6{svhEdxt_`A92e*w<=tnbnRs)}J
z+jRg1l1c7*I?f6%9cSi{yeAZCHNQS3^p3vIUQweROFI<kk$T@P@3+4N$HzX35y6?x
zJe#FMitb8r*xtDw9zykg@uHWL;X`sy5931kW-|!kV7}s_C;^@~so_@BO6912g5c3v
zldDC6A!KD7!JVl^JNU~niyawWfN~=QiTl(ecBq?Jp%ZRW_$rJ&?Q^bUQnt9pdkb?-
zhjS;z9x(iIB<U9pAqXf>N%|v#PzI>+sj;YW5vf@A5?Bf8sX<fML;7kit1^fIo2%Gv
zA@GKyA7t(LK3?-pc5<!J+>%~}=6Q0=DJE!BDxOdPU9x79ED(;dF`bsDEa)F9UDcYd
z>^wmy|Gcq#x1~RR3tK0bK>d!&0nZUL5Nmjh){kH5_KVexKlK?Bz=aJrt!71$0+$gh
zcgHd<xzrOADas?#n>9Ka=%m_Eu}qSq<YANjz>^}id7#fb8Fj7eR?g6~YxWvHJBv}W
z_c20{FolZUQAATxSGP1HS1^oE7te|_oPr&#W>2`A!+}+D>YQmQBcb#=l{;ERKQCNV
zBT8RTuT)5}FXmWT61ecFcZmri;ul(3yYtsmBhjZ%3F<`&fpY-p)qQ{y`W@i}BoZS{
zaA!md_<bV~6<Oa$FG2~OkmSVw^auXG2ci6Bq@Q#6f83z|FE-=fFos}c<^4(|Pg`g!
z<WrULDD`{!LvT?T)3k!N8mB?8)2X+PNpsRD(@UpW{%HHumWVGi$fi7_;pzTSu@bD=
zS%|Y-ZHJoDHsFusTd1tez)JJV@@0`QRykr!yn0W&Q}PXT*drPosusK!ZbIvKN{<_z
zn!=ws%0=)`VQDO)N0L^2Zs!V&?zCZ@-<R2`?`OANWLBmaD6n=s>uDaU!8x=iB%dXA
zc_iyVw-Q<Wf<7S(L43KBq1{0Zm1b`$gr&V`(zU8D<ql-1r{RJX#RbPmM2YR<q#Ezj
z`8r;Jrh^q!G@-uwZ3KaS9(bmX)=^oj7+Pg}&pB6PatfutSQng#hn;s?ooM2>fJ>Fr
zOdX2?*U&WQkukQ?Z18zfY-$m4AoGneuAxZr&0}hjU&oB4wN$8h5M^IgIC)WneOvxW
zqE)p*zd$HV^bNhWg`k+e0BvC4<G1)%;{$UfS48}mdV(z<%=dde)0_IA5SpbRq{^BJ
z0PD2lJuvnsy7mnuZm<CQov(~uVXXcxf8h>PaQ5f53rqH~kGMI-p5pra@KR}o(e?{^
zoTVdI8eh4_#1U4atgdc6k4%X(S^aNB?5OR9dNkEM!f14%ea?$%6uo<;t~GO#R`ERT
ze9KsDU+B*lNF49CTU-;zSg39G*Pk;ZcPEZfsV+*Cn!nWUy0$_^VIX-|sQ%2h&%;HN
zJN4<F%#jM^I-4%=&Y>%5dse}UlB>xHA!0JZ03^YTX#CKpwjlF~zZ+o9pVFqVhLhl5
z^{@8Bp=rWv_|hqrrN}r1PGhn@(k@;@xlgsrE#(i!dnACxV3jR0iq6V%hdiW_NlrnV
zv~CTri*c9<uecKH3z$j)EL{Tl%6%?chgPGM)duKv)4{N$cV}sjv4GRRb}}D7`_d#%
zVCJUR&L#Qpv``99rp(Q;9cPn`<`TUX=)OVHn*A=2r${%k9kO0___hUl!1;LkBqJw0
zoO{OTdj~8Q)1WRRI1sS1V>p&fQ_+#mLuVl{Wh02@-~2}O8ig+zRxg=6^J~@18!jdc
z-PMTs3cfvXI<&@YwhwGC85P~;0H6${j;Sy(_hGm!`e;7inI2#~H10^!k%z}8#NVYT
zA?84)=%{pTL{r;CiY8yf>x%^Eb!=Me{~`zRNaBF#joCc5Ch!lHgM+6KE3Mz=I8OgS
z0QYLI-cNH{(bXKAW_N<~N9@D$m~ml|oYiCNiH9!A26yU?xKuqj$9d`MXSQgy_Ra9A
zMMX4~d(S;;KfOK3<(@%6mfP|^CFe!yc61QsB*m6gBW4LBvG5!YNi&x#iHBQ;*KG|P
zfErVGS56NG!!h+qZk%nzNFrtVG7F}nDQ=W$&Xa3oax=5Tq^m4YtR_NBb6|DP$MRm7
zUQlJiD^5Qv%43YzWlN-I(4zNt%2%d9rB^o;c>kvKvcT_(+$=xJohn$cc%K(EVUg#8
zyz;lGqm-0W5T74iL;cj+$f9P_OjzMuV-^+(3jeyL%G@axs~_E_wh`(%1-se3&Z|?C
znygT&f9gnZtBdl^C<3H!dXvcB?GmOe#+Uuvs^KEJ8V9tFp_DYDInSW)L6-&LHIkDQ
zQ??D1gsd|j-^m=j`(c`4&44EUPfjqI!DA|gLk8yip9vyC@L$waDf3)IU`=tONv@%*
zJooqrHunlR#+}@05jCcJUJ~6%*k)Ycxp@uUvOft_w0QDvPbKOx{j?6x+1s*z-4diu
zn9dNBrC5G9R<ZE6!@$-o%o4W4n755}huL_zw?-SzG}gny;%<Li6WJy-&9!t}Whn7t
zyj+Z(a@SYswAtxTq?NIkMo2SzZZ`e=hIjsMtQVE$=eis*WqiNugY#`gNA9jxRVq%_
zZ^lt$qQ-UL8>P3c)D1y#Hq}Ja<VmX}fqu)vomZ+}^eMR>-S>~C?vtF5M~Tw9vsE#s
zSxlo8yiid>@j&~IgPYHgpVi>sa`#0?4Yu}J7`#fDcrnnJQSMh&w<{@}K+|n%tNuT0
zKaA~n(mc-zixXI;e>d)j|J64HU`Uh-rkL9T@<G&E*gu0_0q^<Ug?38$T{4}le&b$4
zK;culLoz3t_pwF9gi0W(fa)MrO?F6j_{DjE!(6WSgZPDTTG<IwGHBS6Yk*X)o8ZNd
zr${mu>Ft?~Hm2xJPuRD=Zbu?2W!tUQ4vRIZfIphJgHFkDxIo`~Bft23^wCd!A-1h(
z@c08NE^Z11F<~t9K5b397=n<b>+9sjL^}uU7X6YU>Ml(|2?kttej35lrav=GxoVd}
zTkChViHXyP7%!iQ3q$(m8m)V(UH;IP8fV)ca?l(~-TUxhRYKKEAS|5&p?px3Ixskq
zv?xnGGgUal0ePO9ZxK3#n)5H@ngG7})yycky;eXGWu#ehzs3Hp>G*@JAoYahzg}aZ
zy{m;GZA;fnW2FwSZ#JGBqbV+;c-_QGk_uVJ(5lh&^N=55w%^DTkBg$pr_$sBI@XR}
z86171vEs3?Mb$be3+yx_-Y@V;xg$7}&jBf<WmTZq7PqKU^(yL_*Vlb@$E1X#P^Qy$
zh~R2^r4Tqg+5Qgh=LV|L*8eBm+^0L~@vnApgtCieVHiGwjbn_OiKu82lBWE%oQ+y{
zOLV{95Ey{7R?e@xWii8sdeY<NZ~>Bq<qr!4det;cUg#)3g>9(qLAugJXQ~lcIsnTO
z5g)u`kZK6CvT+DBmurA|1ADt7Mzt(90+|RuLSe0G4Dqk53OgBecL8cP-3(hgx#6)?
zrveYcC;B4&wEQ`O9lTtQkc7`c0kv=WH5(th==GjgkmAV^@@YQoy{gTr|MRSQ9XtPL
zTdMd@ho-z6PP543qHOA8q<UrI%RyN-)0C0RXSs4K2<`$>6V{=ac|mWT8im}CUFwjy
zOX6T#lbT*hHWe?OZ?N)P`9*1?x8zoKKg)><6u#tbK!Ck4v!jI%cW~sIGUn@q-B_8d
z1`UF6W5GjsxbI3}ZA>^`oWcmeDI*hKXd83!|G}hxU{f62kpPzdDb@aOkAD~+LKUBC
z*(O`Yy?^u}CF%PzeD(-<Lh+Mm9joXs_}M~0)SM}cDX|Rs{%oPro3_FB21wG1yDqei
zv@s%J3@5LUaP_fL_@j2GFqaAQmDxUd{LSl#CaL6yx8;1L_JuSeW<{t31rcNQaut<E
z-ivh8$wI9Tlh%-98_zWEw=)x4ZxVfk77An-L=2a3o<J;`Kh#P52ls9QCb=EFpR}~y
zHCLmv4(N5OS>O-7MWu_$&C4gp4ZxD}gs05A6?tbXk5Opn<GhsnV&B@;$I8qTf7pNP
zk1u-$SrIV!Q_{^?N&CW)zASm*3&>JRSjA(=-%XmJdp~ajj;hJ)?K+}eraqe$(58;o
zISQpRJB4lciC}<xRih5UDtXN(z7pm@5zAS{oBfSe1t+`y0!H>ud@22LW?qxwYo^Mj
z3EjP?8%=~!YQ@TF*~1OPN_bJqtHjNfBMF>dd(}<@v1`o05kpzS1L2Fy@s}56)r6VY
z{KV8$z5SY@x+g8A{!L=@A~bx5yu^M2tA2M&ypb7~E}JvRLt-NZ_JgZ_{9m0V8VBDp
ziTtJu5BTUKR)I;+?~g)+g$d2I&!)~{CDwIviz<7+oBq(N=h~_bg2W*eg<2geK9<-r
zNE2TWUArdGkN&LKBmR>&G<e4SA^qAk=swaAx6%SP2pAW57QC~|?`z8Xl@H*9@^aex
zkUF_uD|!O-P9-!?^T*bf3e%f1%wR+7-#kerm~8&?)9&Q@r3U)0I20Cui|5JsOc;Iq
z|8AFyhR)>6uuyxC-SqYHf06P^2dsP8x1u7OjHlEEiRS@s41t!tj+D`5D<(JRm{6VJ
zW+jRhRKj%C=lTZ1nsYoEdDN}<S9+NBH1P$P877!K8OYx)wC_Mj4JOn5K8OA57SGK-
zLB$25zMyQ$3T{|tdNrZXyAL{7D^?5)S7SsJw+SmJPG5b!lBS0h4|z<!+?@J%mAoFd
zoH_kTD4wa4Ad{(Km5d=J?+aEztc1DFTmg*37YtaF=^BqP+Oos4_U2z_0=<L+`A_3M
z+VnBeDWWW3!vS8YalvM~&bO|h`3Am~OZm#VSxp5n`i5NBm;1;QjHin1Q1Qef7Z|ol
z@nw{v)_f>Wj`NG$M>($G9Dm8#{3%S+cY!Isn^P|fryaX*{e{Ru4xEjPAzawcXksxk
zGS&rX?myq&tyT6WryY_@I@?AuGPKi_=6m0{VwSpYWHNO+R;oz}t4L!ZR!QwRgl^6a
zbeJeQF~o=HYK|P>O@i6kZbe6&y&9J>{d*#8#LM8E#(SFzD-)oqcw>1vd+bLwg6;7{
zFWo0M=ZH{{%0B}?@o?64y74Vi;%9rq03hgPGZFx^#YHv^vI%wY@)e<I&xn_I_)|&2
z4WnmK#YYP)E3CU-dFuD~=SvgXelw0l%dsZ8p4_<oxG}!+NuTEap6RDjG>Zc2VJ}+0
zygb-3RZ@I}98VhJah}i>(HglRI|P@1(+`8ooS>@2H63Xw4r71by2_V;uBoalGRitR
zakXy$md<|d_z%Gg(mF#O@)36LcHG^WcyxlysD1f>8!CQ5q%FgBh+_1yFgIVh+y{BZ
z2Pj!)XEmMYikrf%J_gxm-I~MMElJ@k^%YE}oH;%Uh(fIH--oU2pX0Zan(sA8=#a_!
z0ti#JF7+uCs%E}qTxQYC1#g4KE_9LhEVHj-!yF`>p(#Ud0bJFJ&PGA73*ffsFUx!M
zWlP-VkF{+PH?4{}*6OkS`$VaBF&XsWC{gPxTl0D#vZS2A=a@N_qhJZ8&5SXB?p7_2
zU8nS4>0JRhX3a#2263_Ow}%Cgz9BJQ$#I@mCunbvM#I}lUBgROcDIe_YE>K^D}M?1
z3)mW^9COCP-IXI-dRn-}=#M*~G~uyf&MIHJX)fM37BP_CFoMruW1x3c(Nb(}1oPBV
zI*pG{&Axr_dIYSjEKtQW%DEPHF*X>MvmIgVMQFJyIU;z%Ira5hz%!a=4DCq={^WiG
zA=bP(FGAd(-!Vu}PB%5-pY+ih!u5Wh+@y>Lk(RV4l5*g7317FzM@F<7Ls-=A&?>w6
zVKT{Ko1D9LQ<sstr2Y@>75G5X_{rD5i?igTj*h4q1>!F0hjU5C&w~GT>t$J;UQ=8}
zx8LrUAHPJ7#K-c`e=C~3PJ<TW)|4td^1~d&<Ygq!uxx+Rn?v`4{N5&Qsg)qK20ztX
zb4a=GXl-0h=WW(4^!WNSgmAf{S*Vtxd+L)J7>Og(!AYwN4Zp1-5q(;<!Ob@Ugn7~|
zej3g4j~ow7A|zV9sxzag7JeZTC7|nUjau-MgNcx89Ca5IZazqrUK5sN70r-Po{jA^
zkmM;-@qJ14OrMsZ*n^q}ey%#o84?X?)w;g_zB00N#rm^t_IPGuoMa)7XtJ@@oyObJ
z#gIZ1uR61-;`L{$3gZmC3!A>Qs(2f{Y{?p+$9I(q<ww~m5jVy|_ank*k<;%T6A_Z5
zucE9=@=ARwS^58vmO8jh6?QeMag5kWnof)Yf~F>mq-Dh5y{X~g)Oy1c)Z}tT+_qMA
z)bCwp{1|!$XE`IdNdz$DDbg04in8U?aOej=Cvl8RIQlJAlE)XIb&tPO)GD<)p|^@)
zw{7+(l(R*|w4c8q6vRlK)fLvbj11^vWNX4kOj?to9lPP-^}4-x86yLwx3on^-Ft88
zM#FOjjN+4W1#Vf+^2tkT67JqgFggPE{gvFaQnPz{r$oRkS$T>9$K-s0NY{q94l|$K
zs4m7N#;B!Rt-9rPN_I~!MybuGsyfjkI&?LOo<~InvfPhb&7Bf|bL_MUUqiyWU<8IV
z78bs2T1TRp#HSPQVqn;tB$XTy3tkBI9Yhe40~A_O5;*2M7+YuRWP!=)XLil9(N4<!
z0)2(v^rr7bVtyuGn)C!aPm0DIZ@`xHbB<l%&R#K_gN^L~Y6kzK>~695!&`EH0zvL{
zg(*Az8^G8*7zl3-=(^}uaO&h@65RI2NK`odme4~7h}?ONR?+U-D;*@MW>h8lc_QnM
zF1aR{(g`+PNQeW86%$l2u_JadLKa#Z`SP0_xo*_n2mwZQRQ#Kj%Ee2!|941A2S=eo
zhzQN`Xczp`arfUHxA6r8R*PVUiKaj5=h4=Ob1P*$+Rd3tc$kEyiL*-rs`IR!73uv&
zMuWj^a$>7f{%F%pkoV|l1gX3#c<f!Gb9M$ss0BO_G*aNFyhPXQNOk-^26ZIvzNA)L
zenP2QX|cu-=yN%5R4yusF6<$yoow5K>`F>Nc{!ge{CZNwY+ZsC-hZ0GpZ>7fA%Bg!
z<G9z$bu`NZm%8fY6@(g|G5TcU^bnEj+|4&*)OFr<l6A0-uK67$<nNW7S>4g~oUEY`
zEfV2vW4jq|GFc7r_l>9CGgMuo<{JFMyBB?gCWLp4_z&;(cZdS-ckPrC``4BEBZji<
zrbx?kE*$34K3%~R)Y10qf8!=voO`1Uud{Y=Y|pwT8ZXA7c~h~tu$?J$jb@2M!`(G4
z#r31}2V;@`L>rRSc-nAV%P-Frk9*{(N{2r8KLxpXdoy_LO1>!SEc%E%jGd3(GJa_(
z)M>u$EUGefDA{hg-?Xq(D`jbysJeZdK{@|#gX&aqCGq5@nAVj;N~dSepLCA6&RqcZ
z%*Cp~JoIrHLitQIG_ADc!bIwKBl1`bZ}SPb@ZmrdB(a!Gz3LeIm|~n0LP@(RlJw|s
zweruVI%G#D87Qtevh33yr#^o=9lr7^t$vgx)5&}x`33Vt+!EEJiC4M}bUtdoOCnUD
z0t(4BcZceo_ua!V@36VH4ti<R`v#&AcoklCw8aO^wDIhY$4_XIW9nBb!AJD51o!Z(
zGPycv>NF~tvB>A?K6$<BRH7`)jkLg!$EM{-CX2L^anR42Y!R&(YYU%vmz>UQq=7l_
zi&OPnUHu<UfpM?%t6IHSe;)bc43#QLUtgjl6vnDT%qoM_IW`#trS|Ms%e90p)qyVB
zMQI-_yaF^0rYO(LOVj-Tc?a{AM&FsJjv2E9)JX{gZ^EnPRjtD1U9dQjWV^&C7vnN{
zlc;2Ad15x)&_0W^{<XUDCjSiSGBqMTJE<*ZzB<m;0M_;La`t<jRPzA;bB?Uf4Z7f^
z9s8)4Pb%JeMG_>4$c1Fnl^jG3YeuWtvfEqx@N2===x*k)2waUrjVK&zNk!5>dky&T
z*2>PTYS|$qUZN1>T<(<QTx{GHN7qxuk2MqUj(ZdD-aVkZ{$@#{D5^<(`YGUp;KUc1
z$&0Z_cWVj7NgK7|9b*N7H-Y6BNl`BkANHRUmDD(@k5$J;c$dU=2_Bvnb@k$0bOTJC
zg^PyOe0m<-_J!(gbM^9hrV)$)nFSs}ctSKdFGYdon0ZXFWby4VmkTv0u2RClZG=E`
zj?DR=K2Kr0Vpj(8hfz-P<FG&7-3{W^1G59pFwCz&@n?9wCyuMVVkVz*9(K;ZI2+nB
z=BORVRk@;h1riE656(1=j)NPKzE+Kl_2n=iqr8USiG(U_Ft@J?A@<^DIBS<r_tFzz
z$d-zCg1NU%CcXq7T_cHWHO4jTtE0ADexs`a^5kEfj<MzS03WINo97{><p{I3#%{t=
z-y>{Xt>@o4o&ASFs|VIrf#wqhly@bsKo!biTGa}@z*aS_vb5lb3nn2aG=V!Pkf@fD
zDHEHu%P=doz%B#T71Kic-L0tyxcSSDl`woGXEFP<enUjP)_@<aN1MGe7q7kO>Z4Bq
zZcT5togZPdq!t)P>~>v^$OKdpS@ae(mUu5yynth;FyXpgU+OJ?(i@76S*F4Hk=pRA
z1&Y`hpYF&s+|n-(O6_5{x7DE<ipD)xx!IgMdpD}SiAyAc@19*r(c5Tnq05)1iHXiZ
z%?+eGxNB!Il^6rhw_Qx1O-t}wChtR|)Ma;nCn%#{UD8)A(0$O%`vA1%*30Lf23h89
zk`?RDL;<3QTs!&k@7QX{wcF$$Q&i;kFhye|aVx{I%ia`K_O5QQcr|oT|2WP8ziA4(
zU4@aoRHEo0X+K2rK$t|@dyNi$3&+>G4Q_7GnkG|}W~-lcl3Pbk)%Go<Z79UuXi5eP
zOnY#_7YtVr`ye5~yn|eNHHUnp4Gh!S_ZZb5a>J(n_{?qjo@b9>!7-+^8Md@tC7upB
z-#ba!*AaCODZ^8zlX5s%w1bb9avXdvqf^e+UI2$2#7*s;oJ4+=-uZA1ZF1YYH><j^
z<DyKcDD3(C#d~>t*qCCRzq=zFFFDg4=A_|K`iC=KL38+|^X~Ootb)M0DcYsIjU}#q
zr7eCT{kiQ0-nLY4re(kL`{mPT1LGqm#CDEv8gQUAeGCj`A@<5%@_SvXoBhi}&8P=M
zKAi)r(%R{Lb%=$A#u>p~Q*?L!b;8$N(s()v=$Mj>;laE9-PH2Xg)g%&e&ZrVI5O*y
ztgUIZ2$i9s*(u>cJhuB3IlgUCf%uRnB=@XuIKmhCV_%>JG(K?0sUi;=m!ySs!PvKj
zM|PgM%LT}l^Km+5-p$JQn)$mQuT%f@&-v&5qkO%o(5&_OpX%yfvww3FhB>X{EGNsX
zS?c7s5ghLwK-RQNm+Q#}xDwF0nTC3vDh)U`jOW!UM(bc7=l{~MLw<d-gJfr7+}-q%
z%KOO9ybC=A^@XzQ9tA3it>4Ax=BrhyPkG)*1F~qd4MAM0#(94y>@OM1d7&ITV`Xj>
z^v1i#a6)`nFO}I@5kk$J&-b9fBJXZY?3S)W0`DpYzi7cR=K(a_hLKKdf2=uTpH@0p
z{0F!JcWLddj0A1S;qF=vsL3gm`djN>mQA_mN*dT+tsXq&<w@RqOk3vEY*tQWAWtO)
z<C{E&SSR&7A%5zhO~fib$&K&JB%DOvp&)tHP<ta7ehV{cn)F@%$yol(!6H)S5Tmdn
z{ai^W4;ePc?WHT?nL)g<#j>FaRoAcJQ2hvh?-29sq#hUT#L%UV1&`P4?dZo7`P7>4
zqb3F;86V&NL%mI+ZR2v<pga)AcN)Z}lVgt#(8rm<`Pe3F_83lzVj7c&%9yoR!uHd_
zXvR^oBLTK*cs}p$4ro~m8TDQlzN<Rt>OEq-Y+sVh!nC2Gh+Ky(Z|!K*`d)esz0LqR
zLIlIR*Ew1y!GSsz5Xg)?ALOu6IJ05un!^&|u<eu%rjZ!=8e$HV*jJ5&47HbMTg*sP
zj7c=}uaB?vD&}x3-MeNV#P&yg|H^o9%J!s9)G9YZoclK5Vg$|59jT^^r+<H_!Q$6D
zC`VBU+a@RYmyODvGVbQ)?c{6dq6oy;15-J2t!jQ;6T##3haF#7rorkLu$~3K|D1r%
zXey)$k5vD`asTW|<a=6uefD30Z!N+lgB-cW|D6W^zpw8pz~MkHmcgN1uORMZ1h^pb
zjXbF{q^>8;P@k1TJl*wa70UnlCm)=&C+jO<C{PxrAhxI~4Q?`VpWWCpe0z;l%J266
zq8xZx<TW<;lRF(Wa=#W(Detsn*p6@`wa!M}G3o(e0GaLVkjVpVJ6gKG2%JY)o5l+4
zmlrtL@0gO2<X9waOLEW9qGev)!mCNfuF%X3#%62Xer-=Nl^VHPXVgPjF)AnT)A7=s
zdsg~hTG2lH^%2VGgHZ)p`=}iB%rzF5*zkiSU`yMHZKIHQ>+Q-=o2U>L(l(w?Nh!y8
zu0>jUHxm<deA8^teEo_zX$ie7G$c=pcz@n%$?hteO~>t^hfid}g}E``k&K-#6cY?;
zqc*4LSuerjTYmWbth$2cmWr*0GDYYSQXLUi0<@*(T8Vc2N)CCKzc8M=(5$n&K5p_6
z^DFnzX@s>gAGHx+bECTgY|Z0<i2`}awT1A^|6{^x`lr?B#uBAo(OJ*Rz=W^o0au(v
zgsBm|fFG#`JKg&&Cp=?_Xh8~9<A$xD{EE7LJ`SCCldI4YBOdRRm#VkMkNIy$%GXyF
z(%ze{aWdxHmtE`_*r9*)wzp-5FG32uUC-hXuq}8PLp8-`LVmW@qP;oG7}M7ym~uO5
zd+37NwBYb9JKgbkNc6V6{#$MSbblD3VY>Cth~Cn#83`yUj{U7%c+T|0g4G(wyBoG{
zoZgYj5s9cxm1NoJTvd5i@4pR-z7OTRgYSc_`ozW!SW^oC0Oz1Yy6Vmo6&)#E@Fv$$
zBP3aQ9>h{1vOIJwey}C2w@a7AHCUl*swbKrsJYzZESirQfR17LOTJV+6zHE>&!zm*
z-#ZG$P>P>V96qjP%;Rd@XTWfTdEL_}|FE~zfcE;7WV`j%pHOwO1ypTIG7tkfNB|5<
z0~`60D+Y(Vo4BQB(RJLti%hS}y^oNC8$AU)bx=7_XSmL{R$@&=q=|azjK>q=0<q*R
z<8RsDh+ls(bZvE-QZY?^)VEuSnN$72-9`t_Szl35GN;DJxB1GMRJ_e?rvP@U99^cP
zO(LGxM&L+btZF%-^fw8G3mV8UErS*?x(&IZ+|yeeA$h$y9;dwQqN<CZLK*ySt85g|
z9M(?r%8cxwT;=2=q##dn6SxHfS2Lz0NE~u3DW+9I4<b`wC=zoM5A*qF&1eS!%*tnV
zaC7b8kVWU^NUn&1Y*zZ)W8s=PQmK>=?W;TlpP;spN(&{?I3=&#qZ<l}bzTpvC!W-Q
zrrWaPW#x?~n68sFRqIsqg$+J>qT>Ts2lvQMCVngDyS$$-Il2IqpFL?0@nr^{26Va^
z`dr;T!kA|2X%5N6-8?7sar10g5_Sbz2@$S8D}J#XgE58OGhX!E96z93>KW+m?PI{p
zz(FvpA#WHD_Vs>nb9CM(^NQbxb~)?Mtr-k;Q}4FKPti5(TA06dT4@j%*|?ERjPcDI
zWuT)YWVGjUL&IDpjmz)ko=95k=Z<C5XSiaY*o=c{0x-<Urne=3oMM&Fo?J@Id3c?j
z#kZC|sWbZhIwYR)mLG1Ty%jcg;8_Tz=}<J=e%PZ59G85<3~MAk=~5GND?o<+z?zHd
z(WG?htxDOb%Gh7=jtxFY?mH)X-AS>(a@tI%VjR`75$7z!tQA6B`I_KF7(2K|8dQ>X
zYz{<M@)561c#mPCEiIwAd!EPk8ARtPAAM`j5fy(%JZKCHAzzXx(8IPo?cObeEYt`W
z%^B>Z!JFU(BLti|0LpLvSzxVV{OOE~p0||ME~H4bXr;jIC0BIV(^wdtCTz`rVYaHH
zuiGcOAE#=0Q{!Tmg;p>x3gm}fxcy1eUHfJnxDZ6tA%o={_5*g&0|v#pm<0QrjxiqU
z+TgO(GRc6y$&b0d_Ja@iLernx`)C$C#+WL>!?+csQr#p8-xG3TPc)Vi%s*!9MBNxs
zcOg}qZ@#oS8}6OwQ|CEPYyUVvzFvv$0c2Xrrc8q=GKmillXIUM`#F`_4S0Eq(z_B8
zsD=NrRQNkkmCg0ElT8QFP?elJy&*_xW)Nx<WwQ9IO_GpB!K5yFzbpY>8oqC$q|!F@
zve{Rse;a7%jW!<kVX|eOd@r;xe1JI{bCmZO<tYiRnzj`Uudfn$&f{UeTSsDg7OkpT
z1jr&uZ#6RcaTGf=|As;%E#oOia2VPcT44I)zSU(5$mrqTE2`MtiDli~YMZ5RF!_Ai
zI;d>B*OZMP6`<yHS1^!H)#gL7?GUF$ft+EzxA%CEkNiIJ4B7U?(P18M@K^At*9$sp
z9q*ts>DUx^U}!9k_tU76h8%gj;mO?TzgH62zvh+p@Hw_kXw4wk)b8g$RsvgK#a=pK
z=wrQc;YFNWhe@3-=bs4C2-R&o0<aE%t8H(ExpK|%*n9!pb;83(tD{dn*C^7R=>I%g
zhp2E=bn7mgtHT6{!#FzCmNFC2f25jcv+8cCNmBvWA;S367}Sp26W0BAGdY^U9(cWo
z#oyLZWp1z9cG9lv05q_i4wWi3pHWuR8VWP38C^XGW6=RP?q`KH`hzovviZRKu$r0?
z)n2K82hDS=i>RxUtK$-7G>LwO*KGpNA%fB)#Ni2O1tubI!OidT<hLp9Mo12@m%0ER
zUB-V>X~5`JLwX!gjbr$FdQ>EbKImHvE9x>mt}I`TFH(P88Js0YO}4b!(FYaulr%~Z
zK1M?W@l(H7`)As_H6G}6={k*1>ryj;%w*qDWb?+(`%YzbQMz}Jl}1|mz;$(ex4lN*
zA(SilEsJVQi#!vvVeH+Bpa(^Y_k7I1oFY-V22=%BQy^g}WqB*tfnCE<>#*Z~hFp_K
z^|;wlu-qww+Q{vXOJ{+y^gC6H=#H9_v}0r8KEczBdu}(UQoDo7CBIJ}<mk=~eWg{Z
z$gdMtX!9cSZ0~Lq&ZzwLsOfrcEVaVLQl1^6Yt+zWOUBCJUMFRHlJDtX(<hQ^l6++w
z)jvs45^>yb{+$w2n#(jV$b!>Ve<i@Aj?sUr0c%`22}GDZj6iM~mBn7%+GNZ6m3l!)
zqDyBH5ZC>GF8+NG%8tAKv-0r2&EWq{>GAD|DS|G^Y^s`}I-%egFF@|^PJS+Vic_E^
z8XDuHW&SR{BstoChSh@Vn(?igf7>&%?3{Wm5d`f$7ONP=wY5(p#&ACzb}fZ+gKWok
z932IkzrJ;Ne0q!}K1*GcDsBGaZW&vSaM>}QSt8NC6jQg9{SFcbb@;z@8#O=xdONji
zFLAIT1km$%rMqOMp11n)Q=gEqEhUpX+OsWd#p*+Q#<F`a%$T|MFt0pW{ADACRTprb
zU6aDoNr4Ju#8&kib%0N_)WKUmf=HJ1;-}Q5(S}B2?$ZoiHCU320BJaIv;8i`<0PER
zvvzkVN$t;I+MvN>jK54sPJO%0xKz06Vu*j4+avR}`+Y7uARj)fAwy;9<x0bA=wq=`
zw;VN0+ekB%RG4_@0=PSh3wtq2uuE!1iT6@>y!Hv8j}_=$OLgr+9%O0E1_Z6q?bB*e
zf?bO(4enAfY&R@2nJ?0&3gi&UI_gT|ViyZrTV5zt+`9l#&Kp55pS8t-EPp5tIoFS_
zm@)0V9y}5zRwzbN8(y=F{eq;uge!JBCmS4K5P$YL;~)z1YJ7$b-}e|td?ctys+hwl
z@crPizg&lM0<F=hh$lW(yMSfEB8=~H&RVdOk$|v&!~rdr=&oUErWmM|Ca@=}=bfck
z?D8Mhsv~TDJ(q{w=9Eu2<IPU=!=!VjMQrmH;?bE^*XWzr{3KxI=NB>B1EbiG>5WWH
zVHwuKKj>}fWq(|C9YcF)ehZTN3H&fFF^)G+@E)vHYlaE7Hk5He1H&QmC0}|hT|!%d
z#Wiszdo-ulW*;LY{&lPAf!gfoe*T}r97JnePuGXLVd*By7fZ%gm)xnu-_rW@mm-!~
z?`x#IXcbmSJV;aHHSRq;0Zh!Hn-}>aQ(RgJ4x%{1pnEcOffn(^qei>bAhu@smi`;x
z)>ys?vWEf#UC5GPEpMXn*v5m>;HjcQHC9$n>Og5jAnk`8{z?(uYiRyR_QuUPxWneg
zhZN^qz%mBl^fs4?hO2CI#L?aAM5;d4`?Xk#3zAeRpd$9X(DEk3*qlKIYkx$2qO>!o
zKgMrtQatczS6(@CdrT7+=7eafu2Gy)Y*PR$&yZvtjjZzz9(+r~A!#mwuymTD!Xvb0
zw4(z@pNNE#J&3NzmZy%LBEGnlgIxk5SC5MC5p!wPGvtz{(0A(nR75QVD_FJYYKFzx
z!Q~P5l9ImlWkS90a5m+PA9dvwLo4V2+S@mi1+_`;<62}Tk<3&}VWV!Pwr`1+=P!UY
z(x}T~7xf5+YL9ya1BvR_j%bze<Tcv5%aO5qP?Gtlm)p?K_}~o<Oj@Yq-%Nw+q#<7S
zH~2z7yh6~TxTSQvKNySl@Z)n>75U_n7M>KxrraD8S@4phx9U`CWE@mJY#tr0ojO`J
z_%?-Cfk|S-W5s)E3Q0$=Z$*oUqw=YuZp3uxg`bgr+fx_S3;*WQK*zy6T`mv`S76B5
zaDU#-X(BY)&vYHVV-RBY5`XK}SO-qm>d9OM#J+_y*yfcc)hT1HN(~cyoGavK+lP73
zv%-m+LmM-ROHe-4snCjQ%TN&etQV(1j|<diwwEpG%dJ!=j0ba+6?5>sc0&Vv#ra=_
zw~rhb+Hb4acWfA2DjnK+9l1^?tI7}(ICXyyDJIj0stjmN-Z>~JZyhC^6h*cHM6C|u
zc_qx7N=<*={bu~DBzGs^WAp2AgooLB&QB#QXn=)Rn%8T5V18}joVY95c)x3Q!|cIp
zV)L;KgtCHR<ORs7aASjsOaa^;HQh8Le1bVgtBX?5$}_H)?ELwp+btnsKOT4+`QiP+
z8U;l7UaFk3oU&1CsTuOS6i4P_Xt%xK_|ET2>AESLPn0O=m)|f=ZeRT0)l~ya&gQ&X
z++6V4ka*YTMrnIJ<CWbzf!@4y!PS`g;!VoKEko-72jhdYGKiW1cU#r7J7=uA`!A>`
znXlKbaHg%^Yv287eIp*(fBtZOdGN1Wgu<Ck&)@Kfpwl4x)g=CkaCebSZ%~mfjr9aq
zQ?@T8?b|{^nI2_I=S-lz@7PkVxyandPj9SYQav0AdrEY}Z8Jaf97sC0bZ6jh)E6L;
zDy(a$M0zsA?9Il8{{7kxegB`Xy~em-+mD`gK?U7-W<)Kt?_~2hLcUGtj6cB@bEH%m
zm$gDk0rpWD1dD|6<EP4}cfRBO?Dq95w0tHrtb#*0$GZ$-RxB`*Fif*deXv$V{HH1h
z=F6O8lU!%f)|4sTVf-}rB>B?BB|=kC-jz`E=hLVHpfF71pqsWzy3LMI<AO)Xli%v+
z(cwCDy5iqIYJFc1juxACkTe`->%>PKX@D*4S__lIMH4Ts+2}C584eJyuUE*VJ7ce-
z&s_`XQNJ%-bkn}#IxH@8ymSTIy$FL`t?mEZ^Sevs-sLc4(`~ixJ=8r<a9=G@IvDBj
zrjEZ?LFkgrZv&`Vk-0r6wbmZ5lkPzV3aH63m_Tay{20K6th;>*vjv=4gf|f|al{!$
z1}U|7d5v)4<e_giyPSQuzmw=jO3(kDz2=yP3z-u>)3BljGOue$&V3~&F+;CW5he=X
zWH%YctEyH-IIns7Q0ANZ^*@f<z?^WwH36EPNM1FP`+HT!f#-f}TUwUt(&|I8U&<Z>
z!VkJ%2@3j^ZSQQT-MkvCw0?DM`%;eVBH%dRo?(`M<tLW59_%;Er}ZhA-O3R{N%TH~
zeX#gQfHnRI$!EPy=pqJ(5$583S^a|IW#q^;p<x8n@tJ@yWPt?a!O>Ae^iO^*7Nu9F
zDQCs%_l-c+umutvQozdp5IX;y_Xy^2LT}$!I=TOGmp)`YH&XS@If$RS=9p?u;6t|k
zr#ZIX4|?}I(H}L&xZ`LA%~N7qJ&$X+5^H16ZJ$061DPUw*G#GC@+xN_=Hz4x5AIqE
z)C!jTIIU{&`8ZnMHeY`yUwphxirft=rW5mk1b)G{2rPCGKX^6I$~083dKTcyT$@@_
zk>X@D2n<=(iRVXILxk)FqigMFXvg0dDJ550JQRywMH^8xlyO-DSDY@!1T!*W#Sf{M
zHk1tv3s@4r6)ekg4@W7!g7w12c&U7Z>>fAjOas1qp3I6qSG1<S0+fy{>y_d9bj8yE
zv`KjFgqWauTZ0%m)<c3mcl(b;scm9NC$qouYM(_#ylo$dYR~925VchNmykri5lAcz
zLEd{~{!o}hq#rHwC6CrSB*xXyDvz)lfzWZ#pI#`{aWexcI411)_eERE0Mh)i<#-{D
zr%vg^0&DC~SDy+dGWpdZhukd5_*jb=2Fz*?)q}ka-^@Sg7ii@x5o6Hw>yu(I4Qe8x
z1*MwDiMIt3cEjtT^sj>?wD6$ed=OBKPCEsW>*sERzD_TaW3p(&9)1SR{_EDAqi$NX
z<ecXBwH;|}2fCxb68nTADAabgePZRT(|Ky#8e<EKpJl74sI4xx93I&1Ni^ZP_<VwL
zr10Rnk||K}^d=KAEn}Iz6A`eX1*PX2L*s-4TidoMgmvz&m^4*C^7$BlQlx*nR=rr6
zs;p()CziUO4Ni%*m8$%Upz-UwU@6AFt)>4sHyh@14(g}pKLZq;0hNrNs6VulfbO>c
z;tzBA?h8oRcmJ9_Zl@nB6n!NkjFkjM6rBD&!<fJnT4;#llQ@fe9q@XRi@sZ148c-l
zXtymYE=>8`p|8{iRCYvv|8<MvO1$>(O(xSLst>)f^YgyYr~O*dN{Qm-HYZ2M5GHWn
z8d8*~OhB?>#5_FSmGz5hO+*UN;B}DrONKoiKjM*~y7^0kaJ&V3-!9JBOog{&Lhc$m
zuvVYn=PX=te;^l@6Og8EI$O~DHQP^!L3yCIn<O%fVUMOaZf1F+N93BFx1HkM|6=Ps
zqnZl1a7`>&P*jj!6_6^ucLC|WBtU4=AxP-Gi3m!h2na~%y@Y@W0Ya5tLk}$!>Am+N
z-yFZWbMMUD`{N|6tRyS^IOjck?{_~h6r0#7cw83JPP+Xi<44z)8B<3!X=Az@fa-Fp
zo7S6`;h&h!d5glL?mSl{8HDT}uW-{W(HTC>>U<f+^!JYd7iPq0mm%W24_)=p`gU&X
zpYabmJpkYNZ5K?@RP}>t8hmUyeykxbQ&(HNxrS~`<S-B;Z0i%W`*b8usy{=RT*$YR
zzeABBh_km~4wq^cNn?FPDNNB;FKj0)V<y|jkf^q3%^cM1YnK3B5-X;WMj~LX)=LQJ
z-?xg(<@uzLXv_`hHA&KzAq8Jw7`WWkts?E`PsWkb4m)p~P(<}lB=L-0SqITCPn4CV
zbkzELsiu4Uh7>rHEG)#R75CLw{un7>tLuXHqZ@O2a+8CS`kV><s7iFm*`r%=|NGsv
z107S5<L)iBL&`t|iaE1B23qbMHg(1?O#Zkb6M><Zr<je#SU|VXghZ1qrH?+vnTXwM
z*c=IZa`{Z%D!<CaFmVwD!S8Gm%8;Plz#=>ATv0nQVIvbhv%gVarH5<KHmx20VUVAi
zAkyKfbMJR7vBF#CwBM<8>>9Kc<7!?G4DXp&EO68)VWhcmt!g96Uw<ed9`6*yx?=Te
z(T3|>ioxsm5z)y$&`aXMZ+Z?L6jv-Qbf8-6+of2dl0)%qYrOFxZGe^yQ(IlF#a9M$
zZBk!56>93S6A<~jD!jZ%{uSEQ7xm}DCZ$hgg(Br>KH`Rc#<>04gB$GqEIQx@=os}>
z<+BRf;EPSnY7Oj<(YRo|av7I^5@&6>LVEY2ZuNo;dF=+s#u`j{7PoX>f~*lMroshP
zrS@!T{<|=9<9pCLh};0ytq5tXIx~5+8(>zpNjod}U=o`_x`ejHvhF5&*<2WFi<gk{
zjNeAw&cjyAIB~9O&T^h?XIKX|5cQ-N-6pIJ@QVli_F8X4d5PADWH9{<cWDCuQ5INo
zf4;HF^%bFkYvhsha}rJ~ed8Kfy<3Q8qDcMjrOC9~YAQy`b~ce%+BCTb*(x?OqWGx)
z4OJC3xO7~iKf{8pWo_HhKpN!%0xDj>H>+j)cn-(-gZ+7pZv9uUqLeT|>i78cIW;=<
z16#C;PKQk2N*O>~{M}9pEA4Ww=KVjm&v3F?B$KUP`0nVcbbffbB4NO*k13)K&3f?a
z8%jT6cTH9plDb;rZY{PZbW6*!u6Tw6(#*_xDy#(llK?btv)&KH$&bGWet`5}$??2-
zVV!iP*~b46G|v5^PH<ybZ7K1vtOtj8^|oJv#K{aO?7MUA{<R8d-3%E7)7<JuvJ&O5
z#+9loz;7f%$ieZl$z!<AXp_|)5FS`4`+)AHb?f;vYD59!hgUM~UbP(-iNsC7A4aMv
zPS1N<W!o}X5OFR~!f>}*%lNT`&-~t<0)S>WFnO4#9#pz%EHaWN+wL@f>MG5uAmNrf
z;iCxabq;c;xR57M6~5sk?ZX>W8`#aBzxNO_bUgh_JuO*4vM^KdQ6a->o+<1?nQV{u
zQP=EW$N1nTfN57mDCs2vCxW&CQ<_IWBjN*a<ow%A7rKf+o&;_Ivbp|ia^AradvmTs
z*mLqM|9?Z8JO6UQ;RKuz`2Tr%{hr{x0?EMI!V9~c1b9)xOX?aP&ipN^3C7;i7$|xq
zX&?Y}BFOJDCK2UjUQqj#-r`kL>W4+?`E;K%4QjIh!<cKcdKKrfzq64w2W3UO?zrUF
z+RqPiZ{^OJsdSy8!ew%rj1HE!D`NL5Z6-#W$?~+%<=%f{V|{rPfVj}^I(yp+e`1tI
zQP%v+Z*WN|<g@ULBPv|=%PqZ4a0z*+^^R%br@gPxRF>AXCB){nBI~3<p~XV6`Jw2h
zYw3;%hRphEO1!MbJoE1ps#`O(E7WCL;uNwZth-z}xIBw06CSD$n8lyMD%k4CT^<TT
zS}oTgr!f{er^0CnAuUC%?87M(?xu%ok2~IJXpfBl!xQ6}2YY(Z=j?Ey2fm6um*cWu
zJm?$Tp3D-U40Q@LyP}Tw;O%1tM`d|zvfQ>zi>Tr*C&qb;#-AZ4Tgm93{UgtTnsanh
z>v2N(Ix@!TmW{(M?a(;StmZIe74PJt8&)srG%su49C^RxGS6G|aJwQU0=pK@Hagwl
zw4JkJ<q@@ZuV(vwe1Nx6+ikdtbjuEW2}S;@c-?Zd_xnZMC8d}Bz2<_4R$Q;$S^t1B
z>b`bE%X~5mjMnC&SK`+}BJmvySpN-;;;W+7gedr>g-%WN*n+YC6Dk`~E3VZ8DxZfQ
zxw}7&YR5ivO|CA~+r3W%Qc957J|xQc4i98O^yKZbjf$Sa%3qnu(->=MFHt0e?QT+_
z?gFtSI0_SOd%t9{q08W$x1fjTG*a#qa#G7fp>Ran+>11Wk|+RMK3tq}QE1^@2>MlT
zO83SM<5$Izev(e+H?=xksST+;afjIlBX#K}U`L&j1NhRMV!ZOqtUq`-UacFaAK#Ea
zW`uV?8ZXjNdmrnD9O7B&UM(`?6TsXrl$%4l$Wx`Q3JuEUgX(LDS38yGHXszIS8<H@
zSGV+MO*;FY|JfIOTaEeJc+BPSlfuiEHqzA1EIjJ<&oY_)QhFZdCx5eeB&_PpxMP!(
zuEvzz)Xbu3#TjYU(r_#KyllHT_Q5?d)-pVwZHMt!3`}G)4Oee`yNCg;8tnDE_=WZ1
zm5!Pj)+S*(?<iW%<*-a!XdC{3TWP+&F#GAynC4A695F}a<Ss!3SNk*8OHN5ZlW6@&
z4R2V15`VAEPueNzrl(Zs&B2V+?K-k6n7DTpN#|5U?oCMRrmIX@TWozHF&^o|1ld+!
zq49i8?hbwFl?gFxNq*E}oI;k#;SzkJbr7jT+2}CCj*po~8ALGQYl6uYxrua<uYE0(
zvU3RqqZs`gDt3=c&UC70NZILk-)&lWh^_vd3iBc#hVoz6mh!S(TGevGy!AO>ztBtZ
zJ}u5?M!%i3#d-)^7n{X*c9tn1`ePK<ozw>8>hI;OewHh!qy(|uHOxQPx&XP=HB4i*
zbtlH__MVyvc|1)BGpn*$Vj18(0KFKM&KcJnU-uuwFeD~qoB53w6{4UMW38sc8s`#m
zuBl&A6}(ncsz%TBe@S!pcPCbl`6oD6dg$+vE0`uYCp_5!(f1^Es&})ZeY;pZOH&K*
zK)ju3f^V)YuYWA~Q{I=5&^ns!$ll(H>VYbJ{<P1NL<UV6rF$~aPhcKG<+{&MX97(@
zEB-y>qaH&Z+V_qItOXxF2N^c?mYY&+X`5COJcXc&RqJV+KbLWYCQBuzNP<$9OE0EB
zaQ3`7E3&S2@*^3$?i)#eWso2g?v#oc3z?>H3VGIpuYMyjA)Ar%PC#rh1Ufl%lDUmZ
zx6u}OD09fTp)v7#Tkc9J<39xJyFOI)9H#|>%wBSOu%a4+xR+u@_!ZNUi;wZz_3-<>
z5o;@0g#_)_x2;m@KI-iXH$f~H^K55z1@DX^h#yRa8R+Ssj@m#7P^?f29kf)sppGaO
za5EeivFO$dTPX?a9=#b>b3q^nYaFOIMo<!nVj;^qoyUV-!t(yZ3Gl%bO&dX(_wPu>
z_6+j$A<r$7$(2G<cMtS*m9(`NZuVRTN=~*6S@cKGq1k$9Qjr&3J@D|KjU^M*9dFKy
zZkPTjvS|xixG)&;>V#60o!oE|AX%k3IVehZNIjRfbbDgN@*?BR6heGluBzswIfHU>
z#XsoXj`Ueisd}m1RB6iTgEoV55XCh`U5%<)8QLL>C%Kju^#FqHh>u=y=D09?)<ENv
zDQDQ0Hj!~-u^MmEB;j9?u{Z#5!V--6XIKD=S{=#Qe*rspz5t~vS^7Vd!bq*;F`#(B
zt?oDAwPig(+(4ugmjBjz0L3#xo&mI$V#45T;IqK?&;K%*7G8WzQ6y0RKUcY&=r%%A
z93tNrZ(S0z+-;QVzPqE_pC6Z6QOJg6!2P)#X6t*x*)6`6&q<f5F)=t##Vb36y90~M
zzZH5`o-M2hY^ntGid{&al&3U!`PV>p(;fMGQxgTOiy3ee>R>m+$%Q5{V}*iiX_W#=
z+jkF{qwJb^4)PbC6FiYG(+sM)(owz}OzZe!cBS&x*6;F<Qf8i=i>5U%-QaWDv*Jw6
zzLx9B8Rj)Gd<}(LW@jUpg;(_`6Tr1HQgZ4J<n?4ZAT0St>j{6lV4lA8FlyStvV}O*
zc+{V^myNt&x9)%`-Y0I$44kz=H!}PDkbV08`w8ZS&$9b*Hl8g~SB7$n_plA>8OFa1
zz;+c%^KFYf<y09&Q3-SP_$Xx>K3Rb>A4B5(PLfr#=Ahd9)lL;-c}7bl2~~`<M_V`K
ztPMp;*t=nVgw-drBNe>gloGm~a(>#VliknWGV)*MtZw%lghW0R{MZ;dt)x3#HzE5%
z){%L%(aVsf3PnZd13MY0dR30#A8?Bb7$+4`ywYj0QfT2$J1uK&BiT-HB-oYxQD&xZ
zyX;omTJr9%Q}+JKNKm=m9zvoOx;{rbHgKR6e@b~ccuzOsEX<)r|D-&t>DF~%klKSP
zGl+Ag^WjqM`{0I&ALp4GCP<$|cZEW%xko|kWze+Br}hojKfQifTTa~~jPbmDLs4If
z;4yq)Q~W_tn<F*WAt>`q!*{jG<DBcD0*0VxHcZr8f-z-iFI`Ryp9(4m5pBCJG;Qd&
zl&7ISCyE<qIx0M;8P<!0IB3iLE+IAAGtHhg6PTjM*vcj<NdHjj(O_pt>@HSKcPous
zyIg7T7SZc=r+YkcCwD$t^>LQ1VZKwK^MR>9L$UyK;`g%KQ8qW|wrjpdQHUOeWX~47
z35<|z%yAvt4Xnn0kPqT$ey8NxRqmg`I2W`Voan%6kX{I%o)EKTYDDZl;s3p%r}{MI
zJEISbP==hSOi>$_8zOaOQ0@n{S>AnB<y3pC3*f&Wd>3b1DLH_YoxGr<sl|YZ$o}9I
z-QycqjiV#BaX;|%oFDBoq3N&IsDu(P6K#ivKWZln``^ZNSOY*f8Nm*VnkQI9mBJA2
z>+9NLu=LvPoUq;1GP&rf4oTD2<pbWnkAj92vdK)@(XshPN`GgasD5AwR?}+t-a;L0
zxj;@Sk3o1ag{^rfk}G1{5hIKkLiK$Wnf5^hL5q3{@JV(pOIsiS6T#a#h2$@%-Z&|V
zyGC1yysi}*#gIbQd0)HxkA7w1QRJESnQH6maQLLSg8Wl_5pCFL)&zKJ(2uR8Yj+I1
z9!ll2Fns?gIoUO4<UcM|GW(?%O&jkKmIfy5EVX2z=wV1~WAkJ9&Qb56aY&5!z=E8j
z3(FbOw@A%=`$Vi;$U2(WO(dh;)l~MPDXj}*EFs`*!FsfZNHSqE%hHy~>4<Zbebl()
zOp5aN_2a7uN3m6j9Zyv<&gCfhpDN7Vz$b|#j+ZGG3mx9N$sX`L+F=G%_M9`Pk}DU=
z@Ck-|Re~YUcMauJCp{Ja>|u{I*E%2ixoKgKb<tsy_1j_y$a!p)x+vErm3*B+?v;)2
zN4aVtl>e%5&(~Q#`dhzIFEQzwQIY)hKK|&nzrV)b{)fQVQ@SKAild}Io!FbM{W0FC
zn2HI}50R*Gqs(Zsiv8hLV*h0jP4H%3n01r9M(>HktqF_GiRnr<Xk$?|<?UZ<P=^Oe
z8Y-9_1N1{tSeJjiU7L*?$cqEiw9z48c1i7_0`%u+M@k1XA-Sv$gu!{F{6)r<|3W-;
zbY?q}SGRUF+4k<Q1h8+Y>;B|cF?{PM)Egx5Exj&!5rIW`@yARPF~J<`9dV38&vTzZ
zpI<d4ifjUvxIfX$o}$@kS+BP<)$F%^@%F*XdC_XyP)aDlN2Lo7&|E=CJv@B<c4-<&
z6p9hnjn7fA6*oZZC8~qYN;G>{^Xxq15=$NzZUD(gVLQf0K{GY7yJJ5&RJW#)Dupm;
zKinJ0ce$GxFdbK=*4%*zA1_SQhH+M<lzvq7juNy(RAek06vAVPx8xKZgSEa7Fp5zP
zn&OhQzmp0OD}0;xiAAxnZY!`#x}B7)r1={h2F}WrjEEh}sy*&F3TK;k>zcFh&l@2F
zfPX@Gi7^DC*fVN)S!8}ZpYlNitoyW-#J#&jb%0Z`fq8f58u}%!`EHw?VAwh*u1G+&
zLTPpTzW?gWx`ckkTVI<s3>&Y>&J9_SZufG79uYk4p?x|wI+6V~u&iqx2Q+{Y{~aXD
zq-g(_B{n_|C}U!&vLNu3@=}q^pvJ2DUq$F$i{B=zcLeNxD(OZ5c{JBcy+(d?mm6TC
z3;h$>|4k_W_W<610d{x#$+h-YW8sy${n^Rnx=fI7@_6SnpZwJCMYxY$JNg1iYQIPZ
zeB-rwHC@GMxmNJ&T7?JCLTFl5Uj!!}YqxW)R*%ly39F$hr#kmps2S(wMN=*+yle{~
z+Re{{xq;5|a{S77r2XqGBET~RCm7Q-z+7=!KFCXHmOkMpbqYJVB))PqdsX(eAe3=0
zrT14X^m_kb@qr7g#@~ZDI7sF>I}$tDw*M(&9#Z)|L(lSZ0(qXhGNNVlm>!t0_}-+m
zP3^r-zB=M@!buswNa_d`BfP(5+_LT|<eovBm{&KG*4#fI4n6<)r!4#S^8rD4?gE)b
zwbD*&_=}yh0;lJ+<s|HzJ_OQyP@BG3>)~LA;?Ekj@9jO@V)sJAoh^DW+!Ww8{z0rU
z1EV#4298O3ufuGD*V<YBL88wyojdrs6nl+|FWXjtB|hKQsk#31oz=^6Y-?sEg!3>g
z-)eJbRpWV?qmWIFIL9;L(LZ2VLg8N|7yu~bxkkt)tF1?;Tp4%Lx^{Qinkl?Wk3J9s
z$9>Q)^#46eU85(I-2EQ{Pf@jG<pu{YhZX!`W}}wF_=c5Rj07%#f@DTUZc>6~K-FlR
zS069an+##U3+r8Ms4*Ss7gY1HUhVGiH~h$BqBBat#_S5Pz$qMy8P>Sa+9;do81Oio
zk+tnY*DPo7-jaFr{@&8rmq1!3<AKGSz7d7qzg(Y&Bje|!TjMX=wvV{NH#SBp$_d$L
zj`xM>?48zwd={Fg_0`9i8EhNk?1vJRzsNc~Vny3lhSy~7aR)9e%m@?_MEMMok4ndP
zr_e_JhrnjyGh6SmOlz7IZYq3f@R_+Ue9^JYw7$m5*COA`;$dJ%g%^xqqZ(rd9`qEj
zr!K=U%nWzD&+^*kx}&O{=-_W#K)H*nHBc$N#ViaSb`QG?@SV+~)e`SCf?{9+ab0(u
zExUwawC-iJQHVg-YYdgJn-0$aC%4tiR|@MQ09%~8-RzUlq3-l#WyT*A5rxuc%dAkf
zo;LCyt7tp2%QsS&oh`OD|68KwzO4+zwPPLd?2u3cZlljDrno$H=cJdfEa;JX6Hi>Z
zWXNZxI%UwojNc>ZiOo`?QnDGK&e_as$3^5Hm*z@&XIsQC!G)4E@r2YbMk?=_t|q;m
zsWm4yw4JWadI*;CT8#eQ!j>`t2f25V(~yaC^k(CwIp6B4lz<cdox+;1!e-&u&aol+
zES!pJaXVKHLuFqWnuRq`XioM`i`N%^-6&n&3bMZN?lKEVt;5tW_ow-uloQa3vod{f
za=-j+xJJr^lJtOweiOxh1<Ue8CKNI}^IKeiQEloVB7YgbEV?k)fLBFmD(G;>ySjRq
zgN)tlGgZU0`^yGK2JWx~RR5I06%(f<z;5OeJ>*7xOrqCNp3fZPfH_(1UcZ?t`?g&N
zlo~??!F`lFEPF(<1)k7w#yrxFXpfZH@l@i@2wbDg_R!R?5hq8(@;qOFy-;P7T{UWT
zk<9<L-n^8a=7^=g9Vk?-<LtgwE<dz8nAC6#c-xF~ZQ0O1GW3nKuATM!hi?8E=A8U#
zp3)R^@3ubr6((N%;-VSfhjnfvCXMe9DEpsT^4J6>s<`v3{${QHYuQt4cSAZ=R(8-(
zZ!g~AAxKth!Et#x0g_o`Ipe2~CQdY6O?P{wa8?%*fL44dV}bIbdamHL+<Qyh{J2Jk
z_jR*Z9mrT@dr-W1JO7BReQn(_RQESM=PR)tZ4B|<aK?xd6hGvhz@bK&YH3fZk6N==
zO<!rTd*k};su9?RaZP!Bqgh<*Nojq->_{F_Wkrd*{dj&#4akvdbmoC$^x<hCtf1#*
z-W^eW2kUBtDbeHUpS`hObR&FtIlbpzO8qqHy4z&W`#X0FVEHadDR?3BXHo`|(dx`Z
z9}eUcOjdwWQ{$|#d;X>_M(M)s4T;1UF;;B~3Z%HM-@bo2_EcX2UWScJqai?X*%j9z
zsyCrP<GCnKQCtR6&<o<+{x=TQb4hg5BfLM;9B{9=KTEW4bgo?U=rN9i(}?WVOrx6t
ze)SZ2#9@ooP1J@)Y&(<>m9uH!oC_4$3Z$W9A1;E3<=AH51&q-^R_yd})M9HjC{10J
zW}dxk?@%^(YU4zerRfqu2Ao@+O&Gj%5<<wt!6Js$jZ4u|rM~fkf{8@VN<28;dyG<D
zLyZaj3TB#4v>ZhZDxaaEo*$Q9us)iLE9zmBc}M-{EI+)ao5aKomKuP&9~b4kfgf*F
z|I^%NECni2@RPDwA5P;nWgJOA;fGDrhqTCs#g>`2g%l$n82&W{di3Ez*?$Hfwpl(_
zVP+?DHhqFi_+#-<xuAuJ#gtczYwWc%SjcoOR;}<A{YO6MIb~Z#GH@;+v7uk9lf7t*
zt7srgKri;coC&U6KubB20rNbU<jm`jaCX!~?{%Ozs~!|l20Xo4{ugWS|4(oKr>p+o
z5cZ$5KIa4DDv{U!$cz1-Bnkr%XF34z8oJ~7iv!qqV2Qq}EBc0m6CbLEjg(eAVt`i7
z3tSC968g3zo9<xe*W9l9c;3NoVE*;eCRg&gbr5}Xgi*0h?~^?}P45trMbOBa%i?4R
z<R~+o?{BkaI@#lkTIQdpWt$;wVyrtLqkUdH&0#>%d-YC{mn^d5&9V(f=9~K|_gL;1
zqJzH&Pfu%4dYTF2-W)x7P>ziy^Q&|kFt{xUhsUgQY=~IBX1j9I)DaWp_80y7Gdgx*
z`u>XPe+WJwpQE3WnY>YYYh-=yg?dV1Ry9%c>}9vsUnCELp15FJccc$8O+Bpg#tjkE
zO&e`BF8_#-xOpb#`&ER<+;jAYY6f}Z!I4}b6O{Ul?%-nx1lzP4X>=?pCETt;_bjsh
zc0nCx)(7^R-I|?Iv?e6q&et~SaM5q(yAB_*+9AtjzlzBt{DyW)=muwtknbe0zUFjK
ziPW4OhC@FN^-MM;m0Q{{JzP2YqT`l<&i)e3u))a4aX-r6P#m>uw6kP|jTjuSTm1~}
z)w*Sl?)8lgP=@yMy-i!IEv2rT$)&RrQ_*THX>735TsH`VaFfXAD~&OW0~Ujat}&%*
z$s}oBJ*CSI{%oP;^5%YTmbT24zrKF1Q}LRho4V|gBI}^Rmk%m?)s2uvoXc_R8(D!r
z&ZW`A%hNV8X8m#DVvw=gPm=E~4==~`AgP-Eiqs(MP{dSiC9&X3YTavfwSuC#uPXq=
zm$+-C8la>}`nG5Chyz%C3_!X{ItuOZq3~W<B92bkGbgkON+(Ji<`c#Wy?zsVf;1=j
zWTGY#|HunPTMO}84n$LA-a^+(Hd&~xh$cugs@hI}q8qQ`PmbeLrr-KLQZDOTlJfhc
ztOWa2^h(&xM_Hybk1sF&U108XnUJJ6$Vu^{a->R&){X<O4!@kNu*>bk4_)EDECG?5
z%1I1~!Q@=yq2GFL8H+(FSN^#|i*2q>ILj7l?l^108*5%XfaFJI^6nzytnpw<Xqi}n
z<kv?@n_OyA&bk{0%w|AGO|6k0nvz}yq1!nv&au!(L~`0Lco<k+Jw3WNGscZ&nL0er
zk6pPMBj{SnC&`6j)eJ<{jHaM|x`wPur$T=l?3v<4E<E(80&%T$LD2W@GUE5C)`TTQ
zlSgtQNEl|kk^i={R4zW8=pZx3rtp0W{`&ZpnBrM2ku^jP`Aig6-mDM4;uU$?-uslb
z+RG`c#6*f{nYZw{W{X&vV{aEq>3)ju>Yp8+hhoPi7LT`YaAYxUtGokF9-R;qbNLoZ
zJEw2wUW?l2N?u)~dq;G7B4DL;V83n*H<f8D=b;x^pF3BTE_A_8PzIk+idVeLDmXd3
zko<9TPE^V8uFboX(vV2TXI`>IleNP;gdM-vjoi#g3UX;He;1376>tAROVS3BgT#9-
zby?%db{}>G+Li@coL8340{oqvIS$5`sHJIx07HcW3*V*f8xJSZ7yxKtFNTsv%l?Ya
z0Gaq<nKw3Wg*Q)~AidA-i2Ze1?Ry%X!L)zHj#I`jFow9cTyxDx-C0ChP&po$Ptn&m
z9RoLCC0>@S@%9v4yr|EyQi&6BGt+*s$rt0h>9op2<KC%u<Vw0<8P(=1J|y95h&Rma
zL&c4uunmztNgKPm+FTN6-2pB^^(kitf(s;=5h0a<-WZNfyv&sTlKQ;??%$s6jC<H3
z+;9GNsQ1A|=*^DV0r^z?{tH>o6Y{hPZJIJZ1H;_ABT^V?rK&JVi@5b!j(B+*x~|J%
zSL<4HF_{yyg7{hgk6hOLQO!L3@Nx`*y(Bhq3Qs61@ww~VUGM=df>oL+{tp49k*0!W
zrc7tQtq93KL+ck7*5#j|6AjC~GO*^pey4YbX(Kjy{MXvIOWaoSOd3s3dbAOIRK-cg
zQ4|z8UAf$QR!sdkrO2jGP|)AtaYylo*|G7}WQJpa=KUf1hp^I)x`qoPjuepYcC=)*
zW2yI$XmFT%sBRh@yc#$CQLlFVqPRZfbE8D=ns<PXgK)$q5N|386$Lm^Vbo!dt>K|f
z(D+=J<cwpow<gfl99v>-mz2ihLO4O`j!q!ntwWP39kZoV6&gT8Q2t6kPHX!V86PwQ
z5UBVzu7yx;i9BRG9r~%O$jU&Rnp!`J8lde4d0X&$QS3<ZucY_Kr+ET-&77fI22*~=
zXUJs6wjB`b7P>9~tj8^)_bo7YRZRNIKp<$_;E-c2Hs`eDP_%DP&rtA%9bUJ)O|rzU
zW$O@#zs&R@`%KCK)W+~~w!#p#N~X}3KbIhHCF>eMX9F4uAuFn%4;bz@>c_>*{0+Ip
z<!p*9?FB)je&C~YC@`6Bq@7cbaVc*18)2%9DQ6pq;8-41k+KUcLOG!@>r5!p{iiQ)
zi{LrZ_~Wy@_T~`hAnbnSZ$5|9X9{$Z<jRWlTc6L__RF|wI$62uz*pE9rw#7=%2%&B
zZ)dPKb9@K>f8pa)!U*BtQs8>oq}n~*t}oUA^sSsF{F0_qZ__Kc3Q+E<)qmnX0~F5p
z`@4n}$w&d^*5;&dg|p`u`vV2HTEEc0uJiP=q$lvkN+ICkYr>`sTyw07*dz<k^Zz4>
zlLH9$&~K=G^JtYw>Yv)vl~Mq4V;RN;L$@?sQh=?-L@p313ph?b3isA3Hgu#6+q;PT
zwXKms><~588mI)0A21rT=YszUCMp_gkPC8@ok{a6q8)FPgp6X3qe`a+QGOX{7uL^3
z%Kl;RheWkjg1CkX_Yaw|$lV_8itsB#-?zT0tu-FcRM84d0VEa)b3$$|C<*qt?I~&G
zsYwS>lgNHhtNfx8f{B*l<oxQu=}qnaUOT-~ukYCfX)X1iCG^{+-Pn`>^)t$a*`fqU
zZQY&M75tY0LD0n3&=RPC_)I$~W%Ygk@qM8s?N9hb8P5`Pg&F&-vH(*71&Tt~23O|0
zH|U5y{Q_AMrQKC^)V&c^zQ^G|fuPNgtK(uzAqi1wXC?f63=9?Gqyitc3x#r4QuVy0
zN|e7zg72+4<<4H`y~Tqpq>ej`Bewb{c9LCQyK>RW*D_AIYu7N~0yJl)abXNTB8oZ;
z%?`cO?;GsNLmy`z$bh<8`ICS&19M5qFD<m;mX5y8{+oI}Hv9MU-&@*yXl)q3--T#@
zuYoKD>ar!D%3&&MSnKMyeveSaVl&V^=L3RulTUiaE8|5>k`*3Qt=Bx|Ckhr0w|6=1
z;nAaTq&DSaRP$;oqSWT6WaL~zw2hX*UJFOkPi{-AY#6DG9ruB*q`p-zVLmV)lpE|j
z|E{g_ovgRc_w)MnHE2aeB3$>l!~banPeaNy|KYOzN@;XlT&W}L)92ZeQ9PKtm*Xv!
zL@fs=knw{j@vvodo%Vv%Qz#F&*32z9Ls4zLN^g~^8Qo*6dU}Z&?E91jr}WAyZr>&N
zu4k&{tDI%e@~6+x6(^I%0bdGVMOgp2cL`E;ZpFq6n!Q8|a1c0m<=dN@SXN#dKEz|J
z6>E$xJg{CA?@8R9Qb*;xCxwS0#UbClN@5Vy>ixKpGQv)e&gn7Ldn+1rM12)CB(<Qe
zxoT$vQI>bO<BIeVAH{m#v%uNS>0*ew5nyWoCCc~I)9F`IN>4p@Tlod`UCjTjl`E@O
z`8Xh1C}Jh3x9ib#_$Og?7SED=$Eojm>O>J+U=q!@aY<KbiJ)!$FQStf__;#hTBWno
zrJ+|AfT&~@7l-)+Cz57N`XK5Kg@mUYia#f~-JoLVZbhf}VyVfPZaU6x{;rJ&$cjgu
zW~2tKRstlqhMLBJTbJ%t!15$j2Gzzhpsh{4E^9*hJO(Z*YhL!mv@fI(oiU@pq`8Lm
zlg3s370&)M@dsd8XRVOhXw6IWwN@?Guj{K6CH*LAm<@6`sWCC@F^1_^xzLK&V={_8
z#PWCXSgn|sqf#DiH+%<dp>~Hp=#^g?x;=jaXpYlm7hipj<kl_wrnFWhY^bw8Z7#?W
zbu>>NCL4wrSK?pxXD285zIYu|Tqm3<IGp~SCOB5`9?47Vd9<U^)MVWS7ZveI8Jp_Q
z4Cx}E0p)}q_Qd#J7$wYjm8H&ScY(KdNTG~G+rk~M4~mF~yvybn2HSoF3KFsK@B7Cn
zI-J7>RuaJ~H*!GYV5_(5J)K_aE_bey@CUt)8Cs(BdSrC8b&6(eZ5|@C-5})yQMHYQ
zVW!Ft-?H;=4p3%4{@j9l5BVRL=R}LOpF@5`j@b#9YM?R~QNosFEH!GeeUI444b3qW
zr>}0c)!Z&id%tb17udZF+kJe-{qm|ag#jX?$$~^zA|mttjw|{rXHY(VUI1*0&oT-#
z^QJHe9)bMCOm*%my*Ch4F?Lg%8D%NCeLucnlvgU*)A~@<6lYjc88nOO8u@xlq~bld
z8BV)<-g81yGGOAZ57e6{HRAXx53#cQ5Zp$m^A?^UvY9UkD_({P8Y~Z@xcjI~i6YXl
z2)j+5yPB-isuFfU>cjZEKDbSx+h=%GHOEQ~a0O#s&Fh!_OurGYUpe2%%(Ht;OQO$`
zKf9wcFMPLMa|}CHmFmFmM2w^L(B>h0i><Qzz$|5;$5(8E2u=1Pcxyt|L#00_xXH;%
zTXdWC9Wopk((3hqma%u<Y9DrW=*I<hm?Z2j7OOJ$r411(51OAq)BvPt#}3G*G(s;#
z3kLy7D-#>qbbu1J-cWBh<<S(2&YmHEu|G=a&`Vg`=&)vr15M(*b=rg_aqf8i>|$zj
zZCJ=$krAvoujup{hJ}b&g}(WZ#SR1d4j*^`0#LTIToCcMMOHjOEV@y+Ouh#fcMYB%
zDe(h2VocQh3H8$a<`etc3wTw&Cd6z03;hjK?Y444^~GoiwT7Z|u9)}$aZubJ*@yYj
zr5q~@h6RMuX#RLTbWB1kf(C+bCwuo4x{Jy_T-q`?zLF#3B)CH!tV~o(@AOb@O%47<
zWuAw#`e{*N_38#zB}wBi7q=08I+?C{<NzS1{|(cU{8u{b8E)PIYUh8h<$o{U0dE42
zw*f&d=q|GZ{$#0-Rr<HJ&JJz;atT<O3jb?k{{Qr=fm)j2{~qoF@4pB8tNd@!OGcLI
znrDaQ7IijNp{l*%jIceTyoYio>mdV0{y;gddc-(|ADl+PoCkMf7OvkI7Mrvp&mtTV
z1*!CZE7{)|U5*bSdv`s0({Zs$LKd_kt%68&;uQbyzHl&AM_}TWS)k15>VZuY2RbqE
z^Bdc)qVlRHp1@bN2vuHGsIvJ@;(_cAc`+^ai6*Bvzd7eO?~FWe;dS+-JX+;ss$SA%
zd+ZH$UDA=k7|(a-SWdXM7FXH<>WC-Z=PvBc>xb9Sn*`T6`kz-)@%DZ_0O_q=)I3m2
zu-@B`UK)u_>dJpKFF%(cI^Dn$#v)NI1^^&tQgUi72kvo6{`OHhtpShJh+QZ*$|~DM
z^Qee1*egtexePq%6H&u8qvWz-O<Z{+b|k8WYS$ORiETi7+Pb9Rk)dI}1;?nf4*@F8
z24`wtFu6(6U3SNdS~xUUYq+XZOZA4ceBrA3{Bl0*&;y>$y*A~1${HM;7}!qfMQjYV
zWWBB^nS0$x`TVuxQSiFhB~3@hy2hI>uJIb-06uJ?R-Wx2BDYNrktubLm@EHC8j;N)
zPE4JX`a0fka=z8CR9a8D@_;R={ri9nWVB9KP+>9#JuvU#kUP=9=tDfMr2dS`CVb00
zHhWR>Tu$1Me3^lmb-vq`lJ?E@*2Hm9*Dh_^(ltCxp<y(S(mO(g(K~C5Ov&x{8aZo(
zNH^TKyQkR8$!x;X<bw%QvT$@(;}R2A#2mv^A5G~POSOoR^!0H`Ajdg`Y;Lu-s~uua
zei#KrEPHREW9HDYMbnAPPn>e15}uQaER7@0ATS*jy3;+=s9m_SxK-4zxKZ-)$&w4d
zE@5mBb<8`1ASd@$*WrT%5Pg3MN=Rgf@>>96vPzc?(o@+c@EDG-Y7Yjg*|~|CW!o(W
z?-Qi%?!X+suH4JwrK{#0owvqJ?9z%_lZw9!O2Vo0^(05o(LDS=M6I=^IW4%MQrD9u
z^j-4vBAhBWr%;NkDSD2+dT)Leho8TII6WfQJ_EO6d~>7vD%sYXaMHV5Z?#9IR<rpl
zkC{r9A6x_#QH#C92|-1znUGzK6-ad|&>4~iL3Mh9q&O&54@E?ijp4m{R7)zNCow`%
z9gZ3HeQ9Hwe9$DvG1W=1bgNKm<Cc=r{LmON!)B}f+n_P)!FzKbbK)mq16+nZgSwg~
zA*uJt!_2Ua+40r&n65q$YsD+KP%}Z3^5ow|xoF}s-EP_U!~G%l4=H1sj@$B>U1zS&
z%JNHbhA#{l?HImeUK=<GBEsaNdTz)k&VAmt7l(Ez5Ejj*1va7azIuah;1}<}KjE-;
zQ0WEC8Ncz>xnQ5)B$(V{-ddO;{|Y5bV3oR%5?I}{waGOaWndFa<L<?wvMZ`LdY@V7
z^5kh6b>%A7o#Ew}%!H?|YI_hDvNpLd9DjvwndeRBhvl81HaahVj0HVlwZc7oSFy9F
zCtk9P8+j)oQZUb9uB;Ym;jCG!l4_DqplgQ*Eki3Le&C4>@~}G3dR~Dt)G}aPG63CI
z0kScy{&KRbjEiO=(!4(Q3goZ!cM8Bdrh46SO3WLWu`3azD9($|!mC(wWadjZL4c^H
zu@A966-662jzv?|L-M0;%>jGC->1#Sux;M#FJODt8Z(Y_jzBt~$n(`s$NFeVpFX(d
zd!3h6sYSCIM}&8%W=3gOP4#vQcC+r`JmU20?`{o_Wcd?B|H3_V<+PeC*{!-o8;HA`
zC<Dz)!6OJhYV2Ua36jlINWGvLkg>FE1Ymu_0~Moa1=8iLLPzx+tVlM$Ww4Cj1J5ez
zCwdPNisRk?L}fw?Md>+@Q>{(m#Vm?-R+Z>pMR78mvFP2FPUvzXSc%lV0j*VP3%-be
ziVO6GX#*+a1vS`AXnYRHB`F7?xc#6wT~of7=G!P&HH#_D(9p55@Q%Esr~G}ifr&AL
zq9ScdomErlx%mFENuG~DR|+lar5bp0w~5^Mi8O2PD+B5-S;(0gt)`Ksj#!7SuZya>
zuMX!evvgo`a88!GZoq6KAbgAOz7Wy{Ft*b-WH#!J2Ns0C{iHl(kXkX4CuWW|%LDCc
zC<C>cq4B@PU9v==y36_oXbJs~OAdf<_u`+d{I7#cax(8<1J{Ar2q+5&&RD!RA-0lU
z6x$%+yjEl?|2Jj5RsU}oyW{#lK1KQuL2Op_SdD$!VLAZ*c+WK};NT>e-ray?BF^-4
zJ1=~W@!T;_wa5wbNhIz?50Bq<Ne(8Q!^ES^?orqMbw%3GMb{mHl1s7Y9OmJLP-2%H
z$6uR#<=0hLnCmV3wvc_@=7;n6j=$M=Pk~r^y5JNPhQ3DdG;X?UN_=Bzt&kZJou}tY
zo2hnvc@7{L2xY2=XYju28aG36bu;~HIe+hQt)kR#h<Fdw4o_8=#+<po9Y=@>$fna3
z!&-WdmACsnc--PLHQ>&>70&t7LZkf^O32zuhL6qk4a~szv7AR2%=;lejh_vL42qbc
zlMgaaPY(dRFP7DeXKF&}*SGiI@N^AdPj<A1cI~k|T@#sAXFC-I8xna<jQjmR=T(Z#
z1M=UOELD4qH;Z0d9GK`IM4<cv0>r0HoB9v1?%a!yeoXZE(WtV_Hq@{r7)^g4R>~nb
zw?2!7rhvSwyy+Z}N7BYJ#f$e5AUmtyV8zZ79`4h4!wm!2kKL(yAcNVih|ebB?jP_I
zrCt#U-Y1(1suRvDw*3NLSkD9{XY`{_dQ$aajcyyJ#0?z_{29g)cJ+;<a-|!azgHK@
zJ0%fCP2pmB>T9y`T7L9bh6=<Ji3=yTui&!!rd1x#MomQLbQ4UHxYry+-F!f7hkNQj
zl!F%3YgZ4eMFz#*OZt@t>Fdc(kH}-Do<jTfDY9xMZSS^Efd+*sDrA$9^D1KEo!V;R
zAAvX|n7-FD;<K%q2FxQelC{^79ey$?L2FJb4`{*lQ)P)M4_y{IC2<^tF_P`Q$Wj;C
z09V;`Z9!N(cY}-AYU*MZE9;8b3y!tv>=BkLqiW*5dxPZPYb3~ERnq!xV5`o~7Xv1Q
z%EH?;rkU$MeZ*(uw#)^seOX3ZmYv-)TN^{v|IU5B`#VNSQL(pTcHLcQm}S|y)dwYi
z59!pzo4!O{n$7`t*pC<YB3Np2^h}MkN_Zzl-a^myP(4?IL$PO;M#L3XL#421CiB&<
zB+D8|R2<)_iCC<YW|fBL#eVlGXZ<DfE3C>>=(?F_=2?EDhZe=vf}vP9vZCGC`x`$P
zg7w~^BU+)$^Pr?)_V#wuK!1>*=T)1DpYZKwEs?I40A;J0J$$`vwESj@d}zA)Ozzq#
zcFZ|1zW;)EMMY-A+{vTuEJ3=|&Lv5AygI6^e}&vBUAX>fzL6~~H=@L~7S<>YJydPZ
z*Tn|wjcFeK8LbBLY^=TVdS-W2lc@nx%}5ueYY?=taZ`=&f>^7rM)3EaC)#S+_SOPD
zDffjF+%dxpV6-TC82bwbs3g*lI+q<7kfGEW_>eQ8@YXBb^r{c(tJy9ns~MW9c&CF_
zb<5xmv!#B%q3+6w(D~$l2u{`&DRZH5S~rOxRS3o<RxRg|TD@aD3=dE60>@i#7|4Ui
zoG6SoG45!u?*fY@-RQ@2+$OqdBcs`oUwSt5E(}W9n3{$+<jYnEp&73(e){Fb$sZT5
z$k)!l-s$8pz7OMom{sv&M-gDTqG&4H4&~)owV%vv+2mGK%I=()H~B741B2<2wfku*
zOaqGUQN@gb;E040z-mPclx(JEt!{}b>pqvbO7%htAWbPy)-JG2>1iW{#IgR=FUevy
zwDIW4UH*MHvuTl(<ywO%TZ)*Zc^L~~k<I?kx@PqrSO+fHhFOVfhE&$#3CHH)h4=X&
zw`tyRWb%uTILx9_!V3FC&=!1!#=bf&ZH*xjuWcQpttY+||NN-f!UOQO4vjbCnJ{19
zhq;xee)*)gpxpyW`KC`H|I7Z|(I?6dxjcG%rT5X7vnS}7-+#PMkw8B4J}`Q^l<qZ?
z$f&vnu=UfW60N+?01IylKeHV-?YL>Qp#ludfc~-VKurd?gISl@v}CI093Rscrxu(4
z=yGN<sUR0n3*v8r0?^UX4&~TPq0=?ADD#+(ouUeM1Iw|Pau&R|CHJ!)An||faL7s_
zF486!emgUyIoWF~lOm_~S0&}Me{`x)xyBBHi0Z`n2ihnu+>jIefk~VkNO(bR=&Q5@
zjFGB%FNTum>2GpA3<h^-vTjDWq#V8J{S~y0QxycymGq1Xe;YR5{L1Crvs$i31lV3K
zLB;^%;SPr?mDlegXn(Ku6&f@!fF?||>Cj67HSHR17@Dk>XaRVMP-9?uDf9=AEF?V-
zKb;k)sSdtP32-7Pr9*|8Ic$(~j6-V^E{oNXjF}C}YTqQ9!-I*Pt##n5AXb)*jFXeW
zkJlnU@$D1>1jpn(*F5LJh@hwk%+tsnY(PY0J}M2GTa-Tp1p4l#7L@Xr-+_-V7~cMo
z4{{|Gbvr4Ys@<6zi)Jc>pt$GHplK6?^&_<uFt*t<-x7_hMn}|3=x1!v>N{Z2;KePH
zZq_9RXzh~)z;Z+h)Oevse87-<SdJN}0RP`a{!e8F@^E$}2fheccX{Z;bVg)W1(<um
zKTn9qs%ES9MsoU@faFC6G+hmUYkqol=$0DxQ=}u8$W}YTHl8Bmkp0b8<4-3<!};)^
z9cdd;n!n$JE#~@j;Uj!5S>pL)yL>@{DtY`}A=%}$0G<-M_n8KoE%CP=ww>a#Vrq7_
zGrND&rstb+sPE9oYG6gX0ou+}2kbDTY%xnmWGVClf)|F}m;WKSCX;Qs^hX+XxcNRk
zBe<z8e|4aJO=fp<kbl!z9YJ+P?I$~sT)EpdIK`NKXsSqYryF^)V=Pe~<rz(`Zf?;=
z6jR#Iby(hEQEeaNwZ}*AFdYFT)q2mfUvKrGGr!KU9LQV<6T=SeVVEDG7y4DCGuU32
z>?`u11AL1_2=D_)j9(rfDW6Z=S+c-OF3f*Y)kn5@E}c8ycwBECKGm7vS!R^#*S^wF
zEtTBSawH(FjGKYoRC&m~kI3a&TR&@WKIaHdu(Y$>$A!&lvKhL+VayMtKd@f<92v|G
zJ(}N=Kx13J9Xw4rd~r&?R%zx}-~M9|Y0o%NS#ULZ7ZIs-$+Yxjr5J4t5xzJh`3qPE
zqMx0A+!VI8zbksCB#W{YL@JLX{W^cu3spcbFI6#ibN?Zj1BR0{T}ft*imCugX9KGj
zSF>~<rGwE4%Eo`-LkAD6R7dZ618lHnzAm7yYZ^MoXE+kZ6^-=rJWA4E|BZW8;}kLT
zrYN`8^0zEEF|)uL{GNiqj`^P}1Ggn_ZS2j7`YNda^q%G9*oSjNH&R|gr~F}9s(LNl
zs~73jrM8MPx(B#AD?e-dWi!6yh>G&eTj?tr^GlEBw%2rkNNW+!k|94c*No>6?T&g@
zh$u+>au;hs(nz;KT2DQ81N<a~CH_O8kgQ!7vB+xw>EyjmUS1gk7S;bB0^)AZSv9>M
zj{|5h<uiW~s-PRIowUpXX#NSZC5Udg_@bT&vxwLCqSGt72yA%17iDlnbLdz+uwFxm
zzJ0M_mM2Qg6n-nxys**LpT7R$L*k3fM#RQ3@-t1264f^?Vo-;bGt~0P1cOiP5~d8d
zwiJB#v0HFQ$2VykwZtFmN}j(E9<xz^KO{{9!yu`7Ugq?*5>ebV)vErN&$WMDDaAU{
z+$Q58fbKsfa4eWRWw|3Oq8ao&w|}@q$yGk;U^*Ak;eI6O1hyL$!mB9kxJ3n%=jRlr
zwXS^boegqgG%*~rt0oOv*f0|bhg#PdsqF(P+1kFR>^duOyVG=vm6kH3HoEo0z3o$a
zE{1`j3n^<aPlUSP7qaW9MkkJ|^-b-#$<c`UM+u?&&VL6RRq0q23BCzx8=diRWqu#-
zE*M|X&)1Eg93NH2xap`qIL+d<X57wli($8Q0{8nuN3Qsb8Y(sn85S{C<01T_VWEGg
zan(V6Y2zCFB?lRt%kLFjmALe&VbC(eM)kVlHa=;;;>12ZCds7ica#n<Mt*`5P<D!R
z2T)UzW&b0H(*xTr?8DS;&cE9ym!PO?Op&gJQ=GQ(oflM6Bi3xu5i=vJInlH*j!7Z%
ztkq~pqBUp3-)ZkrU|V0GT1ZWD3?Mn?Ytta|?A@`<^2Fl7(0D@b@e~2?x02fVZ-*&R
z+*O0G^;bxX<cPl2G9#RL<{AtR1<HZ7CSjXVOFyb*ULAj2RXl0EZ7T&?->>3Jc>e2D
z%(_fYLs<JfL&2!I+>^xDz3HlMrgrsLdgdD<c3%6}r6F}TbmNq<9utMh6N%|HGI6w&
zpI?RC(HH)U%J5Oiy(6ux4LL7wXGEVxc4ol1Ju}&T9$&|}nVRQc$@#9e4-Xv7<`-G6
zl(jFzgGb^a4B=8qXk}!@*-jUno7;NMMp54NVacIy7Bc}+l5}{mGeCw3IUp$m;q^1*
zW3PdjPtMH-*@MrE+>O62ZxXrIpoji?pe2vto0fP(J(>Mfqd2@1>C;hR>BjzFj|2PX
zw<mgJi2i0bsMTl*OEUzdFtbP$D9!j!{5-c_yg@And~vQCd@|5KHRBQ1Vc^6e`BIw$
zsc50@pRtn_nYQ=(%?aD>diL_1(S=z-PNL<8hYQUlo6`;{*GMJpMlxi4aN_7}AQ1d5
ztg@Fs&c-Rx7MoOW`OqYVd6^&N(3$&b)p*n^L}L9G8!!+sn=PF)%lxqhUj=%v_u8O~
zQjFjBXAdBd>Y|^VCPja|%1VTm=L!dwob<hyNu}iq;&OvQb*mM^jpDf}`H?;^w<T~9
zV_|ezIzgMXUpucPf!mFfdyQlWDdmdD-s|7jbKiSdUioc-?MXDdFP@?e;;UHzlzfN$
znmq*IVU!$+Y(6yH<WI|49<{1ti&!=gRL3-yq$N0SD6k8wyKskfopF)c3D?k1q%$VX
ztf8!+>_h!x7ux<LJb#IX*TyTC-!(#)-y1C!nd&K^yVG4UWY`(IxF3mk#uI>EfHlbv
z5$O%oksgeZ7cH2P*{1jQ)2s$J9AfS9W|_fqU~jQxMgFYVXlENLqBBOC*n%nW#>!R)
z=*xa4qj_xi8k67EhtOW=;Hv<2ZN0hm4*g`;P5p&TPJ#;-NF16ObrjE+($O`;6o~fL
z7LKxPAPB<$O{q<<F(o5*A6`z=q7>LP#x5>AbY!_gu>^9=G-W$rmXdP4*$<&Nr<|3<
zeeg=Q*$wO)Lua|Sk8{O&u?7i`^t&7Evc9=&ydhiuyH+z^ta(HS!X&16q=CSC*fawQ
zP@GQ?kcFxnXEI*ai<(CM+RY*o0IX24yJaSEz}M^7kbP6Kv-xs+M0Qq3T$uEBDaE0w
zN}N)7`psTr5beJ8Lfs@@(Ze%Olg$*~a&EF;tH5MFo8_3AQ|;QEA4I=^dfpQ)qg}5<
z*U&#d<azQ@h~S+e=z-CMaJoHF7tO&)siqh{ymTh|yWV7a*=?LL?5Q<8JdazaDiz6N
z=V-a}5*nH6<L@F<H4p^lxl<D`MjK5V{fhDtL!4F~Ufn#d7ib&&x8eQ&Jj*2f>wFUc
z?d*T6yZ?^F(p!pT`u_&ZQUq>kLof1~={KU8%6tSqNW{0+K4rXi-!E)KIx_0xvSJLj
zDC9zV5t`d{r!&XYEGmb_zX=XQP<|89O;W!0EnEUK^Vyg$wz;~%nE@2zolo&D1~>+F
zGs5wa_VfttrLI3ttFaV;YGcC25aYsc4bfpQgSyDw;ME0Mn(MLXYo{yM2Rz;mryKtv
zaE_2oa{3+Uc4g#?H*0caJ|Ov7EqmBG_Ret7aQavTZxg>Op5Ey<cs@?JS`{_-uC^`2
ztN2QhAuNhE_+X>1xr19>r_Qh3%cMu$OF*mZ80(%K%w<enqc-sLY`_PI*A+Nwf5^-1
zG_Kf*sn+=3dNmfHJTolkEwLfImZD`d-M438VH8L8tAefOolM#Ux5!Ugx~1{#ASajz
z?7e3x=Ep@~%*Q3Y7>E7<pG(-_IgW!jUTm2pBPiUitN5i^uc1$)TBQQ(!&Xs$LvEJO
zGp6+T??-zEO~afrq6a5PJwJCI1&{)a4tbBtWhBnwqxy!A`!;q@M>9yMIWSZ^PCF&K
zaBI3)#KyJoqr<bGy%eGf$!7(U&VPb7!zVl6XleV$uEmAkq27kx6{`ic=z=Dl&ALMM
zBJN3#1e<N<_qSSB*>DRH8sl9Q&Ox?)&Zotv5zA}B&x`3f;V)BaL^ye(6FRh6rFz4^
zcVmQEqN0mtycgpdi#Z7DwPvi4zrwenb4$8C>6k2{&8`~qZN?`{GzCc|T8p2>Z!#f=
zY^}%T+UIQk^V5P;@8XwyE&SKYgO#n3Phnmj#ZCLYeY87|>1Fny-myYRyVqs8Gxz0j
z>%_VH#-S`HgIJ?88g`TDB>?NUqVQ}fR~lnEmg7DF;X6v4Sk=PGy>mi}@BFyac|w0=
z=FvJ-0W#~G^eK_GyprziXBO8#3qs~jr@7}0+<rtJbL7ZY;EVjfn0o7=sNeUE7X(E?
zM3hbiq#G8bL8LpDSW2ZASh_<I0bLrTmhM;<B$f^pX^`3lVF~G6y2H<Te}8Ar%$Z>Z
z{$ZJ6cb@liU-xyruEv8)(T0BMM{?d4rVt+Zwh2#ocKD(j<;+bDy`_X0g!eY~Hbv@I
zPcM0sk3eG|(n(S4<W6bkVM;<BBqOUDxCLQgddIjqpd0~I0ZED3>%87+XC>|SSu>^E
z)Ct;%^tYBG)G)DRIL;&As?*}{ge={Yx81j1`zyz~dk`4M%7Xq{>6P=i2!p=Ob!Y{A
zU>z<66Rw%+i5r|bZf>pGA0D8PGEZAP0;qrRw8-REQK!bq(7pFJnUeVQz_i9w!y(I*
zQO8-$t=Vp>Va=N~(O+<@>CO^mYxq=6;j|Q0LdiS{{gF-Om8p?Sl>WTZ9-nIL3g2cw
zLJs}#+ud{IP+SXCC;yb1sie%u@lv(WCA0&nQ$#kKaY!`gP@Z;8W!rS=4G-$2LEWFB
zC85GvVP8|KUQ1&?mp=5=KGH*RrAPJWmm?9CKPI1rQJoGp@)irAJq^Ez&ftx{E!*_t
z;IJ{A-y!_71|nIK>95_R(xgIucA*k8F!^ky+A$dQdwj`*q1<-nRkaf-!zb6i5w=E!
zPuK@fD<dmbVlnK9I#gpO$6<Ld^u3U~l79EG+LlRyZIb}RtBC7)Rf#l4+YghXjRJF|
ztc@5*r&bHzIRy>x*e$HR+9RoU@p2l@QGmpgG2Q&~G}AnieuDwYCj81J)-dR|q}i1=
zy;~6=Vxqp1`)sVNyryKY!KXRl#)ZWAAcS@({;vJ*IDYAvgpr381YLY|G84il?(q0U
z(mByLNBR-R`4GNZw3XB|s}`MKr01a)-;Km&z+Q5|m_Yig{b2ZDLA94l<{k;P{NRr5
zLQ%ZoOs+*mh@?gNa{P>k=XGzN6tL`m|I^HEZ^`1jKW9i2yPQjT{p3wW@gs?}>OK`C
zf=|8oag(v*y6Jp|_TZk8A3sE3?PR4P3J3U+=|9Wu%|Fsq?e1o<EJ9B8BEpy_d~Too
zTzm7$>QZZ+>CNgiWA!UvB)!2T?<VqNsPweOQnSjZ@y}+v9-=aKre!oOHMVPqg1+1@
zHDS9aZlXfFi<CB%d2`xd*eYa+etTMY*#x@BdRmS|js@rc_J8ZQ-k91Bh_TQ%TfRI-
zNQ&M9XM1)xnJ|}UwX%^dJ8Trx66#ae*dtYCvH5#$+6@SLoifj40=pc9Ifv7aH$kUm
zfx%av`qXaC&_DgrBvV;pZkCgojlShgQ3;5AKG{@=Y6Wwoak92P8@+{_VNRTmOfZ%Q
zw~*04>kwCuNHev-c|Cf7$N#U9u*S`U-8L342Q?7*1FeAnEqX|4?k)1Jd!m)6$f969
z|8lyfYoev)5sRiRW5e;45s!Zpn*3-?l({e|%Z=wU*vQ=9I1xPuEGBg3wNecB&ctch
z)ghTm%U`Q!@4ApXkY)uNTj-(WI-xpq4&ia-t0#C}N8F!Dj<<>r?a4J-<1lk$M0!A<
zGC+6d46_^~1~UQn=3t3gxL#YVKC#LNZDWDw4!yi;<ywzBPk_}$1w8uJMiZeKKKtPr
zpt8J2Y0hB;0r-XDFeE=IQlsGt^CXm46@DbSw_T)edIKHq;$21##M+{XwoOy^gTdxz
z`2J;-@I2UNKZqje<6vDa1gS9CLJV&C+7R$h2vYTye_m9Mkc!ak=Y)EBo|vQ}-?Tj4
zjU=aWd#g#HMLyq8EJc(xk#yv+e>!DI*H4Pmnqr`jZ<6>8@ponsbICa>%jk>XZlpUy
zVv8KIwZ4H;kg(@PTm~HU*VFW+b?T>LpuvNnpVAXNli4=83a$QydNX9e+O<$PP{f(p
z_Gf}4Oh=`8vMHD$K_2q~qJwdq&1}v2elm|~n6<)gGkqNYf4d3EEdXVN>P{1|VBmfX
z{6z47D+}Nb@jn)OXh8m*gtZ-Q=VZ_eoD-4kbkdF5ZD|Pg!9S^$Jj0LMr-V3n%7p-Y
zk9Zd4&hQVa+Ghl!E`$Gg!^T-?E-YXw^K<@^#Y=9>G6|b5OEvJkCdn^1YM>Ex+~8KD
zVMzINg?6$%?6E`3Aw)-2bkvi7nz7C?X}}g->`&Vv@TB&Pq?Z+^_X}}iV>d>$MRW9U
z-%@OP7;+wi76{gN8PQ$VD21UU%)6wIcM?9+di56nX3Qmx$EvjZ3KMGxp2%qtBGN%)
zA!(9AIHSyBGexI!N9t%vT(M&PzJ5^`$}pu*$(Wc^W&bxXUAt3y##}*1;Eq8-(3jA%
zK}!Gd>5mU&jcFyk)JFwHY0K|b&$%Sk@%DXZRW`4;splEmV@-2b_k6O+9(E*dZ?kt^
z*^}WhNFTbb#D<?UL)4t*=lRe8b-%@R?Q!u)xMP}Z2ma{#QNwPVEvc>1jarGQbGrX<
z;8`(SSK9Uk5j%awT=E|dxYzva#$(}^i7aAU#)5(|5^GzE4Dw#eG|GEdM!MfBs~-EF
zR$<d!g9AoX!kb<)dupx-RigywORMxeXy1~yb?`m=c6G}WHr6Xzd^q#|>V(i~zA72!
z3znbfZ9gZuJN98dyfo5OMo-gL*#5f6_c?W0*6&r1P+o<lk(a%d-U|h;mCGMk207LO
zSWbzagV-G%#wRk{)_WBgYUe(xk7i;FJ)5Y+Xj~^G2V07X=~=#x0OL=ec+7FI@n;IE
zQ=9aMV%jjMPT#<J>0Jjfthco$D(EZ1Kb9XN7Uf8mvTs>c8SDuOqs!Vw%*XW|@>lBe
zN$t?UBN|fVCqtii4zd#0(pLrJxk3V^OitunIwqEBA#%l;9O7@1Lwj9Prcc#<+D@Wu
zmN#A2n6)(fW_$BQwb$0_kL(8Y^|p*T>m7b{!$7kmWijQ%Nd(_d<k5)PFnqft33V})
zFH<uob8x{dD$}Nd?;F6es58G)_wnA}=J|y-J`{)^Oz-&&Z!Sp_X7$5Rl)}<fXSjuO
zK-rO9ETq|J$BH)cMvPrmNHAgkx?CE2%cyo-I{SCGdQ@wo=SQ5Pt5E(9n8!;k<F|PB
zntg5klAJ(E^4)mZ=w8z;>kc<qW1}-H8YVq0wZzQqa}sZa@u{BpIc6zwzGEp^jmZ7<
z*3rK9#aDUAfjo3`cL{8{T#x^Ie=Z^G1QlcH*2^21SqMiDHIC!)5RacM@*>8&if@zo
zRA=Ro@u!otcJ^!yQwq-WN#|YBXz-+aNqzS0P+~`%OO<xVYypOUYL7*g^?H8CY^X@8
zjI6E55cE_~Iu9D+*AsRIu&>`6I{o?K;Sm~)*rt7063u=&Rn6(&7-{lZ66*Msb`L&b
zrfeLhqw}F~Y!44!;P*ip(Lufx%+Ho*opa+$8-i7#LxKyKI<=$aPtjg!M>$ZRM8k%+
zZN>pO%tWomrlqp^p~XT^WqtHLY-=o{A=YobEWF9rHbFC;YGj4a)grRiWaZ{xs-|=;
zBlUF`N<T&>4fKO}GF?DN)nvzz4e@|lR|fp~pH0VzT4PD&ouA)u;p#`JmX8`RWy0jM
z4gQ;&>O+Vlv-MS`U~t3TqV<wwYJubyDxlS(htfpQP2BWI4{d$~ufE5t@~Npr)5uLF
zGj6cUQN~mInmeV86*9ImHYwY&=nv0{_OE@VB^(nGVZ5jjzml$$R{~A+e}ftC4dt0H
zI0j+X><)6!KL#ur(u5v{=BnKP9&V-W=Lqw|X5c44TPdcr1>;P{(?*ZTpVi?7dfc7U
z{v)uDz4M4fQvmOfb5<O3J*@<wPAsBtcN+Peto4*hGui}HCJwpgr9eo}@<5SGZm&M1
z@%dEatWIS&>9X=-$D2W1-~eX_R2<cE98hJ`Fix|4N|F}@zj9A!-V(t1ss!MW)l;$^
zK!`xvP(0Ti;v%^HL8A#J%+}mru4^KXc39F9RMAh?7UT(ssDiAmO+oyzyj5{W%b6m4
zdU^lX7%Vnm%;DBB$OqJ=6pu8mvHF#yfOIu@@46r$IxBplk~btEWg}zrxf(K+;>;{A
zc+?I#`y`oJRK_qQ&j9?OjYu6*w4Q!)H32JUJ7a7u5~b8#nJR=if$ps%hRqz(eT5ro
zsqQ;2^SqCya-+&BryHP~WJH*#!j`iXfnL2L0AyLWSN?B*KL0*$eb?FZIM~qLT~kk3
zpV&}n2oaZ!mZ}5135=%Vgk%`q(qcbg{jW$p69EQ3NA`RMV!YL!ii4=5YADwgrslk(
zlAg+K<n@Q@kw#>HZ)wVozT|LAF*cmq{<5PJ4KX{zW=cvPq@0v9&OKGPV+^yz1`5vy
z0{Ox!^&c%ZsiE6#f{OrGB!0~zlHa4<WIEzTjI*usx&1HpiU(IFN=6(JL?1m7XF$%Y
zlac1^o5({g=0ANwzm<>Vp8g19v<P5RVxbBmJ((BMJ6fq^VV->LSq8!?v-EtYj%|{J
z*RdUaw|oSwVxZ%|HC=3g`u8|$Ja?cCtvJ4ihve#T&GeP%WU-1Q4mH)5?w+Ak8|Ehp
zUA#KKiW}DEH3S)jbd^{-Nlb=RCa$UIG%WI>0)4Wao|?#Pa~F0PefFiws&M-M84do=
zbin*GPrD43|7yY!C!W<uH}Lx<(bTtg+Z7$I@3Z8;QAitQdly(qt_B@pUun=7`VUK1
zR!?I@qN?-;Ixw$?Vpi-2>w_hC{0MY`G2ipdS7vpzblm1{*m&N7A-T6cw)&)opuZ8}
zHy=|@nP$5!>aslgIH1k)&@X>BswfnMU;A;)63Fo)>rjqoi@B$&QU3eI^mR}`OwVk`
zZBw-S>O@0b4;y88m(!_KO&Un(^W^rl%KB5O0S7-XvlL5oX<nd5rfc0_zaOJPm}?G$
z*^1}>=gixdGY0Uv3eQm41=sekyE?Cj!+zI@?J|Gk-BCVL|4d-Bhqj+1M2q^fgjvs-
zIOJ$f&WBU#waYldM-YdLYwU;?PmbW)BSiimXCA_*>VCU6>9`JlZx}(ZMC2CSVKJko
z?9f;i|3@DD@1Vp(e5LaU$M~h<ZEK(@!Cs$cf4pcl**Aid5uSE?SG6OApjarZC}Guc
zDA?FN#QTUh!P{)e|LAZ0`;LXkW`1EtK9!&ApQaNy$40;J^48>}4a~zoYS6+0%V2VP
zL<@=0fw(&L)zjj*n9_-WQpb;>JpOkr;lw$8tf{q%?@XBwuO=i%jvWv74OPe_<;v-)
zWKz9feS@lfjr>kmx;G!!?toz6FqbF|o@~Sz+*A^IpH#3V-s`@@bN{1yj|7N(3Z-7;
zu!pt6)F>%JerZPrV{{_j&!FaF1{%66>^zNy_5f=mlfDi9xoJf^nEIDekdeh~DNg%n
zw7i1w>+NFOrgIxq-D3R*J=GJ%ZKw;qY-W+|kB9efc<OkH^)fL{N?RJ+(U{wN2h0Z;
z+2<vL;b!gk)$#p3-mFh#Sp}<py4Z6nO*G1r44mafOKUX~ekAP6uin!#;Lnk`RS9vk
zTQSHBjD{Ld$}TER+9zm7@pRK9Ov76wvhPF@yh)g9#mx%@-Oi38rq_6yau-q~PX)D<
zW7=yk2i*QJN9+Brcc>@79wyzcLG{hx(l(B(Og~+|z5fHWsB5Ywo(jJTEqX}((ss6k
zqFdT}{4p=#o{m?mRde!RYNi^<3I5*VxklswFhtcN<!zw`o~rn`a<eg(Tr76W?=Lq=
zhyB)>?Yoa<Y>fiCQ~D&D8!ZjVs-H}j(&e_XRd+3@6z|QMe3kf*Kq8>}bB4G1Q|0^M
z@5r&zg~H5-!N7SF^s>j7{oWqCur+4JmW)+1r$At~=#Ae}GwK~BHZ!KQFXX;8bC6_E
zed(eGV7PN=UP>>~DgHU+Gu*MGr9WYKM7}~MyooK5!5K-Pbw+-EAJCSbq=&<sGgz(6
z4y+FKO*K(f#iRmV%xbh^c7Tg6!SR`GifMaSI!H5^D-+v<XP-#1itb$1d9;U#d9V+W
z3G5LYs6<B&11qLyV5w)+RYQ#hPLd{m;w^b}FL4XGmGjL=QDt<Uw6OqQ-`Kta2C6V!
zs~jKDJLVQQ6jGssrP}kxdJ}I$husu&X5PFMKYfF}^NpG<2d(_YIiUEM0;P&->)@Q+
z=55I|sdckszAThAeoJIrSppqMVoT(uhn-=()xqC*zs8SD6b$|<r;Qwd)-uvms+TsP
z#z0CYuM=^5uG#NU-L9<pCE;>N)micmqzcQrK&nYks%IN8^=t9Bu+{XIcLkV~e1#5$
z!^f4xcyxL!b%ZpGFC%?CdmFa~t`P94$F&E%r5ynRcuk7-70~8Hne8tWxPJpg*4T=9
z7#pz5ThgQDQKGtcYU&9?ZwXL{Bd`@h@&<@Vi=m3r%c2d$yCczAoigf&LmghgEy1UX
zIC{3=Sl-g1;$bQWv74NW#Kx*CL;A2tqegM-uP;ecpYB=IR@Q;cQ`UCOk0fRZkTMMa
z^NyFgVKsovw&@F-3gTq2$z`Q81zK@YfRReW{a^k1EpPkfF-uuw)Gp?!Fn{0FC!+ra
zdWtq+?{!Zszudzw2A`Df70F{}U&qaME@$e_F`^_Q48HcR=Pn9}eq?R?mQ^9Luj7E5
z&1zl|r+L{-O|$Lot?i!n;VHsRq0#v(!N>^S0Eh$|06sq0Z4ic^Opw>8=d~p%@Bo8G
zd!B#<x8V+Q<%D#5m38lteL-2A1@2jS%lU;Q<#yN26}iAH%m*~|a+fPaP`^`=@z$@)
zl;3BC<d{*dasW$X8&Cbg9#tK)B)$26tPqqK%+fgAVjqx@0s;T}a7?SV&>f^kLQtA0
z5Z&efhb#ZyygBpiq^3lc#CElZm4v7yz_fcv<i$!0QGDu2*{8Q;!~xVXwvhTUpXQ&v
zo!F`_0Xy8cc<K6$;3<I|&b$m>fuHqZ*nK^VxLY2l?~?{yCEB%7c(5DJU$cdobe<q)
zi2A8=E9y`OXE^;vY&8m*p*i&x=s84Zsb=@wOK?+ZtQnwI98azIX1(ziFzl6kMoMRC
zwSE9VU;jrvx$}SDc_axj%O7Z>E~#pl)Hqg3Ogk|GCpdFK!53zQvG$pt3Br62_VHrZ
z!Gee;7Ul!C%$n>#z4DZ!zCAg;B^}ORYl8^^w}lGOMY5Z_6MC~7w5{!M6J~ju%lbdL
z0o0dPhizvmhogO{H~$eh&Golh^qVkC(6E6>wx()*QECQ<swmEqWyZ_uy;aw?k>?J5
zjX|Wp+^mPR;3{7x=Bd`~{??X-eb$hhF%cpe@)R8i3PF?g{uQQ}K$YgaExjwaBaHhX
zV=7r0aI~O77C7RSzc<@Qq>y848S={Ivo$DE;!)f}ZHo4!QHL~!viHgTnJ-(BRkx~A
z&+iTg*kWBG5_+eL9iF3sw`^_Yu)_2g1l#SP(H-jS?a!ohJu1Cqxi1pzQx)?$(+}_W
zvhP0vYs3{AlN5#U*bO^ir-o=<vi-ZWsHqYmv7bpmw2+AnSiJ@Z>`e~;kTG3Ig^d^g
zrl%=>u<$tPW6*hd6X)E21hnVS_Vi(VpJw#%ioaC;gVMHZMMYz7>s0eB7e`qmY{<fL
zQL+MxK%@Qr>83X8Sy<|06Wb=VoCVdxoUlC#$k)crgYQ5?fgq&p2~B*26dBu|B5lN?
z_++ERii~sM&H|!nKWa@+SxK^%?=BcP>{HEl|1d3AM!{}wdF@7ZPZ%id$zN5^|B9Dk
z0L#?(Y2G`gyGxp|G(sThJV$cgib<+w|1D`-AdVkGdk(OCg|EHA^(sIJj14W`vPMcM
zAFbyx^$n4fzK18P=m`N5yTEa<kzHSjo&;@vsT<jOOTN9ojP&tIxh}gUJkOcyUXdG$
znLtS&22v;hSVV?7fmixOk_-sYWs?u}qsiFG@ffv++eHJ#XpLh>rsX0Nvc3asa-pp4
zVfoaY=ql4B5k7Aw%_U|u>TF+VF0N7N8oi5rs+sw6B*1+)#eY43)WYE%PEZ8hI^EgL
z@~HdAE#GQw(pu$c;}6Y~cgD@=#q;T-D!yyOm>)98Dum>G?7R9J0gL^|V%D~>GdTe%
zUEY6>8`o-&DVC~z!JT7Ssy{BLZK_nM%?WlEJM2Oee7Emf2AkTw7<vjfSWhY{aF%f}
zq-EXruLi8>k#9*mdfv{=)}+N93ofLNoBtd-_N$Wpov~FUEui52wuuQb(-iG>J>`(a
zVrv7PM_bx1AhZo#yyk=>El*Jv!ogS>@X&`KhH3OCzl|-cZxAN0P_oCRdxcfKQF#y9
zP%8HwdVG30nYzJRe&u_pPdS9t1ynL^#z@xlC>IO*<39x@xHm9%#5fLz)?sc2|2%-|
zR|cEZ#>LLH<P3)@YnewQgjwpBTGO!c6huuyTM&ix+|-B)i#+t5n9~!~1B1@qk}9P5
z^Y9&9hi+|DRU6m|Y#wY<^^hZ0IvXV>U*vZ)(<78>k_Lk6@n0Fzs{q4r=IT^*%5UzY
zzLGnb-j+{**ja%;L-jVDu;v)2O^+EG@M!kBzS-;10P^(*O94!n*UJS(#(TU=9nfL-
zed9D$2uGGvKZwR*{5+1ix7fL{de77Rn@yA}4Yi+0v)gt<6W=?F;9psLOxVa2Up049
z<#S`9i8A#aT@yXg7qN(MjJ|3N@LUj5{r5Sqc4QWNiMVpxg^Q+4=)CjcABuejDXEwZ
zgRv4Ou7sdTbVK1wwV^)EiI2jIet9pN#8P#{@3XO_qV9GLA)*-HmsmJb_CdI_f1C1+
zoPS{6vC&a#l3lKQV<h$*zi;&@iw+kH?iE}I+kGYfoDMai{UTu1s4c%YXR<AfgS#G}
zehDzfM%BC$96sczj%V&&y==b1QZLm)&Q0+49&eNUAGMsFR*G4a&#vs{p3b+n46GU?
z!7*Qi`63MTCJM-#P)3y2$G5h&9Y%A|o+*tv3TlH40g>1&ez9Dh3U2!!rTsu6C_9;Y
zONsYZz05smG^g^{5%|IENBD$&FqYozY`IDY09l);Qgf>WazMY-Va}lr+sYgXYXGCh
zG$|bDPmcuH`hMq?bs_UkkLNKzajNfQP5|~v#xO<D1Fhz`=LQJJ3~nGw8}FQbrT-%A
zBcPuC-ZgFk^sX+Uj(ZX2O-4Wnz=^t=BcmKRC#lXZ+Zf`*zqBTGLV4Pvr#{eLcxun9
ziO&+iGZqDa4@~~OCJ>Rg(Ox+>VY~6ewk&GN5aTBseo9n;=Z+$SCk_~2O$bdvWFbSb
zMtAP5BQ;F)23snTdk^vprRV}*Ul~6UwoVm~Ighplujb#kJNd_e;^f7|7%x_fm=hAZ
zA@}t~qmZJM;ML2{eX5M#*Gs=-8neBErpM)}^b_+z$BW%jJZp{}!RJxMI0jSxIG&=W
zxGw*x2odX3Ybkk4BMqU6+9aTVO#B)<6oNK^D?sX5nR!WUHL;DS$l9(Bdo{t|Xjy-j
zQ`2v8pf{M7O3yjcRC{vcJP*z{3<Y-cy5y}X00>O!aU&yIg$;|a<%TM^i~-=yO{O1!
zlzazh37y!>zYA#0z<m00TlX#jlmE+V$G*kK;_subh+7+*ywcLf4~AD+LP<JQ4#}2P
zV-8l-()Fi=*c)Rm$b8-NYX)m(k!SVIhrMV|=AMbpbfR^f&gUi1?S)f_fVpTW&4%sa
z@mckB7_T+dB#j}q)im1Q70&(7aA!{?q_#RlY4Rk%&!Q_&Oh_nK39yYV?RB!bFqB-4
zuJco}+v^LFFpKhKG<$MPXo-l5G0(povFADC6>ePbMXn6tbvz$!HshL4d%Q<qV3!j_
zEw4W6@_IJP?&7r=R`w|d>m)VU5L+&5-;$bY&wH$+brOb1^WFBKc1=%w`mZK$R^v!3
z{_1N`8l!RH*~W%e&#v<+dJB@@%|kg~w2_i>#s4iMPIsU9yjN^prDHlrwa3gWBpRM>
z*K9t0*Hb%QXOWLdz{C`%!A>E>w0D+IhD#ahI}GKqAZ^|O<M+(_gf1vesNefHAIZs#
z5q#Udu5Os}!b+Gb_#c+jxw-1A^k_U<G}0Me(uCV?X~fMt5f6Z|?>`$ACAzVbz-yXr
zXVYQ3ZYxtC?aUL_diT7yG%pg?WMduSaH)^@xd%N$8T9qjc@u#CyNC0*?#$UX!>una
zAcpB%nT+5xdWo<reJ8n+t#~=8oqGfCy?3N1l?(d6WhR&3iN%ZTC=M`C?f`$oCdFB<
z2b9soUWU@*aQ;x6opN4I$<W`X{wDOpmY(sY<<p#wP+qRsK1{?(;XS|a&vxCHOFvFO
zE|3~bfk)N_g9pB<HNCan(dUEV6HqdQ`Zp*<gwxUXZqmPf)*kRfd;MTctX<4ku)+7r
zi&041y4xmi?xf=9+N0$(kc{71>2Vv?#UFj5%ossuwv<5_VGq+~D9@Y`(e~nu#0mM6
zYnQd6Y#Hto&?yHgSNA<)#HHHV3#aPp(;-uOKNVNM<<6=J{7p=Hc>U4s?ByJ^Vdk_Y
z_;!#V)M7VO%QaiENVN7eKDB)eW@$^Ej^wQ(k3U^%w#ps0jwTEk@eI*_B&Vn8J`>V*
zpzfM5y#+aI{OFPoa5YV88V>4IR}=)?UJS;vBU?&}a9;$jfN3zO$;~Z(?p^s7_niY_
zIlj`mXs^|GMFk<h5SFgAbf`by-xa87mN`FQ76H=SQLf94A6si<l3?R%gQAk8Q%#18
zySIEIZ)azyE1=kE=10!{;wy$RDI@05Yatk5Ga>yK+kAL(HEk&6K87X>Hm+<-e_8$(
znN&Ul>7gGec#@rj@Zzqp<%1-7CIi}T{xx?P-9vCAV?_WAZNlJ2<xo%A<QfdttM7<e
z)73wP%-AnZA&2OaBv#<Wy}e$dk5n{^o1i6+t{rC=y&8<DE>*k~mPcgVG#ZS~Cb-}I
zWI#++&R5!BLLnBUJ@;x&@aEIfZ*i&A9nC@dRHA>E6XP`sc>k~=bzl&A!s=cfL8<gI
zkm37>BW>rlY<WT=&@#cQRoOyRBD%!!xlKDXIEG%pk^@<rV5@%#&3_DVIZ@1lP%UL8
zu^l7Dp0IO}>cZc9-3|?GKCsh@`)FFZ?OlVtE#v3*kT%%D$ym%cHKV}b&V6d?_J}cT
z|KL=Th->6qoT)$&-xP9>8p?I2a{77t+c&)Q9D;K`rQ+SS<q@z==S<`&h+q9O6j|8#
z%cU*9+G~`u+WZ5GEcl1#BD}%&WlIp_x5j^|@5&({>p8xF!Oi2<7W1Ky*Yzza-TsoT
zLG_E~s*`GTfLAwz9%t=#u(%u-{Y!vF2I-t;dB#k5B7eHGz=kOpKM6)=eF3w_i}rc;
zs`4%A&FK9i<7`^ydF~;=iE#+<5Uy+k7Ia%yJ|!jb3|}g^=B^g(=i08P>{r~jOy0B~
ze|_(R)tZ{OlFr;5ysm2oX{v(We)(T=xFAU*+YzMdAYfALo=&X^R5U_DuCNcXe9dz;
zN0aq}6kqT17AI2#rD<PbO`+=1VMR=nrsNAr;ZU=pho0)lvlq>mZwY)s;FgMNui+wD
z(g@!>LA#(%sN-KAT=R&SM}BQpPvE{m8e0xKsf)KF>a8eG_5_~FAw0HpG6P>Kf>V6Z
zJ}EICby~V7{$zhlZ@EX(IiW<V6F@jR%q#Mp(hbvgNF=LJ!SV|Qu#7xV?~sP9jyMA0
z>e=utH|-;epa|H%>`*7dsvmYe%4~4pFz8TW2gcjFkv>gZb(7ph-t~gy55RxepW6If
zZJK1usNQb5&^?HhR|65HnXB7eZg+)gdc!w>s2Vn}H?e9NJp>!|C{pK8l$STXzuNBR
zhN5`Sm;^XUibCXA`PJlPOxrI!qtkMb8~f3&t3v0<O+8XvLn1&TSW807#g{&w5L9D@
zG}huv;)VZL!hqP2z<?Qp$GaRMUrNZp2sccO=GtOI@MKYoJUVMbzaGD@{-mXr`Nt{J
z({ithd*d9bKho4eOWDTMfd6-z!}Rae9(dHU{nims<d`=^-pfxZ1jr&|Qh>+n|AtnQ
z|4}%(3=+K+$=ZlGTiieAQ8u_oEzq}5gR+|ORrqP(x%cf96nU7n;vd~fb^&v<fMn8=
zQ2u)9DF4B9l>)DDZMKugB&+1#x@RWN;XY^1ms6mo9gdmAnFv^aUc9a7av3pNKK}x{
z8`CcB`kw#(up;;>XnH8<Q`EMR_(W!@oOy+yP_Blk`AjI?I`R@~$_0uoU4<46iWI-_
zdq}9$yVuaEKc9^`oW;<*Zd_}$?Sj-Yw^n}a`4)`%whl0L!RyVJ>Z~8|ifkIyulgdF
zkaBxr=R;Fj;xk+N_4kTt7$#u2ecgfbuE>Nny*Y{Xj9>rj_)3cF3Pz~7%~uSzrHkLc
zvsVn3`mvr}MH1wmA?bK8v9~VLYC8})HgvQnd&`@`Z9fWi^`^7ygT8^*&fZyB6OV06
zzpae3dtsc-WLRysAv<QzA-EKKQd%*uIx!TrjKIf#C06+(N6y<h3k$^^6%Dzy^yg3f
zY7d%BeeL<z3(El9-px8T#gf?HTF@0r65fg8C1*dHVS?Lo&lY>!z3ZygwlTj@n(zcR
zN?&;xe5{zV{&P}MyQ($`T_;MS91r#Je)ysEXCG!#wttFl#$fmt^GI9l*>WCZq{&=<
zE`}h6F0kdLO;9T8C;w~p`wBmKpU3_BCL3gg5J?;i@)@~L#4JbQJpm{-7wX8wn#>0R
zn9%Xr!sbY7B@-@oS)(WZyxgbj`Dv177$!GriG5owF|pU`%)F{bR#h$Kj{3+XLzQEV
zc+8rOs7Noc1mXN3X69-gGzNz{7f|1+7b+nzCr|BB)6;}a#$tu#qi$n-DP0bq%^<z~
z$wE4tYZ}oO&To<&U(2L*Nh>%4a**;;7rc|XC3TkL*pJ5g@?}q0;fNzMwwdSpyCOTw
z>C(#E5x%0$y4kL_HK{Ddh`{@!g9}l0z9sh2B=E6hdhx9F(6*v|gB?oLs=49)8upov
zQBTdm-WPGrvA<SBbznMWRWWTa6^Wn2I!{#p^ol{I+Es0R|C;#95WCqvy1#}_dHC{_
zdUC7oPCje8d;=Mw_<LdMcfH>$D~KiJoK?J^@-RxkB{rN<i~-t?#Ex?&p;fN5BpxUi
zf4**twAPhJ<?COEc-zuJVZdrN7puyeb0BSkZjG`uX%zF~hk!2*xeW9n-Hp}CEVEm&
zu4t+3)IA+i{Ei9jnxz3%aHV~Me?DqLgyG8`uS%Rrk}cUZ+yQ_Zi0U-#i6D6Va+*tp
zXPw(rG@2(4P0~H7iet6gV;KZ8Hb_u<Y*E+kMx^-AZj^rOT%yEwQW(@V0GDlq)tp#k
z?)yjqbsy299WipF7VslW<fTE1;9zBabSFsQ(T~^s*}P++eM;Ts=)zEKQ7YD7dFILI
zL!GmdK%=r`^+-cYD65IIW4mZ?dpKtuW3a(W*Z!6|zKFS_bUGO{H*zbF?%V1F#bE)^
zv!Gn0x8!u)cDmubHd#NudA;&BTj=2FX^pgR$%s$r3;q`%9aW3mJVzvCq0~g5qKmqM
zEc{0CmOPL(lSAZmbOk<9I|b|7K!%B@@yLy75lpPdHpm$`@J*q9h@93)ME`2?I9~)#
zx_~KPdbXysgKh6nUF4sWepur<GR6ma16~ukd7V!GN5sS}B+ygH?E4iq^c`76_BZM_
zmi#8dR-=D^k{;HFUreWJt&p`S#JEqC&s}_6_wCkP3@_TD<dq6pJ(V*#{*~s(E;`38
zAAu<Cl=zb8${u(|bmP*moxq<JllN;KeG+)qM0Bq%;ERupiIpxdemlswvjsHy!Gd7=
zZILFk634N%P3hopzqi%DFP@F!>(x))vIF1-@CM!4@lH!@g`kPB7SkJtdj1U)y>`Cn
z3-qDr=y!RBj;_|6cSkD&bE+;AF=LmRM(qiYZ<{;iDIQ$BvZ2_jkEnp^veB{gb{zh8
z$-RyKM?1H}!rF#~X(_ufvZ)$-{Pr3S%U;>Bd2SPsXLxaYm<hWhH#_n&9xL+@aSOc=
z`y0^793nU2<CK2ye)@3A^{yJnKTwAZN>}=eris-fuMLNsijlJPJ5DWRbFq#Jns9ui
z{zYB|4tJwDDL=|~d1OJ|tnc+|BcZp<IizPE8_WthS>8)pM<bsnRW@XnwJ}6#V`+={
z`f<>&agAj;aRxbYgiqQpPb&@YB55Cnb!irHq;gxhzS@PhNkC2C>NgFk!{%R?G`VNV
zbvU>6&W%q@6a7a}qa?y826M&8G>@>0*uqsV)0yQP{w#Cur&G=O?`!EC-(`!E<O7aw
z4D8ZVCrnyjMAA4=ysd)!g-v<y+<dtMxvq&rc{#UEZ%U(6%DRHkhE~&B4*>|UFhv8$
z6r#^*gaDKQQcLSQDYIQjD&`vvdeHSY?$}5Y@koVTa1p?M!=REYMWo4@&sgWF2pJ`y
zv=vaU>z4fctiawsxfrGj<)PCjRse|IUz<FQB!~1R*76G<Fs_f7CRWca)f%-Ts(zb$
z0_p5KTxT0w;MJ8yAbmIUAA#c*c*Z<z>t@UHo-j_IntzGlGC?y<XCW1P!S!>$!b(lN
zB9)of{u{GnHwAffykxd$_(aG4>n2Loe2=?c3v}Z&vml~GzLwf(*H975WT&l9fd-+v
z_dNA=xS4rgv|M`dewuV>*@LBfX>gX`%ton}BXrO&Aiwb3keFe4G`Lv6Ahc*91JYaU
z4}$TW`&UtCxd-JkCU<N;GAfE+1zEsD=XdUq6`1DdMu<VBpib*7r|C@5Q&qkHPG9Gg
zW$<9hb*#S(|9}2*W>uvT4_5jxP=Np6nkbH`5aobpuW-*3SjFQuNl?xqIu}VLH)03z
z1vT6C#2+Wb1Q!V3$UF)JU5s2q!Z596eWUa;m&q1RjWsbAp)*9cMWc{l=+BW&_#&DM
z4?^Aq9fq9O(=<%Jy`4uFc7jPNSkQm%wR}T@-=*O1Vpb9@xO~MtdtGvxxcp$Lso~!+
zm2EBm7<9_$w}3L6p)G`B{~6#awr3^FhDWKIT}2pgwD8Wca8q(L>HGl*IG(4aVriw{
z=GC24QuZCuoQzXBeL(Z2A0}2iBqN`PFaBix6%z3no6J^jeX@3GO2vzOPb1&Ze4cOj
zPT1M})%9dL9H@*9OC=Bqv+DtUGc6N8Wps*W?J@o!P%7h>i}qAqTsATX5{KtRBbH6g
z6j&8uyJclA2)(iqdxb~9%fDkZ?0|hF@Mz1wsf6vDU7xbL?TS$oi1()57ZPbl+We#9
zojashgWqK>+xHx=(c-s^psTwKym30}L$Ps4vn~MrCx&WtW=F*i8PtW_#YR&!;3sbK
z=KIE9I-<W{@bxO<4J*_1L!Au2_0xxw>T(Ox<#dRJD;0>U?f~}G-#`}xJ59nG$qnzH
z%zwkSvF^ls{L&k<Su8{!=I-s_wRQf~Jgg=uFSAVbKz{rr$kTX;&mvWo+jfSlw<v@q
z(ehA}=_Yk~ESw~V?&@bB1nJYZq%>D%`%=q3bu}bmjCvHxD<v$sYDA-Sqxg?`ivhAE
zo!C<8TTUfzE4Td>&c1V{&#}kh%Ii^3R5^Bog(_rvDh>rPpR`(6*m^&8OQwAyL{Cow
zy22g&JK4!48LzRLGC|DHrOKl(smG&UvX_LVhV}gu`;Xuga*4V4iDqfA&>R3pENs8A
zw1^wud}FsFk-071nOo1tGYaY>P9Cr~CwzbdZPCID1(i5kuJO!*ZZjHO^5_+_tCvl-
z?25O$>_jj06f`&SyFEW^Ud``(ZPE~zd)?cUMza}gYEw3+#!$ouBOil*Q4!O1VH-B-
zQJwsOr198IA@(%Nf7+KQw#OR)1|I`r20PVfFG1W-j_tcTt9#c)R6~9<Qw@cet)$!G
zqkN*p9u01()Mame>b2=oMxB&Wv1ykLqOrtPo`Z%lW7x=#E1l4}_mY}L?|v$lmCCT7
zD_~uMo;}WkOam!DanQbnYcxdaapU*^a<BmU{ma*Zsgwa~_K*gBk*`-NgC*d3MN2q5
zZ|KU%+%FbeHm58Bw-wyOT=_5Q$i%v8(tGfR7A7<Cy}?FW%~#H3QlzPAW}5m#Q-EKX
zl0!?>nlfaZptQ_-4C0@fn%Fgb8sNKYFqN5={?1LyBtp4X@=3EI-^M6YK%E{H^vT<V
znaNu-CLq8{CC|qE?!x)(C@P}GsK@i{_(>&L@lNwL=9>FVIKz}0|Dwg$mW(E#tO{^=
ziKK85m35f2azQrg<VZMkqIB+y9PqYto0dR(D|aHmH;Fm#HCMA;f0r9i{7Ni`p{<=7
ztMBSPCEhW_n^@*TTVNogbfdfG5pKK^Q(wvVUw;$yCk!`KW{5}Uv9};=XsWaUv^A)1
zo?MH$a}O)-R}i*|`c&LX%?_$9`8haKsRuL7N18~!Wh9<_Jy=*fvlYN%?NV0TdX%af
zWJ=~(&YDiY+~6>^f{rW>B(03ij<E9F48Wpgbc}i%TOVZ~kMj0@Zi{)scU*K<aj_-T
z<4*PdjdQxL&qb&CL$&=dbnsczZAnR!=~zSCw~3eCfsQ31`v?Br(xY**N_3HSEY9=d
zf37=cYIL)G+T??&b}wyP&ff2`9=w|%2j3`rKmKZg*9x8gCVxKF*?)mN%zAxcv(!?(
zFIX6t>Q`CsrLlQ8WWe^2@AYl&&HWgGpP?nDxnZuid>O!l{yk+FHdn6c&=GE;9>QqP
z6KuiNPs=a;L?kv0)G-m#mt&wQvIaj-D17gz@L-!oGIxJj^2-w2im}oeqv>z9PbVBK
zr(3zA2T9?Oy<XW~`5P_VmQ`-L?V~J5>IUxGo+rdAN*@UD37Rbp(tnQ>a-8YK_q?wa
z{Q5qrE2Ojh!$QpIL;30sr=xmPC1?NIoH#QkDuj3aY&I{R|1SHN^><s5puKmRW{)#&
zp+a}zRD2iubr(AHEVJ=;ajGIPJnT%zEK6XaVnBVwqT*q!GaXH#Zf5cEUG=H9xPL-W
zhn49L9ycTaQX|HG!W}z`Pq{5%>?UP@C{{}sD)6r7t&_%Oc`bh|lt{|@U&4t>B1J)h
zhb&w*EM!bgj55P-xyS*hrQ?o>P{cP~IJb3d&i6W(yc#3+ax6QT%z#<ycl!bz!~`RS
z8H*sPEH};tF-6oniM3jF)Q%>QZb=xXdiWN2jd?@4*WGRes(o&%i=xZf!wd!M9~f_5
zR_+^#1MpM88&1zofdVSbemmp;jqZR4@f#OVi*LQ^-F{*trW}v3{BX&q`dte>iv1Sk
zvjskvANtoz(o+YIc#y1!o)7E_J`9RF#hXXa^h4DayM62Hzl-E*_U;Vfmj2xeq!|So
z#mdd6KsG+6l;5*vXrexA(lEazM%wK&T>fK?23J}$wPf1!7*LiBVku=>xsgwaBk>9&
zkXGy<pgrfpZh+$4{_+Ky;#`6=tih=0^cLbyQ50OW8DscLBDrFmt~uaLd0aLCGpTF{
zmY$>!qKBWOxP?eMyvIlOjg)<xs-2Tu<Q~;>o+ByeLS0Cwj{S`wl!J5*A1SmeWPccp
zTW9v`0Ms5~eWJ=-6cXYDaXmkuEI}8OT5H%ne^tKcZ8YM`+%CXh^#o)!JRC%N({$VM
z-i`AxwPzz-mdmGGE$*ZBg6jK~6|B;z-e7nTOX#)xzv(6gx*G%e$ik7PZVAz^aZ8EI
zf8(n|RYE>hwAILZCO}d)5O3(-*fM_DLb!x<c_Ri5*B!<aco~jtNvXY@vmVVECote`
z4Ad8SH$ZpEf3MAUub&jV^t`^*TZYnCKUz%xceQVfcr^sv81Ha2HwU&YBkdE6v{DqU
zFo_TPs+}c?Ci6;odljOAZfug<yb?}vP$7695z59}@5Y(hY4g71dE7g5PaS=eY+7Hs
z^HN1*-ZI&Rz84Dao*_rhl}%qqSjm&@H$rkAO8)FJoeIf-9{>@@7O0(fX>uF2R**k*
z>J|((oOOIzhL;fmr|AjQYZs}4q<U2}>%>&^eW<NCx*gG4WOps~iT=4mU6WGhajYn<
z`1dGL4dUVNALV$GpV`0sd#tBv7jPL``PxW_-~xmHYquhPsyOCYa9-N|Jl2u!(sM*`
zDk+pE{Jt6MUy|Du!!Oyrm5ghw-TL_X^NN9UPwf$0fvbqwU>9HfgVX#^R&8Sa`#U;u
zAK8#0-zUM=svhktC3m^zTHK!lmGOQO_$`Cso}BmAJZ`HZt@r4NP-b!S$IrtQ6}q<)
z4)fCn6U&+Q=!;YQt%{pUbO(#*bQ{hi!CFPDaAN(Gy@Y)R_E9Dl=ignvb&`)T7ZyHm
zY}c(D2s6Xl=s)Zm`;&hU@<D~_-l=<(9XlndvicWO|9Z%vcD6_sE?6|G8O1qo${^Mx
zMe@gpxyrHo$hV?k<Y~1?yqE?V(JOk6Oxu<K>eH^>r_23SAy)4TMg&LEJ)JMJE<W$T
zFBR1?Zg<B15d@T9C@lL$hppV_GJf91^RjAN4Ix1LDx_Zf)*La`G}H@)&)%$3en;)F
zGoMgesw%5IyQ?>qAMD8}#jc=vx%K+0d{m2$p*Mo2%&Fk}LmBwGtt5hYfT|6_(#I^1
z-aI>ST=PuQ--a+)zwTf0fSpXdTK5<Bnb%?V7<au1K=pIRgMUgaZ7sA&Rl$7d>%Wb-
z?_j2;GRCY_UwqVlIvf8wA~T1|43<5W+U`)h+rDYR=pV|bJLNW?5$epwxa-tM*9(19
z8EkCu4fg@?@D%~=*tXuen@P69gLJltKWO`8*>H1zo8~CFo2L+nbXwB@^>}9#kNKuq
z6qY@U9ppVE-hbHkvSN!CV4hSm)Hh}}hTpHdx9dE9>s(u2rbbsPvVyP!n93(*ya~(2
zq7>XKAP$!X`<9BvxaP7z8$mt`vt;fS8%b(sV*xc5*}tAsWjO8M?EXeg28~Hx-?KfW
zC&=nNh0;p0KHPKZ?Gz3Hi@drr$$I8Y=e2W#R9I=klNE=^M~XLBzmxvd4I5b$K9|S-
zrIBZ2!2alP{uK9YF;$Ab0vS+L_TWAyXTNr0M*aMWJJ<Rt2s=C6Y30Qb@*$i1#wBRV
zRq#PyhX?zn#YF#>>seC>xb%6!c@@&)_lgshPQX7KYBm1qnUaUN;K>QFoXbNi(Z&Zz
zgSiAquWFOf9<{^bzBiyjta$E(@64hs=4Je=#v8KMK%8+V@|jm)len%=;5N^7wPPa=
z_ZPKOjp*Oj!J;w&cK;4pA<c;zxn(=rAKxqn81Dp8;ongOPptCcB={`$_ArTNRoY}s
z4n?ilf=I8C`W^V+cALX2&4Z3MbkE$|Xr-jQ4KRaix{+4><O|H3Nj7H=kIzLMwbaL~
zc)n|c*^98=73~R_b~@ecg2ls);~E-Zh_bMD$@Ps&;T#+KsMNO>VIw<(+h*a#vbbbg
zTjA4^cE-vyxGI*HMNH4%xH$_$!{iSs+nR;MlRM2j?uPOd2kfX`NG@@}pT(8e`Ca$i
z?)0oo^UIQyF$Sxc*H{VDBA?>n=Xx_WXEs=QF6j70O0R)-<J(WQ=KazUx7)K`jnT4C
z;4tZ-BT4x?A>;sOl6CoWyVml!EDKA#&LO1n{rl6inn?xX3%{YCttyRq_M}S0WVKI3
zJ|OY?mKzZva<)r35{z2vCNGP0vP{FRLfgOljJyhsWs=F#k^n9R-c!(l*6ey}RN$<L
z&z{qwSx=?8mb++Fyi&ryj*#2xSK@B4eXZ1wqYnn-*VG4-c3n34G$I7aLq@NipvvF<
z!srfN9S7|@EXdNW4`0e?Tmh>8)6G%ocB^V>S3tIU<sdLou_@0M@oQFF(qw<VVtXhS
zTtO!U!HWR#HLGHBLpjs@7=N}D8BdKQfCVc7PqMT*%|5XSvX}Y?{$755&2Z<OGJYJ9
z;EY2A1DSHzehQs9!Q|0dsS)B8W|myw?N@_5sLsxgk|?_!XXbA)!@!T3H!t%w=T-Ms
z#J|3O#3QNiBzfVfJOL|c0UUN`AZCqrj5f}Rklx>j>~V7ieIo|ekaI;aLuU-)0KzhS
zr-W5A<$=~{;HKYu8RZ;K$QX=q#X&WwjA27tisMunmH1<TA?wUg01G?cQjE&w?OL>a
zB#)Ny2G7?{6e_>w_YVU$^X%>g=M`spnz_>|t=RiVZp9(zWQJlQU;F1mka#QFkL^O7
zwV{mtRd1Pj&r%LXO);f$_Ki-Defd4Daex0KxC0~;e$`ffAVJyl_>p<LS{>wm6j+J_
zd!JVC{T}XPRpg#HDTy;CiZ$s=F7UYfmTTCj$DA(Z^~iT7laMo%m5fE&N#(J5)aG4p
zacGv5Ez3d!76@0sES-05WWjHtMktL!f5=zaJSgxresouIh~;Yx&*hy=4tcD<n+Qf<
zAa)>s-q<m!aYos%a?)Zi-%;vC=R)d}f%wxJxvq)2z2{`+$HM%x!)if)nwsZwTr%Kp
z9Xd|4b9hS!v|dJ9irgTayGH+bK=z0JfT%K=7W@4b@Z9hm0nb85EzmX6^*8^0;}a*7
zCIyNRu229Iaf^tY(^mRSu|)iy8dRQDn{*uzhPyYhZF4W+sj&@6xv$*%yLb2u;V9&0
zuC*n3bAThRV{Ub$;=DqzhO;#(bFN;DMsO}!&wCqM0h6;hsdD6Ecj}#uvvXVId(xnC
z-Rien9kxGFsilw)y8Bdzw+vD>NfY8%ZkIT4UJ;jSaJ!<;&k=XD0wPU?ot)xgvsb2)
zC0OV4m3fAKuIKcCyoO)?op{Xm#*xZ#wm4aV-wj3%C?ihX?-G&Y^}6n7?;#HBEAb7t
zm=9)Adl%*378+vOYI|f4T0pPknXU{&DlmpU$3}KG9r#ytn|~w^33Xige3vtuYgEcq
z3$B1yhT&iPv~yWQ^-??6^vC8G1)*e-%*{vxwWGhs+Rc_~O~_y)y0j_=S(ZGMnKdma
zkBqK(HjoAT-hbfL9kAP?U_ePCAE$9FMq~ppTaBwr{H+e%{_Oump`}gMxbhW7zn2aO
zng3L+lenig^`Vs_Gle`BldRG3;r<`#M<>|JaM*HWY+(D5Tsf!GGXUyT5c-=}K{T&H
zgGF!gmYXAJ?Q3>?GZpO-#?67GllR5XYBAht$vj%=O<+FRxW+MTDf$}h_1lbw=hvZl
z-C)sBN|<cp+=G?LJN3)6wca_MXjy*x44-+G?>;FZJM$zP>eJ7^Nw_f|>R*O?Hb2!i
zZ{jPz?kka*BNH_jZ`}9z<Yv{OE0(8O#mR)<P$<zm8hK!1wcXX7{fM{S@RG&Bn0X+f
zH}MJdVVA9%Y)*5v?^ku49%QjORs&jgp<=6<ANT#ugy4$+vv_pa1L6FHb{DCR%3lw|
zY@q6ezlNH5`ZOb)nMfOOnI6K>J0;fTc4w=X!3hgU2}ue!Z>DYaWXXrc*Gw3bZAl7c
z6MAq|kRh93-;T}h#`5!J_fJ80cLvr6$3EcmmYI>`D%JRzKG5Of&C|YK3eUnFh``9g
zX*@XKx%aseQ@D+@YM!8{HOL8~HvV>FL`O`5?qe4_wNtezheHX1c_jN?Nz|S3W6NzW
zJyEeee#14tRvwnrF90*7HpzM%zH;H&<7W+-Y>#9Xg^tJlww_Y#5*QYEaEgaC9$Ad^
zr^;vcIBVE4{|c!l*#Ijnv7$5F16qJ)wItJ_&|g`~si?NWiW3l3P)EC{YupVGjd3Vz
zI>5tkR~@bp9nQtG7XJ}cMk@U~+_C84h+S>TG?Xz%^Zc3=A?@-7($$wn&_iuBnz4n5
z+E2!pncHuZIH|8SuQ7+oUK$fy5W4jQXxEHhLo%wU{(%z8B=s5crj~%ov~<_VSa_;%
z&ZcW+yzwW|9;0X-otYDWvTgB%-^_vQs6x`%qr$o{Z#y_>Y$ow0q=i-Ztmg4;se+Rm
z-`}oJdydXYr#wcFf8?57lkRqkhCn9epIpcvgE25e(WD#ruR9i#HIN&hi<v`j<BK0F
zWGHX_Kw_PvZrJEiZ<iiqyGRqSu^V_1t0vU;RH$c>>a^=K3xCHnh%r6*OO^*oVKC|T
z4f$b~!NK02;`9-M6E1e@)y8WRrG(y+%ugXJc~E>d`wNG72460Xg@@N9KmQ|WXiSps
zp`PS-TP(YnF8Xuhd2ot({%;mZzo!!B)m|1{y>7xxuVQ9pa>nQfq>P2^7hEE?zja2V
z9{z5yb*#8DazBh#O!Dh1S5<+U-Kb8~BuUt)r0wd2*==2|7u-nJ3*9_wOzD-LPkHdJ
zGm&uYskzp@R<T0*qOOjoCllmQ#u8%9x~KDxoI({ppWl0VW%@mg{^CD^YNx}>0kKrp
zd7!7ZE<oWdyADfjq%RMolOv8VO_>hR(kzkuq4d!Cgwowp5%Fm673MZm4M$H#L4VT^
zN{9X`$-;*J4@u`4&*uBTeXW*OjSke-)+(tj)K+S5k|1Wq45Id?s;Jo1-dhl=#2#(c
z7K+$nkD{m%o9{or`+j^q%$e8qxz5jVyx%84+0#enjS)9oj{YB%FP+~u>Si{}epzji
zeLYns`$X1z_$WJ-#GkF+Tr;(Qm2d;zJY|1)j0aLiY>UQ2=r%3Hu1m#&=xh(|O>BBp
zh5^QJ+9~;fPXltd<eI5+1N*_k^O?)Ucx<xTzj>~;wm0!^H^76uczL*suCcwgoD$Rj
z45l?~lQQO&XCLj3_HCU1BO%#ng{f?2$iTx+z2q{_5--hapS&r_h7}PPnKhQL&$4<i
zMq&gqs;fC|Wb2Hh%lcnrm@Tn27dw?ZpT+Duhuj-kHh=WVI*xY=WvVW69hb(ea;W<}
zPI%PqTH9%>rR4fJs??!j#O?iV{k=3GQ401cnDZ?Bg|#Hwg}Ie7^D=CS{qJ0{FP(LB
zz&ozrk|OH;K&-*Hh#q1Z2vs@Vf1TV3lb3S{)Lc2sTscDgSxO6!Kdx!7AEBB3)39~7
z{>AH25MK>x+@ePRKfkyIJS^7x8N)LpTSlsV6BV=BI#qVN>oHB5*0_`sL{ItLl4s>_
z6kU~;Nx>6(lms&3@zZu?H_yb6J}|m(Lak!3HZ>zRXNq(78w&o_2h#gt68?&PXlYHB
z?>-Qfd3kgIPrOWECV%#v4M|+e`Rb+K*-FS0&J`1%8rf?AUrP4l7+qDWkKldYzj2o*
z+Gn4<bSZjW-G|ntA5CppD)tK{D)BB&^@h~`9qqS+Nm52KE_MfnFjahS)lt}=ZQn-x
zi<Yw}O?GgqEf4&m>wN+w+lSkU&}Zlp5x<TNCt?o);@j~a0fvpBV5|JQ{PlK4^Nx&-
zs*@SmyT2@EiusWq0u$XT^S_PNx{3M6`2ga+sHJe;Et9gwtv;)Xh;Bb~Qy6O+n=_@2
zL^5wUkBQoay}_`p8mul9$mrcl)=J8oDwEf+`9kvfd$l%Esv~RU>^VT<oWi!cqhyPU
zx&51DS8r*zTQ*B<?1#f`E=0z@zR>)1*@@wn{;o68R(G2_G3ed}jwrMnOA+;AC?>{v
z{POS3d>n&LYnSUh|KJ*;p>m!@w|GXbsr-)P<pSOqq#jzZEFY0zqwPu6#l9>H7(dRV
zoSRduq_}tRr?Bj|^AsFL{0o}gEjPZP*i6xs=5Xk0jwee<tj3pzECFlm?PR-k9z8*o
zp?}sW88<T(r*=;lJ&HLg8e+rbPOTd;mfuY5n>5qIE6Mh?IpsA*kZX<pwJbj>r3<Un
zIiFvE9iLljv{W$_+2A$0+G85DbKuofPTCB>JP}xl-5xx{q=Sfh(sAc)F8p(6mNzD=
zLTg;YkPMW`I+5(-Vv3EZ@X^~SOQBrbdp@U5StW&m&(B8enuwaFZ{G=eqyx6wWDULA
zBEC+IZXL9)oFx9U{|=4ltQcXjW9a`VVa25Sw?eeP#^ETQ>AF$giR<jva+bM-(jzn9
zO1*gw6|2C{-`7WgXGbf#k1fygjN5Qt(*20NeJ;n%RHY;7JpZj={wuwS8$RD1&;eWH
z`>CfsIvO!+ip}TxK_99gmM4blW^L*7TE2B;Ai$L9w{(-~uYJqcu4l<CA}addUw#et
z@B3|npq{7ZE7H={aWs$n>r+0px30TMjL>X!6R${`cUAupw?iBGqz*?CAvC9O;Lq`f
zZ2K4KlEn%k)No~&D%Ev;VspWQR}!C#d&Vl@W2>_`nbG9Z3ssxWt<T%OV6qi<Cr$$l
zyJn;6>_bgrQ8ewup(n(5Q9?t_)Qe(H#qmR{iK(lS`oOA>TzxEuw!qPkvafiBZU}xh
znPE$<^X0H3u?uw9vONqdQ|`PwusT)*p*-K!h0Nxk>?=fZU|+`;y<SPL`d`U6vFPXZ
zY(a*}eeUJ)g=)^|7lCC!i|vQ)FFz_6F@P^XpBrgI)93`&mDg;8Dr!`|qsBL)xy$cQ
z6_9vLpHxzmxv~HuaZ2N7@RjP(R+-;Q<`E*S#E5FfkV_&29Of}wDQeqrdYqr2d?e*Y
zDEn2CMq-za-s*nNWqJ#F{gUMH#%%m~)?ki~1gy>zWpiG1?t4<vxxKo1oL^jR33=S0
zOL~cMLM*0xR-I+J23{gZ$%T*elj>u9iTYZ!D*FOforgl5;kaVc&#WR_@9tTrCo#o?
z;s~7XR!j2B;r(VMu1iZoND#Ywk4qVb5B#>vf}F;5@sgjlK)i6$lB^DV+?uv<W19yP
z;mRB?yP6Cvf$i`IV)@S33vw91OFOJ#Z0%SXrHKdV&cD8=t`>ZkziO!0w@maKm!aP1
z!(Ewc-h_#G&+8b>A5Y8JSu>xSbts-!>Bx$Hn51Un^)jUdwPE~ZCtVwBm3+jNNeNtS
ze#Y$8e<WrsOGbYiKLl#KtP0-WYQcPodL{9&h_!?8<8vG;H<vniC#sZKn}4g0X=VZM
z7L8?zDKXw|avsrG1Kfvuy3t)uSaabGDk64`23|0%GP@z!gqTCSjlmBzM)xYY*R?a8
z1?k`qpq_5A?UJa1RkRRSqYF9=vRckzNsKP#=*+4y0@REaN_6}uJQJRT83;f!_-XJe
zquQlwzv;hj^MZZ!8`6IjZ%)QeFuK&6gbJ+gsI#oQwPs&{#j~vXfe+jGV>}wsaC<=x
zJL&M7e^~dthc_=B<*AP~qc3+pg~J$c5mmrUt8#*Y@5h1}|0Ip@4>@_nRQyxFb~*}M
z>FvQft&}#tN=xobmAuA(mASQ#_5R(}k~s()?1piu;!UG2^#X%&9iqM;s*=Q@S~Q1>
z?d<bYVHT>G_fDmq1|we&ddvJb1;6YO2@}V*cj{GR86~#|E`pkpbiOp|Urkr~zm+Z~
z4b2QjIdCKd;9Q$WJ#rlytxor~Q^k!>PSG@+pr_<iAB9as2Ivp{JOYd~SpZ%%L`O)P
zWlGd>by(MFUfZGOyO7xe*YX}9D)xk6>0b>oHcmXepktX*^>gk7R0}!Dd7TuGs;7^j
z;13(08d%qR<}q(yXhtHpzDWi4<Tn$8w<Jy!2b_&*JSittX0FZkJTLza51g*pHJA51
zHtl_|xB-kCv)oP+Rp%U5-@{jQq*a*qh>8FLAZ`9@7om?FCu2FwzHxc<ziiD({89nO
z_ssoa)@Sko8RGbrWeJ!{KkuCl^_#x_>baktWHLr8TQ3}ooYdHB2K#D4&U`KbTba6P
zz&$%RT%V|MfbA2n$m~Ydp#}OgpA~slvlp|_!Mm-V?)T`9_kV`Wa8GH)0dh=&N8NI|
z^Kyef8MU0*f4>Nh2zA*7JsOP%E<n5JCsq^p%?Nx0ro`Dz#;SdH*j4eLZG7Z1Qd(^_
z7WS8eZf+~)3BoAx4?KG`w)qaI!QJJ38LFC*Eyi|l`gr8Q_B-0!e!vA|GjfvT$+k%4
zmPg-Waskcb7~Mpb)kJgJMWwiUY1U(g9ydtJ$4RQ(k`$tw7>oiwXXp<ZXyrg{zOIL>
z`_@+bSU2r6_@ek-Y-Ya$m8yMX-mp`5=J=7v$h@7>_-jeK4y1L|iT;9lofnlWn5Aa|
z>bO;tAjgLX{v4)}`FgdpaOGDO&p&l`4cH(aQ**Jgb41sx)!g|hs`Ju-tL@mQunBqF
z>&N?=wGwig+$a5p3mx^p?E<17{}SeCg#Tj?y>Oil6us}w^Ly`P1h{WO(*X|0xdV0z
z(RYHq3@k2f8$CuaibIgHmhXw584ST0jRv9#G2N&blXl*-)6&(2L=K+-CcE)Qv3;@e
zz#nxizwbB~7_n|i81!r9-?if>!i=<F4Poydgqz_4BIdm=3}=W1Ou^RE5a>Opuf)$6
z`A9Jdta|IlNxmCXrv!#5vNvH6)KFnAe`#fHN({|RS~=<tWyr#1Xf#NImqKPPK+bb0
z$9j?l@6~pR_hozK;xTp~D8f^++NCok^zL{)u&sbd!ynHXiIFTse-zdir31620Qzj4
znW7@LNn)qtQt0=yW{VMUwwe3@O3O?Q0y}xf^f=vm#Sf!g%&e~_NE)pEgD#r7tiwUH
zq7d>ct=otc?ROGdu@AbnQ-xG|u~}ZtE2}e&v0k+hrQArg!_QAZC@Ya@&Z>4QYjCxL
zwohXR9N5CY@a^WeDXIYBjcafyDrlx!SRlAR2-`8cBDP2n%!Nccy434x37<;9Gx+9G
z+bR0qUKp6$ZpA@}3T*cYs8&a||K|CH0pFxZHGli-3TH;C$+g7DEs$L!-UKsUJapZY
z(S$=k;DJWJD+^wwk1c39`X2@A7bx=~9>BLT?o&d%XRN4GYDwdJkQ=`e|9JY$P`MdL
z0Hp7}+Fd>3Oj7zlzPiv!48l9H&x%~G=Ka!~dC`&BCgB87Ei-k0c<ah|&uJ#am3@v$
z_p^V=$VVA|zj&J2&I{WCcEX<2%gw#1E>Ux`^t|+w^ElrZ?S69R)c1O`^tpS7R5@{O
zSIxbHt!P<YMxcRXSAdb^*Zxv`y+<f`_D$nDx!mzqhu6RXq1=aw3?Q)9x1-m9KQjx}
zwc~3lik<1X`}{s(bw}H*TK3No`P7c~=K4wg-B*e$iKeuob_p~;Q~Ig;cQl6RGxD++
zXsaZ^lxlxwIk$9dWt)oi8xlT-o40Sd^*E-xLSbJ=ce^Ev0POnOmxHNd6}<?mJ7Ldv
zbw{vBhxd-t=SK>^{a^@DtwXE;^JMImJ=q7C^?B{?Sn=GE*Gj}s&&@3@bF4xX?A+gt
zfSOR5XVb+fRuGF)mHNv>k~xGYIGnk&<PAl|L}f_`yok*Q1|-_9h^l5eaam~)9Q{Ae
zMp=p2-OtnDo%h2{vhWz;v_Hy@v#>ZCS>C8tYN6OrRgwvD^Vp;iHbQ{}{?+Saa<m;b
zbp%6Zg>Ll0P|kH&t`K&$NxkKvS%ZJ%KmWetlA^V_QBMCBKX{w>$oA=-wF&L6%fduA
zK}Dm&O*TSl>&eo?;FPF4DXq?PMzW4km2?Oju0BNASY@wx_jgYFYIs1(tG_`3U~boa
z(lfpLW{vQlq8}U)wBkPA<-0ghdor|{D&|e^<VO|ge<U;*?wb2?&d$$ZVsE>!H-Ao1
zKwhTRP0qbJB*{jZB?1m~X$`8F|K2zf2hUMQNeNd*A6D~yY~cGPV(ac#aLYyQMsN%b
z6^4AjhDT^a>+Btcd*D%twcli$hnXikwf-Yn%Ddrn@mRjo5*R56dk^dJqwqJ{$eo|d
ze#vPxnfFg;k=@LZ{h^?YTz-2;0Yh?bw~B`eDB5j$jz$3b`I_EX)8BAXxbco%%_}Rf
zmH1cwM(F`YFI5WVFd}>xhSlg^INVTydsMq@oHx<?5iT!a??9i}o8@JqhrKi4<zJ`(
z(4mUp);z{P8{l466Xe<TqHMTZ_CUt>M5zg2A(03pB{Mepcrkt&FyZ`r+=WmUa4#pz
z1(I@dTE5mwY%envl*9FBe%WNS)w3EWLig;w<+fd-j#z?$%OT924zWDT!#Og-!%2%z
z!eiJ%a`eyU4h|Q)7MOACV>t-6`Fk_TIaN+aou&ukMJq%luHwydNpoUj!hE>g$95)$
zU?)j76#zrQo|1=sPv=8m&1e7hiIW2Fzj49Fx_pJr-LVC0NBg?b5Bo$%s`;EEAJB^k
zZ!NTcX$B`AxhE$RQNI^TfrH800Q&8|D&o>59zpnR61E;;=CK1tFq(W!%qDh0z9hDZ
z*{S5gIhyq2QpD?AFV|UEz?kdBePM7-R90et72{ZgaK`|-1wiUVxQ>5nwqce&0U^Uw
z;6@>0MX2`nRE04yoBxO+&&6;3E05D!3)_WypB)~=pFQ;-&J{uDeu?mcn&mlT0NIUj
zo0F^6b{k*+e_edlYE<LEX}n2sLUEtmzrIooeJ}4iL;C6MI#={W!5nf+fL{%t7;`bn
zzp`rvhITdQEKk&#YdrU)RbBK7a{MRpGnb|9usB9#=7yxO1BmlrdoTIdX;kk1m0@1j
z1)<Gg0Q=~5HKnqgZNawH&P#O<j^hxW(O2kaU?jtY>1Fxr%4Ln3JJ!uL_s-&2VJZ<0
z+>ay8e<7zYt<PZ_I~E+?^B1`mHjX{r{y@{Zw$tu+{F7snOQ$HPon?!gZqAc|;S>bf
zo(?4I1mCLTM;A}o9ic8ZV>`EF-1mpLHZX!!iZr}=O{92F)*c}wYeX>Zj|jwU3v|)o
zmI9?8DyFKJFL0Kv%znicP@H17-_0{fs1c3?J$F|l28GIuz9%EMVpU;Dphul(->{)%
zHtW?L3$GPft8*Q=wx+?+(`QSUQ+HRit<IgBCB?F_%j%qN3nFcg0ayQ#{1YH72lJ{6
zdJP(nntW;MV^sl+1saBg3T$3aR||e0WmWp6E7SKk^z8Kwc<NKxE_b!xd5WdIdgWfw
z(|^};-6b<9GeFS#2&=F~O$l-H@mBJAgzvtE{PHA~8%3nc&Xwgg{3M#us--c~ZD=6d
z-)Pdk()i4<Oxhl^08i}OHFO;F3qu4$y(Wo}RO?Y?$iU?_Z)~&U-VY*Mg=!weWb$a7
zd7bDh5GBjk_9k=rpAvBP-cv7?D(NF#A~!iAOf~pZYXt?p;(%@m_cjG35tm%_o@iNu
z$6Gi`=nd0(=p$ZZNd*6`WHpWLos=T-?;PFbcGS7696`t@7oO86?xhZe<QAdvUL+F7
z7dYo1a9u)d*ab*L5Q`+^5Q0bOS}vFUrR|s7lm7ie{M-n439|iOR(rF*qt_@#f?p{l
zT9!~I;2<~|GZEJ$hoG^9H<wy62NcvNl$O^al3z<#yciNsdHte5U_~~*%fH+@40(J_
z?J6gnOE}sO_j>kCU_-{}o3A~XGAfd;>oq~@*_b`TYQFh)9b(rfXm&JW(=!lcD7U={
z?**BJ0_&(#z!_<1gjVx$g@|I!Z-~yu=V8-GOK3amV!OoOuCXE~b|w9xLGM2fU2ctM
zm5L9NcZ~VM;W&2@ucC~w+EG=n)^=!|(vBIS{RaWfu;-*m<&^*g7U8!*MR^NVRdcPs
zwW~zbQZ6V)QUx`}Ei(%px(pobq;KaCmCIZCSo^QMID^l<h+=1;gcrYF{Ey1H(qy;A
z=Iph!wQG2Nh*-eqjr?fQhGimHcuf79Y!TW=!XrD*gjzffL&_nvE%$}U#KUnbN{)zq
zV1L;}teeQrhiD}Gr}WE!G>=#pyA|%CcT{%gCGf@X8a@hT#cQbDo8Cpy`S{Q=Q{jYI
zQTOk@bQMKQ_%Qw`hCEQ{@{w*etUKIDi(SB_p9^b0mi7`y@_sg7MZCOWSt(!yK3f?4
z@TOQg_OLfi9T{-5C!cPe)kqwH;pIvOC8d0cQ^*judL%`>hfBA-FQLf&f+;JBqQFR2
znJ<-9^1?uHD7~zb?3ooxE=56&;oC+D<bi%qrvytp9Qu{fdm@cnzYaP^bc9zTQ|~qv
zyENnDPvOe%2o;G9>pgJawFjpjinj^XP6kunzYs$00Fg0_OUT6TK1eIlT<Ay+(J_&`
zD84#Bu@f#L4nkajUwjZ3dHL)Kv&RJ}Nocy*i?;Tq@SY62|3|Isp(BNdX12J*S(M+$
zs(xOumD)ym6@qx|HGu+i!AAw?OFDjbU^t8h8*>BfbS`uJnHUc(v;lXVTfSeL3|@Wy
z9sOV*AA`ikcKRv2{oUYFnZ$=37gbI<??{0xZkz{9?8$`1r)|Bt`x4j7BQG&orPE=(
z1cJw4v%%>nH47UbBrh?uXRLe)dj1Clw7H@zI(H&PZfqL(YE(G-2=rTME(QlU#6<|@
z@KS#LM(+#>M_$eJEnYz24HKu8kpD;ohiK^3GFacfLNoEX>k`4QGd{nb%F&!HFVCe<
zgkc3|Vw6ynB<#a&Wtz9mBcH3Dwi+2N^IPvI91KTQL2EKRT)Y_d=OWmbp_<k(0Tb@v
z8B<i|<kNQ_75etEnhh1ch!s;T*-t+et|U%(`B@HN;b23X2VSeP(MLbCa-D7m`l<jf
z{^=|B`H`s1fqtFKDXKUIZDi@M)gFl`aIeqOmP&-pt)G@vLyBU<yowk1L3b8x)ms<J
zi5gO;oR0cWP)7*^v1NGXJXysd{_eh(U6@&Tse0h;P06YZ*V_aHAo5`I*R}MIi@;>O
z8LyQ{Wnx}$=A`h(^ssK4A^LzPy6Os8g7x#`dqXK7lpY6I(w#LtnGS-kxR7><h9`SZ
z=8FElA!}Pa$-J&0ShARz{mJB?;|l$c#CV(NF5I!8`FHw;sk9?$@aWA0zj6AVcn|aX
zt(j8ai5>My{`hw-2iL>V%gn5@#R1XLKkZu|OLAU7oHI3m6?oIYf6enS)>ya9jS66U
zupbBWvys%q3x038r=8-yQO!{u7kOJ2-Lkr5=v5S{3gZ+hT7R<6m74ZHlJBjPJ^7Cp
z-1`T-t2KWSE!QMh#A{ps7Pej~pxF88&wMXBxo&6DE_P`D@0pl@c<4B#O@e)FyAuzp
zW&Wc}`H%ZM;OAZ96KmrF#jy8FLaSwBgDL>=gQ<n)j49SA`%ibjS49<2F75`WG`gAe
zi0memezjKw1m&%*-Yz#++E5-LE)y{}QEKb~$5n$(jVeM4SV~H1ADef%6xXx+^?=2C
z<0^v?pW98`y_qDdK-`(XA-q4iIY|4_qMq%w0==^p#pb@1owIcd-r!cfTGK31QuDde
z-VF^E)ot#Br~YrB3do9ce6^RVcB2C6s~*=2g^Dl5TuiYCJu;!ec!11x%0+i+W+GP-
zaAIG^_ll`r#CUZFF#+ESqh2!*YK9iWJ6>gVL9Ybjz`AChHH}w*RojXBbS~9f>Gsu5
zKf4Xr(?sOz1kt;^hwY==|B=jUCe6J5i?N>46XBMiyf~?m*M>QV1cY2-D&JL#y2Yoq
z9gVAcQ5`&pInqfzbzd{$r!KTCculDMJx_nGz7ZvU3@l^Sc0de<c-M4!K1HanG<P-#
z5pToewi-5p<Vv;U4dvcT))n(`({T>QYCp{FBFS=g5Z8P0%*u?{C$F|MQ=fcv`Z$TX
z#U1RtQcvJlXi1>aIEtt9&E)5!cVtz{&7MruF<4iXDMob<s4^csF?CqG<fQCy3Ze0z
z{YR-d)_te*RjWJ0#Z)SBw$=1mp=lV5dUEz41bQsXZXCBYCZfR-kDv;I&?8G|1LdS2
z?QVM2?%ptGvZuavYxAEfJcIaEGh;c!-_&ok@%{Yc`1^a{{N7(ti@{~5_-}u<A_Yv`
zH#v@L^x^TG$2sClop@A1+p9Ud9g`N>HT@nEdEqik^#+BrXGASdtxEJ2o=>(@z%kYR
zx{xwXn|MjE)os@skhr(AtKEyhxonM0hIOl#+hwUNk4w_(*`@TIvM&68yRaBp&qgoF
zxbVK_V%e(J#)qfz(AW;+Eo?ZURILl~^0p-N#!2aaB-FenJ_bZMgBZRZeVA>4AW|yC
zX2DinV%V-N1xnWiuhQKzZ>;J_uw4BuMlPeZrOR>JoEcf(VQ&Flxu41I$D6db9@pCS
zYA|YgT!G@dYz*BUZp+}LJGAi?CP5YS=PIJzq)fZ|je|P=;prF>l_7B5Y|*?dT%C_b
za3)sM`;OW89u@k#bcdlu!Lqr>sYCv{)8H{$BuLQ~Y_{47C*<8b^V(eoOTLn=%q;CL
z8ubh^$(7Fn25bJ<H%axr3p-cgTT3l%=JY~Gpnu~(!j%)*3o1lruVz>%)L>pa#&tGW
zq0knC&pzH2zx(=6L*+G^Y3JH2B=v2aTw3x@V1~WOlwbrd^Y2|j4pu&8J7~j02{^o~
zrAmpT=>j+S-l&}e8aMHDW^VF>psE{0zw5XP!v*&>i#s+`K8XPuHZ`)P*qZwI>lIv#
zbMF@<{HEQ}Uj?E#-(7$i(Y519rS_5UpARTax|Qo24fsr+HuzQ?_Fpdm>NgpBT|37L
zY^_@xvi-*zLLv#pM<LPt@q52XO-`!Ckv4fR)<Q1I?eO|Btt`1TPz9h3L9C)XZF5Uy
zqI8dvxJ0lLIB2B}{XzU0Y1&^Ls7G6yFSeGxvTa2MFJ{%yG1Y>Hc|URztFoD96tV7u
zabGsfarcw-df^+1W+I!%rB!(X_r&gT7`|~PoOfz?gw$z+`4y-kubO23BWYKELNy~M
zN+U?RZ}^l0Bg8e=>^^VcOdIKDCPCX8{)~A&79Z?7mSy0=9JDQ!nAB5RVtf82ZsKtt
zDU7VtG80X*{%&#Y!T@nztIgm10de!CvuhvpEEzeNTfgo+=`y=_*#`Q`c>5sp8xnp`
zZjW!^EA5ANyOLyY6@dS+OtmU7+?Cm7wZu~#HlV}TpGegc_-#ND*63gHi@s0aVCbic
z^CowGY{T`E_MQ72cCuw*V6GtaJ_PZZ6(F@@Ub(MUTn)*H+SgNUeSaP&J<9ooOWiK~
zOO_eUGH`rrXT!K*L%I*^J{ckwy=x9eJ%FoS{*XjJZWa;2kiEJLi6~dSsgd1YKcbqc
z(3m6kh2K0;nP{kKcRY7m6SRc)HG0|gIEA=gNaZ#$mZtQH6v>N=OxQjYw9J|YiM+F6
z*c2mJQybN^KjYm7q~K?3ZNp7O7aI3!*Op?g0rcy-cort*U&W@h1lN70IA!Ttq<%f_
ziwPxDSvay0W2hC^22P(Q%2)EG@YKs1=Q_=X<1X+n&lCTYx`^q~rgvv7Q}P8Hl26#t
zo<`DFe@qJTt?iE9*1<D@!fjcge`H}n;u2W<2X1>9oLE3*#C7(sv=Gv(=YU#o7dg_S
zkXDP&zf)?sO>^hjTVl@XV~Y@cxP;8mB7lL3j+)YOl}VAC9LY)x_s^3GQPt)a5B17S
z6=kkL0e52B0zPPZHUFMZ@`YF|#h5TFAGe(afpZ@0$m}A-lk^6h`ECb^;`)J)nea|o
z`>Q*a4*hj#?)(Q{KV@R@1|P5CK3kvTdoigRh576${nVDvCf}atKzfC%@k71(K7t$k
zzV%~vu(|)#Mo6!2TxM@9kT+reTTd@V^PkR0|5M@%hPgZWEMlk6PeZBctAX!I2lK(*
z(UwVD=djCpl8%&d9XvG)1lesovB&z<ZbgUGe+o5k+`ZIE+Kb?#s@Nv=oLMRkj%cpA
zKRdfMkeRb5R3xl|pWE63wJEQt11mQYJM7)!0s@SR$O@}{$4+C}seju1B2ESLI!<33
z?GC%B?^hctvVp&ne{^abbDnkbeBFCFvH11~5GB@QjLe<rKl>cTsp!WRiS~*{VBZf2
zV?8W5>g+PS1&oa}r`O^(A|jstg<bYk*K61I?U-Ch)xUo!?;8v*=Cq5#+Ie#19ZvWC
z%GEY&{8TI{3%_G29K`~g#Bh8{SLlUq*lSsY+s5>>iAaE);7TOw8yNn-N4Zj<N~ad(
zUU;0L1x#d{9sh0bS>~^Fj^`|!F_dhCQ77Z@udl1&>heByG@X8u_(NEX+Zj`2K_TvE
zM=ZMf2{HOS`!)Tu9yK}oO%Qud{fGN>oNBOeqpv7sR<ozKxBF0+_&Q_%Pd3C*+W20>
z5?e%?o@ENwB_Ml)Yqogw5D<?mxL$ReY$aokAU3&y(kh_gTfiK(hS+xo!$WJ-XPlH}
z3|q)knoq=bc=1PuU8vd8!Ep^s?n1S1fL@w<ivhZkZDJD_Ddz0|g|Ee+MQ(KT$~eu9
za8>zr-0Rc1TuA#^hyB0a=v0T57YZFsz(1LmQ8DSWt8v*=(FQpg$GDxN+-0Y7L2{nl
zFpP6=Ii~~l^{)$~#H||+{`C!=96&+XFVaCXq1RQCCi7)>-ixzJ-V(UM@)R?~Rb{xj
zlV_eUXmsap@BuJ&Q&mI?<=Rj!YVs6X<n5B8bZITlBRJMv^TfAhQ3>N{ljuVejM0n!
zQWb+&`zc^xJM!4Kl>Y^GJyow`W~0OZKq8v!K(0XacF&zlOxCI{>-<^fL4x{COu(dE
z=9~&>mo4S^wX;-d^{JdZbFA0(w6n{bz%7x@y|{@n{!4)wtH7_l>>6uP^BJ;EN}7pf
zdKKQ1ge{BAlm;0S0oe2vph6Bs=_1{ss+)x*gwQPX)hhI@e~>9h2wTdXg_ubNRVscb
zJE>9q*ZA2s?!}|VC#R~=zh(TdU|iD|^2*JA@1jz^{jzIi-j&jfWkrbiP#m~Yo*^BA
zhpU!s4)QckT%`6)v0^<nouojUj240XlOs;f!(LLN!9fT{-A?7FQ;XIv-Yy|i0@xT8
z`?{J(gTBV6cvqFM6>9(OZ!SToI5IX6xyao6_(>jxkrXMroR-W!ZPAT<$XB9NT3soy
zo{|w6+u5bv+s=16POVG=jM#=f7rStL|MFJs;W#fdf1Lkl%Lj%TJ1ld!R7U;ruEM<{
z_)5|La<ICKM9-A-Z(?kfnm4azGHr+!kGn-bU&BdX9#2QCeWnxb4@856miJhliy2Nk
z^(6mJk~o3x$;;QLC?@R*`u)~jg*{+%xyCNahgx=<k@f;f5|!lxgS_%Q2LkeVMH^=;
zDpQ?}3eQM&wa8~Bh*N)U4tv-YJ^yn28jrm1RIy_)t1xAFr5xk9_xEfgOS;0ICan8_
zk<_zhFCo}hil*<DXmL#_J`-;`i|R)#qzup0MswvKm-3feOjL3zjTKH~GW2=T>4=}1
z{EEk9^&}7WkIQPXv&Fv$AGUbWaXJ$U2beKDGhLo5p(#ZP7syN@K3;yLXL(x}`$B%w
zpIhaJ;YM-GLGYhLXoY+Hr~V)VrMePR!-jZ4SqJzM7-EYmO7|bXO4O@oiwHHRr;R98
zX9)f>ao2Id>R4=OFGS8=crF|4ZDk}|VV}0%4Ys35-#7pCC>3zpVTl$_e>rjx{H1EH
z<%m#jOrL-M>Bee1hx4v_lgeB?6!F9012o=6-ghoDR%d^ObVn<psKKGcGD9K%-3Als
z<0h7|gWmS%ca2`&yBNR2p%y@4UZwJeEGAw9ZU_9HuT-=(wNbJ<E~d&*eA($8`WWV_
z;p|N3?ptQRjum+;A|PdWtFs#kg7LhGN3;LPJr-Q}UQC)vtl1(0PK!(C{v+AWzfZeY
zqD<Y)669DrCg9^X7FsYE@xK@<?!jTNK~<<J15oX4mdZNfXG#X-)Ad?_DiJ)%X`TK^
z?PuMds8Ml+1eL*e{@F4x)mVk;S9l|}|LThbrUv1tr?)a&UcTq|h;w?X@~G;}%FQ&U
z>#rXeKFRIVoW?WA1@vxoHrxYM;#__)vM?pm3KUD%W)3+%@T&BoY7U}l*S}%G)tMT@
zsBp*SS3e9?j|@n4b<Kxp(4HsTUuQcnHN(tq(hC|cja2!{W=P0MCXv^a`yRO+RyyQB
zefOR3k|=~bGvH#JBb`TIvC*K=6eBN)-)PaKH^N9qDtCK$SNWb+N9$hiwxV{7oF3}=
zDn507NUnu*n6Osm8}9TA+OD`N!d*}dkq>7XZZKXz=z15ih`u4CJILM)RXN<#CsNvT
zb=_~<Kym0Dt@4*??~~ESp~3qYkY#F>LQf`X;i!b_i)GE#$(G4*&IsYS9LZ`4)s&*-
zj}tR|Qa?7>M91o*qU9Yr?V>APl$I>RGN0a6$)jM+EqL!VI^*MA6W_U%mG>mg;@{o<
z@M>FZPZ~GOXlK1qCJ6AiG}6V7@O+S87xJY!ze+zzD!howXhs6W>_l3?_)QUBoESKR
zcBIcw65dj){1RBZr8jsF+GgYeYJ$Ill$!m^gO4Q92d7j@VD-;$hrP8a%{BQ3_Zt7@
z)I^CKfAnPsiT0A8Sz1MwAc{nQ(VyxfYp}Z(605r!>IFW;7u&J91*B<xWYEHNwf~d2
zsF38Wh^l`C)~cm$B`<C&$=qD3=Q86y<*y~mso}4trmCLyIf489%@<Q&ue>MQF1J!F
z2XECKo?c9SW>sjfP<&cso7EcR^bI+-nGxkb4dho|Pw!jwIa56ip(wDC84M3xi-&en
zXP=Hray<9PiHmobb?fT6nd?vFg5u^><A=lV`sfW$o<!Hf==yx&><cQL0bAx|C7$oh
zi{jygKQ-Q)i6;2zEZruN?Q9lg*$USS=cQXWSnXwkzE=~Q<IM~6K3zpwr=q{vgB}87
zTFMD3d@Bvp%nNn9;NTG>aG2?&7M@XBqn0E-tHi~mFL*fPaw5{VM^evfqlHRNRO}^5
z?^L@7^RHGrO1HYNOBaTJLBjL?pd-Fsl4$b88)fN!;n}3NM{kT&<yT(H^vh{%+{rS6
zQEa&iG-`n_kegm=L!xB*Uu3&&IbEK<OH}l+qPc08T9;|+Wp$`)Vor(YXy7CKWiEh_
zKJ)5+W7ebP!JfuqU4!MxV35zTYe^hOVqDWcoOXlMy#LtGi=w|j3oK0{_;hC*?lER6
zpOnsPSg=rxZrkJ55gfB_3zb+#nUA+B8#R3F%GEZ#g`7TW8vc%M0UihG|Ftu`B7!@Q
z^Jsig_S@6M`dV9r<cG_}61by(L48_Ss2PSA#ZNs;mm>jNtlte>rd~Fe?PyQG`By{t
zo3JkovS%Rdn@40-0tA@j(SF}bR|1YJRiJgF%abPK&ph@au!%K3leuU9Oh>Kp^rnsn
zc_LQ#{sN+><ajbt3Nfp)KT23+)_;$cisZ=7Y<F;-<v8Wfi4>82A<C9}LUMd-KDW69
ztZ4Q)=6nkI*<20ZJ`B>(J`(mOz;x%<YfvXO&=iI-h$?o)%}baX`^VzQ$n`<kAAlk8
z>D9~?fS!zUTGk0VpujxaQuv-J7v^Vzbx;*iMZ_}S;%T?pJoOtH+|mx$bh-8IuES?3
ztN@HNw3LTu-@3J(9Q+V_-#}%!^>vMGZ3bjf;C;|tFFDp0H4dJ2PQv0fz{@EqATFx<
zL21k?bMT*ii>)R8_bRO{$?Xd4v4OSu+rdsM?s{>bJZr4~LE2j!Qsz-wUMJPz9J%T!
zj*u0QwBXnhFNJ?2JpOwJr}5rutf_yao(`bm03G9>`jJ(^x`>Ar&@EPY;{SENm$d9P
zn(fC)d?|R+;qAa{ISkwK)L9UK=f_3bNnGA~8r3)XEb;=h)8~O&()?Yscol?TI-k7g
zEK<8ZLc)vkN9&l#Zt=9N!>wj9wV(g%|5p7dC;5Maz*s#|2W4BUKsCE4DEso}m0_cy
zDqvsXLA<{<F8g9k?~ZA#Xyb9&4P&Vf{t85-irJDQwZD6<q%`LjpbsYyaFuk7=$VJY
zBVtcdI{|UoX3+OD-Hd|bIT1%@AWAoaQ5e9^>#5<p)=x^sp<fQv6L&I6;$-|jSRGk(
z{X6+}F0B5HB=bm~#`wBnS2O+#an+$C%3=iz9j6<VaCuY=p4rZ1d_WPGj$VI{-$@-`
z6YgE8hy1DWUvrll12Sx+FG`f?nI_@4y_#soYTE}jo6ywTmHahjbbiJ_j+gLTm#vK2
z%*tblb^shCGC8+x!lYnsEYX^CkVh#Vt;voxsfv(0KC5B$&r|}6@9iKffffBY(APfp
z0u(cOi7e3!YTb_*oc+APGq(f%_A>Zm1tzl*quKc;iguy2`qjWh##()l-vy|)FW^A2
z2}}3995mmI1rawW5SVKkEhA>ydp6#ol0wPQ8(FBknY!{v%olvCB1;&~8q4ie70UP#
z7i?9=cae<)eGd>{&NTm7bi<-wZ=h&R%tV?Y8TM`-@)p65u`f5$!_GXA;B(?S0wLXA
zJ`#LP%hp*UnCcU>igZzL`M|L`7I7Z5<?aKcjow#?tRx!9=o;N-^S<t#6d1bPetKgv
zHts{+*|nwFW>2Tc+A|N*zg|=_uul-ufj_kpMkdfFyX{<o5CxzH**~|9uOl4TTbG-}
z-eRA=>3-3;e@kAPiR8UOe~jTu&rohX!e)s;>^14kgF{7X3c7*Qdd*N9<Xxg(G_bj9
zC*pxo-rh>xUI&6e@+bh3)9RvK83VEH#E#Xw;6$xI0FOIUwf`gO+A)oZcrDx%x^Dt?
zhIxy~<y2a%Zk)5PLnpaaJo7|tCZtPe+dM0i17qIoniwUMh@TcJiI!#!?PS()7Nk8$
z)V;iZeSxD(%%_IWzpjWfa#`h(Ja^ksV4BsSx024B>(HmFg8A^2*l#g|6Hd@Bj$gX&
zQ;-U3;03)mR0Z9ho%<i@Neu2AXtk!t4-_ggqv#URRhD({^-aH47pZqW>y^U{I$kpD
zuF?2!djk2I^qc%AqcUjsLHF5Nz^rpN^Z7SGqN`#-{54h&@``9j`Jf_Q`(vZ7TF#)W
z+DQW<FYvnCzON*UyV$2TdmGM*u~P>0!fS|K`vtg@$1b=R`0CZpmICvIsYE83)RYJO
z({!nE<K@IpNB5LOQx=MSm%|Hpk&k4e&|lrTC6`=R;JoFKp4b;#c5_ZLg$v6bT>zHb
zPABfewrzFlFzK+DRza~Lau?=9sV$Y7>;rFxkLJiZ=q?2>nXU|7qb$kmr<(JyxL2w(
z8~tm$TX|@e7ZS*%Kc0GNubKmLmm9J$L43o+(^?}pB`c<vBvtcb@=M}`x{)iiZ^9I%
z0;XsNX1SeuK+_(CciZ+!G_vxgw4@r@;%6iD+P5u#X8ztY8+#R}JS%D}^`(bM(r>6{
zs92hm8C~{U8)MP2H#SrKUc-2n^9(aCkeyU!?|*V_O}b9ho}3*8vE?k!p%^>E+TV-@
zo12{P9)#&>##(#b@#+C!VdfJ=-?)KP1m^vcP2|4u`)}b`f3b<gC8IpCV<|k_OLU+-
zN2|cemf5_9M4bf0Ioq&!f^n+fviYrsv0X!4%xZQ`p|tJ;orKB()5Ni=Mf`)vc|W;{
zVaHP{s;q>&W}_Pt50e%p*-~~4Q&UwFAyV^O^^fWI?ffb)TB1@FCL)W67U%TJ$;r{K
zaLvn2MuzHeOq@|^nW!uG!u=JrY5zuG(P7nG_^<DRr03vhUhy8IrW?KpC;Dz2ZT(hg
z=5~ELvggx>+sCKX7L1hRHZi+9=EGTJJNITH?lIC0y9Usi45cB3`=mRWFup%v@5yOX
zzlOObPiB-0fn5E)8w%q#=}=#tB4G&tIq8xnG9E^Yi87SHnl)bABQfn#l?9@53Vb)t
zXb6Xqm%@#0Z4F|CbDL58g2-<XtbCdi=Z(}UGUWoW`Q9TdH$rh8B-%eNsG@t4b16P+
zvZ71$6*Azokb}?CM);X$g@LGF5S(x%f!#OxpRx~$6y2khT5}^*@DpRcSd*MaiNWuu
z<*5b}{T$z<=^#M(`$jyE83@IyQ4FyA_Nvj=^W^Nj#k1R?e?>F(5{>+207=V~)H^S3
zNiD;1wWx^HS_f^nY3cG!%WDv|#(b!`#NNsBw>!^w4LP1n=EwSYqLMPQ%%dUaJ#I&n
zDcNBN{~&l|I=Lb*hdsWZp(2VqZ8lzanXn6S;e1K_jB(<o=kmMk#G_xPYf*&tp&xN$
z+rN7*s9!)%k^`;sl32;#n8j9Awj!A6nw_49C+yyl&=`|W@)yVnJkW}NbRN>;G+s}o
zp6|;I;bPynYqM8uV{>=yqx+Moy>}7|{ztCh4;fuFl;qgf=H9t9>Y|KHGQ{|Lw~TFF
zwEfh%_vfz9e<YM%P8kKJcKr}D%1NGX7cvLL+85REcL`l+PfI01k*O~Br|PaDEGlLw
znXvhQsrIBa?=2aaB%@?aLZCcKJo!E4w{)DRwPa0E2kRMH@R$(0HvhxTWB}k^1aR<-
z5htt^_E#EM+ftILmRFR#Z?2+J@+P7>9W%s4sc~N@g%2-rVD9a>4-AVy;1`U*o~UMC
zni$C}ynz=tKmxN37;iMBV{WFM&m=ML_BUi+FLERlCzjtzQZe0sq}Am=UJTQoC%52*
zE-7gzHVV+P`H!Qr;>V(m?@i4=(Q+Ng+h9ivclzO&tW*|-Nsmg$QfIHJB)muzuG_b=
z6!X&Lp`I~vmT<ex9%(&k)~69R6nG)0sWu7jZjw06Y6tB*{;83lQ?+LSl(efQ?jX}l
zLus{rl6YkDF3)5g*t=KD5lN_~_x?3Yl=f|Rx5e%l3*lGMW)bH`llgT8NCW)#zyR}P
zZGJNw#zla<x#0dBsyH90O56<?!m@UVYIs9NMW>*NEH|=QycH7)uax=5gAtL--V!3!
zPF>K7s+iJ)<yUA?2JJ}K78RPLwCFcMSR}&zla1cBP)%iy-Wz>8VrFRDs4iP^Z%=mH
zE7^-g*kxdRRrMG((FGb^DN(KEKY11d)LhTZ6oZDO8Ckn}oDbNs-dLhV$#Djo?7KmZ
z;x+AFj8yZEttiLnBSVIznLkwgVyTwR7vn9}s`46&Fvr1e<_J;bOHOhzB6XGTK!>Uo
zBRuay${mWOa8u}KOfh`DQB-0{2+Y*D^_ZxXB;_D>Up!l>|KhS;y1DK|wN*Z6ZNcf3
zA%gsIt!a5DEjX7WftR<%iR>5mzNfN5sH)a9#^U@3RL?<kI79Tmm@-ybIK!L|^lJ?B
z{}OtnF~}>D;P$Q2F_lgsHVoe~{Cmn5!OpP?tu6%I*1CBgStF0{CNcB|YOK1pn<x5W
z^|CgcLAPX!?do?yekqD08)lQSnW@g>3NDV@bgHMZaNR+ht3_y$b=z|yxRJ{miyY%t
z2^uyr#hq4ZPR`Ga(A9)IH~x2T&R9g;f0Wxyjp@xCkbc|nu-Y*t@ogdsky7AtxOemF
zV&Y{m6!-OW-!GVS5O2POXO{V#WTn)BBn8cIdl0Xu@u2$SwcGIOwr&J-E|tqzkE%R?
zE-6X;ytdvh>V}}o=560@jn(FtH3}J$=i52<BTsm;m%L>b9yv5q-KrY4=K*~?RDx2)
z?C^h{3pHgbOm<5U>6l0WzxCbpddx-T&2kBrsMH=5A`mI|7%tkfiwXX+$!eV<1iW-R
z)!_vSn#h#Vk}9R(ymgpjb)RLb<WJGa*!*p>r=-A*7A+^Ss8-~0I~nk1B*C6pMa9Vh
zVLhxj0PFO!zBep{IU$x!$5%f5Yy3q%YTwiyQ68%A(VH$6%-uonIC2t}oP1>0&as>q
z#q+PD85Z(D%)oT04nX=l?q_}Q;X1FJk;iGk``6Fb*a)m{>}0_&OanE}LQ0&(cQ>BK
zzRib4u&P3LO+HQ~zdUSZYqJ#DP*ewR>O+_}D(X;k*)b5ZQX;o_-(<iCT)6?%5vOHV
ze(XZqa}Gnf?d$b*A$aF)rQ;VoD_rjP=m6Be0F4qznz~=0UE%_aE%Elqv+<M81Cgx@
z%HOF(%C$>PZJ@GH+$>~-p|1u@Gbhm#9e@UY>;UWBL&N9-D2mZ!foPeY?X5MQeQ~%X
z9Yl_bqQDf$`g1p~HMKX4$6I`a&9^rMI=>8KV0FqH|KT{+Vgj2jrxNq{W-88&#ks>z
zB9kA)>|TIP>1G~~2<Jhwl&!129|Ql42e3SupZ<3hcIl62>@#p{X^GW;>qE_OWad(}
zICFt%TlTr@oKjr}YI~X_wq10`u4Uy|lG|fSuihPjh(wFiKdRi^y9CX~%&ji?gMou8
zzt}L3ZIwJUyK1jX6{JsVlpTTxtx<7`D@~Snx;NmDcr2mE{3a(?g?0RO;<F;1!c{w4
z6ELp29{&IcdifCjokjR{wGYV1nAwf=H8FZIp}EZBBO*wv%}&k^?Q$|~4IM!{iy-2W
zS$~~Q0{=06m7<KW<oAym<6jloC3082*5Z6C5^ImeKD2samKQL}Cg3QJ8fY!U5&y}d
zHkiT(#*kki$blcLT{}0TYkpt)x@m2aYh<7}S9rEi`PV@6<unTuirc8fq$xpT=GjVR
z<35@4JAUlSuef(R27!x~{xzHdWXhZp+NZi2!#z=_f~yN@UJ6Nm>AjL==O2p;m922s
zb7@o;Aj+;rI_~X!_6@%f?>rW)K98*9&f)m&pACB(Rh20_VA@?2R3#ZjJ6@ZU;g1jV
z7==%?A-{?L^sAbl{(Oeo?TZqj0s1_udyqo$WM4vS^5{R3qMm{?#KfTK5SCL_rWcxQ
z*_I_?+uRFuNDiL70Vn!xZCOMX!wZ>>8Ec!oYM$5QC>l&NcRU03>;J~oVjYSB?0Q$C
z2X~^~Xv5BgFQ$sm7UtoZye&czRym@U(!TO~Nx#0MI9vh|SAZ=vwdaJ4NZ%P}@8Y=;
z$Xl|0`n=a5HpyV__Uf7cR6!sRmpl80>1JOp8@YL<Tp}A+@GMRQ>A#Wo^CGksSExcf
zOrTwyYUkF)c$bh7Rpjn0hC9)w>-j+DT4_4o(alV$|4rf!l#O+AY+IN)Wq`oC&BZFl
zt2XVLmV2CQRW`2zvy$2zE2yFh<BtDj=yS=W9r1w!f4Gv&<AYHT-O;a|Q)mmHm$1&*
zps+l*&n2+lM*PZ(nBS;8(sik5+Q=Fy$S+_9Xh8^Nb%3Dp)_IUA@nL8bm&y?5j)9sg
zT4gk)#PfOmNc$;j_mEKD-|LL?xMyW+*+L~$3+d**|3~x;>63cMt<M-gR=pKdJv7uV
zSz0>>q3>C4;f%Z?)pcPbbgs|p4wR_bc?hfzE5BW^K0$x^H<^E6V@mvbiu@@R?G<l-
z*^?&J>np3(XiB4T(n<XjBD`<mxGYa@HapnfcXglW{U!uGmgfQqZ5d<tz_QpH!Ba04
zyAw3~a<@T}?DgIEqRoIW=6!!0KmIK<gxF#L(-XW^3vxu3_t3XvsUn&cG{D?g0#dE>
zHc<)y3UTxeX*V32Tz1;?=Y(c99w3|2lqQKt1CungBXs9oOKpXLmOX+8E6wMBWJaj2
z%ssz=5Mxz;ewJq$d;NOe0?|6Sgp3OXM-Zv@H1JfU3aLFtPWoexwzk(T(ie5{w71Z-
z-K`Vpds&mU1$nOitq=D9I2O8iSWC8V<2;Z(@jC{usxXBrc56PjOH7n}s<>JJI|k^D
zh2tk4=gSn>c>|kqFmgO{Xu8|;@yA3^IhOx_%+(LRT>@^}+lzQM9CFT}sK1n3`piwY
z)Eg`9vrH&|SYO%A86vo_QM!`;r~=kh;xM7&S7Hk$l)P(p(<1W==~{>_L%<zs8*`x<
z^?$pKyXOuOyDGDya~Qs_&fZu~QsdpCc&NG@BK!A1<&K&XBiL;Oir6-x4eY>;l{nlg
zax5;kO#R}a@c3=90DQ>z@Zbu3J7^wsaMi_;>2RNesNGanU(eIm)(B>vEGG*ISjf*U
zyCLeg?_WH*(I&AU<euo(lk|JNvt`6{@jpnhN&Hm1!c+8_83nHPa=JoK73$5`TGs@A
zCQ*+&f>Bg>SBw5w6-(3p#OXy>PzE=8I6bdS8BHs}>{I<+y|<uhZkWT>zqHEc9sCKx
z%%{pHsd}NmwZy&!Cu(Y}a-K%Vwyl-vzl^$NeQ&T@*=Z{gVJm&ISq%`GTQ_5x_!iSM
zaBEBZ7S2%|O=LinW38Pip@t&^@=|}C5ud}SJ?G*}Jeb1HzsV7)Ev*_oDq1swt?fje
zQunKjUZvww;&Nl<t6gbMWIu1;Yr5$_L*A(4YB_AQnmlx;eu$u88~{laZSV84lXGMw
zWye?98B8=+LF+~7XB7|`Ri);Yt&Ql%kF<$={0bi`BhDRUi3HXFq~XrG6ACu*o>&my
z+oaR6OsO>(WF|6Hz4;!_RXEITV4nex&&f0MPmXJp$H+5E=2;FOx6SFCpOcfd^C%}t
z);GTy$jRPCuUHR1T61j>CRCRxWiPW!$DcYSc2KcSv0MWzRoWkZ`|7-cx^*G_AIY85
zv*we4=+1lQLyUIZ0p2{F#O&DGn5~B;a*^!HM<Vpfv3N%>kAEp$at<GWypH}C`Pu3~
zKJpb1;3v~polniYV>X>`f~qbn2v2fz`|G-2&J%(sreG@2T`BR&X7s9lDM3QRlI8jp
zjKz2!8C>F|y@vV=SCRDjF$2-N`H!0-*SF!ZYnupF+T?lq%QALoUt8M)c7Cm?FDt>k
zXNh8#G-g91b@zYN^o_)_y>NZ(J9`32w9&irunfGJ%zGG-vfR8|nD%Ng)G||IoOxHn
zG5v!{ow?Wqenf&!{+npOt3G(IW8H$131z`+yi<M{vE|g^^?l1+`DZv7_ML9TA*v(z
zuBTKHqOsKeg%g;n|FXWpxfUOw@6a_c3>TRbkf2VPp^9UQHrUHAp-99Lb%ndY()+6j
zfcVa4WPJJs5>p}-l;<j0U0#=|=5>1I&kiqvasN@1RW`|pRy^MVdrn-h3ZfrBme!3p
zxva*|WqVfVK@+YkSu1Knj|rGMM{>q}h(M>G&xn|+wGH8SS*4nHJ_}1QCN}8#;rn1$
zv;*8AGDAUz23q3BwoW?v<?{b1I_s#W+cu7)C@KhwI+0dD6b2|UkXDrL*nm+=v%%<a
zl**$B(nyVtQDg8BqY<T%9&E&Dq@@`R&-;7-**W~N!#O+K`Q7*T`d*(8+e@sA>k?zr
z%8v-upD#ec+8?^cIc}l*$lHD4z2maU@y)%R<t5{`Pj7G*x3zzJ>#5((MY>c+Eqt&c
zqhd$_#{cIwwz1(Q_^|CpdcvJA<`ILY9J`?CrE@Ub84ip(?IaS)f=rG}!1?mWg)yuZ
z!p5I9MFbaalZ?)qz~VWw2+<v$S=u9#0E=@PEX@=}vOJjt+tQj=8KB3$OzVPCvQPs1
zC{Lv!U?5^_6`Zex7_Y&h!n*k_o;rD=zsIn50HWFVqpsr1+8r4WYS2+>;Q}8KLREpv
zn(Uml#klBtnUVE}Aa(PuI`-R*9Bd6Y`wNQ>KXAXVU#eJ09bu;8Hl$cVIhWOJ?beXl
zgQ$sVub4DrnPi}|GWN&0;6*;azt!h}mv~RM#A8)m_=(Nj7o6aD9%^1K***TuDlz~$
z7oc8k8!trXo`GCx4mnw{tnxX2Ycz}7vxo^~?R)H^tFqIP%CXZe@ivA|{c>aI#+~u$
z&5qr&W8)udGe3s&m^)bWbY881TOoLUq}e!k`sWIyDLBUe+J~+F?A+u<mx`+${!pZa
ziJxzqR>f1>_}x<1IpDFU7E3pF3%Z6Z+QDKhlxnVrxLrJYCAx{`EX}2>L{+>wyQQIB
zkRfCI{Y-I(Zz3p<_xr`)G#7dCG-D!~E3;sY>?HOwLKonTnXPy!PtK-joN#C77dIwP
zh-cbZ{`-TQ038_5R@|od8+vEaz@ywrc%>RE+=w4iKVt-&^smR>{f}mbzv+Eda`aTB
z@k-U=uu`e?*(#Vy+kIEHG1g6Zk{c@>AR9EZ`*XB3*gkd$KpqfBD=w<qkr}|E{Pn23
za~`YGHT01rwK+0a`~tN04f}Zkv9PLR@yrEmlH1F(tv&Es5b$z2D9#yn&Pu|L^Z5e7
z!z|j$>lohy>(k-8s+e1}Pj)mqw~LH_!p^5~M8%5H)10JJ3AE8}mfQjNWs@gomBLiW
z^KI?${hD*+Ad3!kL-%3;Akw7A92>si+fHZ<6t?+~=7->j0;iGr^@N~r7MTaZ<yZ^3
zXNFo#`VzK^KXLo;_*0N#J#Zys(S>ek9x$Iz2=T^Tv_fZ@6srCVhMw(F_N7akoHh3^
z<IPqzr}MR(B>NGa%HxMBU~7t&=mDsS6T=jKReFvxXv~ojYGfVprN(Go|2-HcV#_eX
zY!oyIj;!*^GudtRX3d8b#uEf5dG`?*X(?Y2OPxvp#-LfFr*_!q07MpHC+7G|xW@^=
z7hY(7yPcq+?0B(>ENd8!&MIk$R2+lB{_Q=rP+!<oC0Cbt6Z=BKJ20J`!5U?Ms$60m
zwX+swPlaFcUC0bYxwJ}ymdP>^v4L|IT<JwZAx@~kbvwRp2rB2Q+f{2zO7ed+C#13i
zJjrra=zE9f>LboIDq)aw4Nol{XfVaO?MSlod=#(v^or+C5W%4bGO&6QW7pw^44R%}
zFT&sbYiIvvMi?d>{$psFpqHZIeeZVka(56S)}&}^B01d2w7L{*eLELpC+}C-6C}YH
zMcJK?62K)ExD@j#0$T7XfSnnnioGti51j32CQpq>+;av3Q!u@6haNgkCL4JZa(&HM
zFyqJ_Q4^fQHHD_4oi3KG%TB%lwF}&Mk)fB2%Y(#<%p}gDhTsXvV+mT(8thFegU?ec
z3g*Piw<#>Vz;XmSHdW$6u0HGsoW@@EFx#)d1wgo65`>u67&Hq~4-0u(y>R+Pku_X_
zp>RhA=EA9{@8MGZflfV#LBLAGy%ccT1@sNxm;K3R*X!<0RSm58eRCSWRfF#5C|!71
zGJz-juFGMeuu)aoNX-Mw{1oJ2r2$=DFpyRdO7w&4!TdlAAKxJNu)Fw87pZ0-@dL-z
zRZ|S1HEZG)B$!XN{q5Q?OUzqpaZ|V0cnx6GNhZjZgP2|L;k|<{YT*dcK3%rN)tC3P
zxu|vdrK@*C_Pk1v-d5-puNX6ip_g;fR#*dmRb~5M66<llvug0}QoW|hMe7z(Maf=6
zUB?adbDOr^o#^rM#~ux~X<wuZl=mbzM`A0ATImyK*xgOU`oIsmr8!QGb+L;<dhJZ^
z=GEz76)7eH+0id%GnvDt2Jdv2M;kcky);9YX>`PI-5m?Qg%~H@i*D$9+rZf?<3J9(
zWa*kvX#{lUjH7|uo#L}5V>`y_JUbn^G;fHKYs#Lny*@Yi5RvxHppQ=EembXwr*5Rv
zr2=b^&+|2z(L6s3P}MlY<?Q%~QH%-Bm2w7JxnODJ9<S*5-M7o(zG~LiuhGd;^fYQM
zfO&Vgip=gACCF*&(|fM|gf6w9yXvakT`kr3a~rRP*<ydLz?=1&-!*?-Xq&m!A1OyE
zl~~#x@;`RK2Em-9uG)n?5V?g7{CLW6wBcnH)?;mE>$`LB;mHB6h}%i0-5{wF<FM+Q
zEcgX0N&2~NxOx1FT0#kv`@O?mA9&e9<Lf`w+FH2k1S4Ty$%zon2r*D`g&Ik=wm42Z
zyt?6jdbqFWRn|TaE<tk6@cwbW4-Ji4Cjq(6(sR#tvje=e72V9-pOiarM~h{Vw+zob
zW3ZLI`)tE>he<owDse*E^TCT_?ST1W6FSxb^h%?5u{sWE_<JM1K@P^58qlEDUHTi%
z+t#MB2lB{;T){7i2=)_MHZ$=rdvSESbk#3<>gfe*PN7)8*6cqp*q2V(`tauWiCf|J
z%IorYuE{;E_$_-J&^ENSJr`6LKe4YDNTc4*|Fnd3yG8@d)ANS;CTb=6BrY}*R2C+(
zBO6n8PK-5PES3lS8~9s>EmYfef4eEHohPosyT}U|z|SM^913G@sFtplpJ%?AP=J?+
z_0)#cmrez3seks5jStUd>EoV;RKUbyqu)P7PF{UbT=T9bX943)32qxM>ptHkA)PQ@
z$D-7dnUoSAVW$I%aLs<Hje&FdldRI|ovS}EFCV<yFMXnC_qLOqMY78N&_f8bB<aj+
z6AA=BPt8E0y?zzbtA$I;QX%C?t`!eO!M5P3^WUpp$-Rjyv$j!H!hJXUw62;n_j$$#
z_H|;t(K$G2?lyom{iG8-W)R22^tab!yy2sk9V0A-NldWZh47KA?#i`RE~H><LAPdy
zL2U}2>0hir$;3K6m|XU#losF>l>~AVW}KN`g>Vw(=h^xrqcTK`ex>Xq5jzh*doYar
z@*FH5ER!KMPOgv;UW$DMw>4m*3OqPGF?_3j7d`T=UDQNauzD?vFY-(8VUh9jir^K%
zy8JHH6<!!KYCS^FQl_|NbgW<n&#wsf-!Iq%DA|<1>S!T}bR_&iLiX!HGN>`L5nmNJ
zR9vPH`!GGU8QFLgksy7VOKggti~gFaOM2w<`;X^3W=C84Lv5Jmr8UwDTKJDojh>*|
z^$uTO{G6~Be`$0XFFC5*mQP3Xi9GJ}&YLJ`!l#IcrKF!HgH1JRzwBIUDeVZqOkI}5
z(v@1l+D`O(Q?HKk>w2WvtmO5T%yabql%!4JQ4q9asY3<Xk&~$DAG-I7#)}t^?)>oV
zN}_Y!H2uQmPu2}-!qgFr%ibL}@dm$>WyrqnFGWt(Vzc(UFS_^37>;7|HtWR-ZA)f8
zPYWfrPK=nD6o)CSu7>w(8XSzB)0#o!9(M21u4*2LZkSp!j%Jgp9Mi4h#4$5_xffJx
zmtNQ@+t9(I6~WF4UaG_lf!>E(lx3{5DEF@o5qqkKJ$;$;%Y)gTouK%)XVpN6;uO%T
z9FiRUH<y@x;Ju&QFY&(P8~uuIMYL;w^#5N{>zbT?yrTkda(xt7bb5^d+l%A^{%);v
zczck)p%K!nP?Z{ur!kq#_!8f1r4bt3q}ni@A~4xZdrPAM)k>_$az$5HGS;zPt}B*j
z`igQ?xc16uDY*dtMYF?_sx$vrQEQYomYDQTB)rdL*F-jO6ylwN$qRoCjL5pJUVAzG
za(I%BfAG$3!+2xV_pKmNN&c#W(Z>XyZ$L`3m|lysRO1wY$&dWymCu<Pfd#f_0mzf?
z;yp9E4MXcF7}9&cWYtCT#8_WGlVE55x`v=DVBR5-cu@-yx^CnA$sK0-%w0zhHxXr|
z?lJx6P#UM5XdZ&WUO~c=L`$zhUK*!Yj=;<l*9G@};m?bnO&1%QoEY+Tsoe||+?O)w
z9Jw%3XXqR;5ZpXC8G0+RLH#u42b?0R9uF9rD5%S8GdHv!YD0U8n?cuAwe5r_(xyk=
zXOIT1`GFvQ`=TmUXL3QJcr0H}WN{zcM~m?pp-S+mM5l7?HvsMfmdLsV&{Mshh&{D8
zrgA&PG(5pmm!ZLRU6sA5_#1o2ByA#<)5+=YZ|R?m7ky%dA9ha<3y3Ie^kL-3PVxu#
zC~=+qO?LN_LE_cFj+db2TEC_?{%@X9yLrK3$i7rC{PdDkKy5zVlqN-DWkm1lS5p|5
zDg-eZM)x;@^kURw9#82mcPrAqWM}Nw2TKBT^W)2YCuGWU*0$@5f@O{Tf0z>+p!V_3
z-d91JEdFoVGWvXt*c-cF;QpOsx(26dcHr)bhac=D9{ZE+3Z{6fH$-c)`}FS!iwkzB
zZ<uFf^V=$7UDphbC4Vljn>V%ERlL8)#K%OiNcFVG+6#&6yA(Zz{@jA$*OFV&q^=-1
z&EQ96uL&V*weEMHPpAKkLR=pp*7iFt8Hh2hoXe8pc}BlJa@_Bl`q*765`gLl<9CF+
z<DS~s^SZt=4<L$5id{AOe5Yo|KL7J4YGDmq9Avd?sX~ZUuHSHuS>I3Ff4A#z{K$Gn
zVVXMX<O!-??Q^+%$3>GlA}WJA(3|V^xo$wqShMh{7qBk!%T~MY;m~9JN7f1f6hZwv
z*am`DWiaiF8|N4S`7`%6#B;7IXf1=hlL>~dDV4yp{@J~i?z{rtXV{0&#=CwuZy4kK
z#Em{zVXOR~Vm<QS|1+C58V3*;@CSfGx}`E<Z0NEnntp;$ZIRHWnzqQ`aTY=@G!->v
zNgt!)emqR>E9`GwG#EFBK^s%e_zUB4!zd=&w|~6KOpj#e9kvIeCFEc8OGYh8J`QB`
zBgAtHz3b}p!7%>%;>a%j)Z-SB5V9gtlvyJqQgmLFYdR$`$#r}DBw~1MnecuYKyWh8
zEB-Sf+_7YM;C&OF_Az`vkT5?InfEskB<-mz#_v+q{lK7mm*m7(qD`kkQ9*+BYIjGa
zaW!}5Wfr~HsZnA9KSklV?@_uPmV9RZ@88ofQMY|JSQkuPMff94ea_LwL;uk{kV91a
z7<5$J%zSNiJwRtvC);0IjO7|$ctRw1F{&VSN{z2xxaThUTYzwf?o{Et;RsXUntjB^
z0?nuw063@&+FoBTJGM`kM>5UZ+mGok!i^vM%psqYjLVV4gl&RfAM+l+$HBib6LSVf
zgx<sBj<!~fY_;u=bXw0RDtYE!Q5CV!;dy?AR}j=(b;l1<3A<=foqSpcZ^$p;%9L9>
zFcCuAU7X?mfBd@KKeZ}!^`%kjyy}evtvyU-UWffi7H`OwDe>9nYljfH0#|IR!Kajz
z^(0dZ8J^ad+G5_%H13!ok{xu|=3$hIyQ$vdUPGpqvHs^x8`*0=o*LQuY?#5S?VY~w
zikkf6LSmn;|MuWe68g2|50aBv*BolX?6a++3u#@`T>JpAhi2M-|DL`492a8FRmZ>n
z6L%($lytgpVCP!7X=5^~U0J56L1}RL_pC7*(Uzh~1^o0~He#0nhs99t^VNo(_IFM`
zJg&koC(TO1Ri`o(Co}pDh?}n4_l@k1wVP~C(xvvg8FM?tdf8dI<<bD94xhs!n}WJa
z(aa<P-eXVrsOION#Thz4GpNsVZ0*OVu@10`-R`><qRB-1Si7b-gG;+BD7#y#$iL+W
zx9e@Sp`6P2h?{7#YnyK9=rkinvi4(}*jtgmr5@t@WxnUBjXck&NK-rAICUeh{j!Fj
z=Ry5w#kS0U4kcHeSSH@swoK~@l||5IYHTEOYOrm(y=Hzo;xo=2F?81W4DRQKv}zA}
z(6}A__1V7v3Y-CFdefP`lD^#%-1gnvB4z)1T$<sVVhz%__ln(xq8&~@>}8A+ot{oj
zCrWY=G-RXkrWV;AOhE}RO`#KYeGF`6JM@v}H9h@K(AS2SyI-tp;Oji~TiMMW<crpK
zMHuN&Q#XudD;SYuT^fJLLhtcN>3bsL#;nM~$a4jEUL2@Hc2qlW>Z+andWJ$m2fVK)
z>Qi+|DT6Zwj)#al9$L|^d9EHZb4ye?)KklB${XW$C*!~Liyw{lO|z3Dz}K(L26_Lr
zWEtg4eExOZFzS{ixPoa}E(t7nRG{mfFfOJ_@ZzE+-_;sZIZ$i^CRc1lWv?E$H$1<Y
z7eST`6|GFa61wGexvUtJI!1M6fp_AZE^OaaUW#3}6h$|WC;j?i7i$JI6G!p(;!cl4
z!pwU5Z?TpM@9@O>^)3kdWB9HOSzS`#E5RKI*mYlv#s+dTw)C;dl+t?j1qB31?A#tA
ztfZ(-G>u{d+h?25@;2(9l&vQnrYi)5nTe<Zhjk#~U~;OHxTK)1R)?Zmm=hEviv+B)
z-vP}4Elg<%)KhnA>lN?WuU9ej_D&_UgS5xgd|RSx{_&^eJE+`DIN&3h>PX(QMg=-$
z*onL66-hjv(dsUoP$Jmh8-TxQi@WLyuWA;Y|7t@qyhkxl;PV>R^=KTzK~-uPjV&=0
zSS?f6_c73hfoe3uR8&vl+G{;iEXP+Fwu-oC_;<L4CV5nf(?coWTX6HgYvTJQT_Pdg
z&d{z++Thy~KkH&Q)btcz=?y1ZSJ8CiJxzUOf;L-fa&7JUJ+%^VPd%j#udf6XD=|t<
zex>BKHtx9+%_$)%gb<yr)K&iw`29UFwZ!b3Z}sXMW5)4vYTg&uW7&(<0vz}JeC1#2
zEkih1P+RK_{*LzHXM@_|mh++T>JnH!4B#bVnqOc2UZx-ZifgfJiD389ZcXre;=tX2
zCV{J)!HP$+p0VK`MXm@$p&c*)6!-i{uGdlp->$VHr&ewBu~$>ElTzHTUS8f&8B#i%
zo?eRgWi7sR1o$sl=*6{R!n#nQ+N?J{9l`xhetG)`s~WiaPhDiuGjoFz|KYszWr7lL
z9X}VSxJJY=HhSm3|CG*#k*ASNZ~jXxJ{dlcN+QOkf3Q=nO6X!eDDoM;|C(**u;|hq
zoK%3pvpxgG2c781AbreBDJM#}mse_0(Bag%<E8CIV+d*rgUfS3i3zqTJ_8o{ME;{;
zv>$lvxXCU<7TOJHLmTq*q2>$KW~2z(LaB(f5?Edi(pz}X2mf@zK>W;GgDk#yfWNK>
zEU64ub-0I*Mgri(F+B&5eLU#+UH3W+2ru_4MJ8wc$@l+KarJ2(#)9bZ?0Rexb8~BW
zWU{bawIDxx(LGJJi20J1Jd)qBGpW{SkS-z3E(tQ?xH27J7yALR?EKoLlN-KS@;2&)
zOJjv*NwDM=oXn@5HW0Ra$Ab1ygGyJ|@S$Yfid=Q}>REcwMy+?dP33*oLYuEslk+8S
zvcsb_aA&4Cz3uUVd*P5O(!|X3?$R2a=*5BUQE<WL`xl+$2H_`MCQ4%0G>H6sNkh^f
zZs!%?JrBA4Hdes7BtAv^4pRnicuTqpS$Z9Iy`>hm{6nysm@atUdS&Xf97#alA5DG@
zPGB(IJ%t`iF36Gi4fl)e^So5Bg`G`8+TkyqOvO6J1H%6XfZfOyC_Qa=$&I5^f=>o0
zMwkCrH{Mz+%D)}nIRO#i-A#OK__6xmg}ACShs&+=H#%Q6K*HDq#?BoZUc*+r;O5w=
z*Id`I-V<8W=HD_%ZB8C8arJnfcMU>FwIBH|;#Bqn=ABC;K@WPI_);V*T=Z8&F(+?p
zp?v|*fMLk-&MRc3m9|afTE$g`R=OX5e2(2M108q!GRW+3N4!KzU@rdEBl<ddf!%<{
z!(x^_m*^1lBJ;ljR+d8UX&DKN+_&Z4RR60PN(&<UB|-1IHOQqZ2Mp|lb~3A!hROW#
zUJ7$W>iSo9T!{O1O{zasf5Dai!+eK74~3OJ?JS)DWcGe3o2B`0#|^hE?RyTw<T{p0
zzw2m0jSe2w*gH^SF*ODg<E2Zh35Pj$Et|Al<E&+`oc#<YJDO~`r_K@o2-oW%+W(|x
zI<D)R&33n|tE%Rl{p+b`AxbTBHWzvUf^9^a-td+53V2YV$(9I&u<QGU`*-Bh*y4^q
zb+$XvW}qudkY87S_YRfb()cp;jj?IX@jn_*{vk6}e-5JZk>}(3BDwIooSLM<Uq(WE
zRZ+<}cu;Uv&bwdNK;`jc2j7+$^V`}Of62gL8STlmBjnGNX9u9Vt;)nLqmrF^vBpvT
z-@T*J7yKmRYSzw{Nrh4~iZNsWy}ExI`0pLnLE<wECGFtHX<h(k&f)dr1<^lj8<4`O
zyZ~nNwXvGi+9<Fz;G0bL9g6eWZr13JySr-U`&(`jD)}phUmB8<J$PiMy~o4!Q-n@M
z@B%BjB%0O5J&l8Fuv<vnd9E>Enddgj#OR8x5^b952kTiU-JyjFZ5(#jE`X56$EY#%
zNTMDylY6{*?^;-bH<uOgF?KqO<pT1BzG2&4!9Hy1s*HVwrol3QjA>i(I8Q!iOyiRN
z$O6dA>?4yZ9o|v0*_n*bqQsc-H#}k==dxWQ)=^ph4$mEJRGm7+FABa|E`yvrn$=E1
z3+ZJT6>y)UkBsM<n#WxJK8o}a-zcjHVnyH`t~Pp9OBHgbqe(W7xxFnT<{OoJofq_`
z+i(0FI_^&k4sV#4O$(T1<wl+-kG6(qHyCe(ZZyqa-6?BHoq33AsL}pC){oplo%HRW
zc4vQg7QMs1U3Lwm)2CUtKc;Sm$+XTn>xm|(EM_#VuJ4weN}x&^xV{2IejP!x_cxId
zBO2LGtny3(mjb$ao1eiwx{S+<$r}WZ_a{(n;JE<)W#!Fw(ddpmtLeZ!qlNzSx+N?y
zKo(;wcM4X-I-1>SO02N4ZuX_(!9vSIZX<V)W@Rbev#`X3=cY;Te}WPiI04HDY|e7L
z#~(KDkh>y!o+FhgdF5jaxM}-G-hozy`2GbKbzz#FWH&s{B5m;g#5iw7=DV>L!`o$G
zkFlH2QybxLnO+p^<aAcLP*HcTag|WVM0b7F@CdadJa<XQDC3qhbe&=)_u0MsLWDGg
zjIr9j_}7jg=tduL)2l?6<xK@;R-n&fs9K50OM_dT#;tE}eVCPXS^cQ8LmNr7MM_`!
zO5=UOc2YBYEfT)V>;nG$HqE%3N1Zp-=|Lw@&v6a85V>e0XHQaI$!+eJd>LqfY4^0t
z_JY36_33*vO5Tui6%ucEas?EXYZ@c5%jBG!Y-x3t)W>fKwa)EQ?u8zX-C1!i9VyMz
zDe>CfSvP!7hu>54J=awg2rp&%BQV+{GZOwp0ZR50Ae?AgQxr?79Rvg9rQ<1_!S%r6
z^oRg^LKff;Es31F4RoUmA16OUXDs|Nr|*xI+AovQHN-!Pv%r)+?xq$%%26YD>%KGo
z3P>1SGk#r9-Fpj`_b;M|Kl2IJzK0jH=lWVNT`hXPpwD`PSg2mm)l_`fj!kc^D~Rg&
zsz~sh$mqUnrjdTBaG@>{)UPdKUjvF(u4$f{E)kkbdbm_8YvF)=TaNb+eEH?rjHhZz
zOI5Wg$MyRT%w?;_`^5=aLgs6H^Bn92YX9aW?2eJteZ<sNy)O>%!Kig#xb^d`ef-ot
zV`=Zas$k#neSW=jche-8z!T-O)$o-$DL}F!PuNo(#55N=`O65~Y2Q^<ld#`tTH2&K
zcQ<Te7H4Z6PcY3ce*TzuIg%3qHPH_-5VhcIXxQoYR9rDL!On7IV7^aef<B=%_YwP2
z?&<V?P$pb4^u>Ihk_cLA>HrJuJe&kN=6`@5s~{X#AL-XH{=!Bj-!U9r%zxImp)+L^
z>K^e6!4kpreaU4iQk8#AOKD`Km7Haj;;9qu+VoIpVX<RDo6;{}yo9|FB7NwezDT&0
zh_ILunK|=sQZc8GVVjYhL7;|fI0Yz9Qgo*yAHp-5pB)XgtKXO0`#pHMqoE#JXd@;v
z>n6WeUj$Sy14u{kXO(R#Z_-Bz5>J9YZE#uPtt|rL6p;QNC8qkv<g2IFDg)>TIfkiA
zTTtV}(0d2129&wHJbjN*F!yV?&khdyRJ;b1ZT&?DsOn}(l*~cRr67;@-alWEc)d$4
zhFASZ(~xifu8;q`v#!u=(R?_)_<oOwp!TI?uKn**pOPdl$q6F<Lf}_rJ&m*U)AZ<m
zBsPI+>FOJ#!&`dME<a0MRjq_@&nK_py@pjQHAP-@;*ocGzceTH*7{6{Hb64ZUf_Ih
zawbPnbsYg!PIyCcujrR!RjqlP%8VrVwKj972*=${S2zuE_=fSTssv7Vi;Q>%7w?tM
z><#lEjID|a4b7CeMNquBMoOpj#@V`Bas{&8`vI>Xb<4z8gV{YT6EW5FfR8uv3Gm6>
zGYxvHc7VHLW1-%GR8ZkQN0G8UihnO;pkV>wzSb22W<ZbZSgp81a;wRhoiR$rtGMDF
z!Ze$1cwNchOP`FK!@Uoc4toJ*#CqV6*9LpJzAL&K{^!x@IES%uR__e|aAXr$4W$(^
z^EIaIAd4^HuXwu$IKlM*NjUn7Ez5C>P}gmW0*5*mwcnlOEibp>@918hs85#&earGG
z5E)S(0}kjEwy+OB&C@~Tw5XC5|2=3t(QW@`;(M`a>*c)vgb<WkNk@sI9zQx2U;CDQ
z1{*|4WoFnJ+|tnjQxk8rnsuCeTsm7%Wv-VOVsC`AiYgEW5J}w}Zz>l+MS)%*O$Qp|
zWWV+H+p7l^l`md`LygcH!fDsqquGVi?qmh2pihn4Gw;bcqbvhU*>}oR02xNBx5TVw
z?7-`FE8(MD_^It*52eF)$Z3aQ{G!VJ^7oJTBlo<ih0b#>wTS4OMD(cooYk$Gu*O&U
zPWsE|f=1U|&3#KpzU#8(ZPkqh2p_S`>xRdHonju8{717~6-KTMk_ccYE8LmlV~cl<
zJb7XD9}P4p@koL4>gdV#Qd@tyQp=-@e)Y0%|IxU(ekK19GgEwml0^?M#sv@^gz_@J
z9<!f2Ji*IMDOGEWNfyY`xCw?oQ}qmcE@L!`>_9%Y1@ljnu)Owim@G*NNd*h_0Ctq1
z@U3%0ZjBoMPYn&ww&ld7z}%~;PI4phbWz%!Si_pm2~W7O27SmFc|DNHUK>HF-aJF}
zHD*|}aGh3~?vz_5pd`7hYu=eX;A&+j41~^s$KCtHwck=CGyBu$!V>pdvuJ`W>k3Oo
zRJs%`Nu+w|$93*Gj9wueu*0->m8-j_Fmwys96Q0${Hd9c?>??vO8VQ<WUo89a#Q%M
zA&^psSB>_3zF*R3+SuT1Uuj=H`3Lo&sx8AECSk@LSakJWdK02KxujQ3ZWKDZA@>Ct
z;Sq@7DdWL{g&AFx9=Rew!T;EX|G~J78hBb*R`NgMOS=>N$Xu*H(yedrExB@N41?aZ
zpD}BkNc1@)_`G>O25@Srwq{?F%b)N7LLxDqa2exAqnr^)cG_os5x3CfnhN^nr_0k=
zwKl((blQ&Zuh`JPFW~G37^2Gr@0wX-v+dMZNC5l5Tk+|B`H$b~6Y4xouCvq0IZ8W)
zT^GE>hQdya<F32d;>Rtn8|7xv_e|d4P7BLx1Ho%YUPp7o^NIJ4b7>;{jr*E!2$L7!
zajNWU>-9e}w|iTbxSvMkPkLHVGYa_c$e2=}`qi&O%IK%yy_as?zX3VIoxbst@Y9{Q
zEJqc0JMSh&%a4q1WPaWEjw9PzC1@;!pY@zq2XR-i+?oDr`uOuUtBMDq6^<!Vrev=X
zi~g>woO>yVMK795No)$7-E=3%BTh#@1hU>Zc2e3bsQ8j!2BP6vG`_d;M}QoTn#dAS
zIsU_ei4lb;@eFxEd?Z2soA>MNB|$+fd$RMD%wHx+sVc7UIJUoD_cvKRF$B8=XR>7`
zYq}fI5JOLErz7pk`8QH8s7AE06AkkBSDxy*{q$KuejT<4OZ6(wzQ%h?dZLbF-2&Iu
zQq;|<^7I!<u>k<*l7191@x9V4%koxH8Su;axBIymP3<aXUzYt#t=&_xoccLop<ZWa
z)1uQ?(qlcNKY|iGmhVt%FqF#mb_3IRx5r<o`>UFKpX|pQAo*Y8!$%)V8tdqPD?4l$
z_F`0Z*%xDaiJ!S`l2d3YzmBPCj|flZKzo`*dv0?7qagU{RJlea#ufUj9c}M9xU+7>
zea5$QFVXN#;g}Mjb|{pS#tT#Hb88>h>)Y!SRUfN8z?4KOB*;9^<LPjm(%X~(R8~Q2
zK=9E}EaP*@mraUI12y;Z&he+*Dg8ggmC9day%nws^P};v)5wXbPP{i0isE5{Plw9C
zEq0<wU+%~Yf?xPhae}{)U-C-(yWSw-VNWgO*$G<s@PBLJOFsUo_x_`)93{q|%AqmM
zb%B7n^GAZP4{IuahWu4kehdMvUKFRRCp*?3ucNz3lCB~k$Z_bfd`6Wop|b8d<iL(E
zlC#1-iePHH=muFQ`3xx;>N&79_`4c<J!ksKmj?B!Z8XfT{e#t9JJ>`*EZ0Nf-7owD
zk(8jF0b-V;=<wyw^T~ZU*~s&CmHU8mHF5XS;C$7i2Fe-KHnmw;OpBxjgfP-L(XO)T
z5FSE%efJpbel|9G4m2#uPB5sm($B=oUhU^u>g}AsZBo`P>`1loS+rwhVRC9sRu=vc
z{L9ThEfCS>dyQPV1hZ_$s_=tHhVRzA-VP5F0M!_jHF>c{oX+^~V*#qumNL&-u0+aZ
zKn~ihAxX@2UF{?%RXe_9NCeiVHQ6%c*B;LoL>~nf$1vH+7ur#DCLCQch1bm(^egWd
zbnBxEFGGN$IaScWh0~virP1a>SdQm|#fJZ*2?f1XjuO|afhcYpamwr%{B%uGSHsmO
z9eJu<v8Nam0AgN{%3n5K%FDaF4&_!@(@&zcnZL6mi#wHp7MCuH6pQQJwe`lk01F+F
zyPs+${)C0~0v$#VQjit8sOmU*3+^+tVSZ5fwC-~F5FwiLszHt2J|67Svuadx*GH&J
zQe@VzgnhN*=m`%2l|NRixOR@bJXfRMXI_0zjW3@|U}Mk7dAub2&e!jUIceW7A-GsT
z=qXJ}#CHTzg`_P(cag-MM!&#^hK+s|VdE6GF`x~W85k5+7+iyf)17t{><|VLyHVKz
zgcX1Cw>P%#Vs<)$j3l{U)=KC)uRhCS{OBqGEuePw8CWTw0+jPn>49L&k)M;pzv5p0
zQ$0kiPZxPd|4vuYce~V86D4JXS@~<~nchf_6@50{TCXUci>CisyI6tDvQrIBx$gSf
z#BgDd^hpf$Z@RLbT!@bw(tgKRWCkE%n+kK!QA@tOs<kPYGs-_M9u6q*-GMq8l{FKw
zn1s=A^cfzbVo|kd(SvzuQsY7*9tHEim7!fXZ?E}XSDLBn+@FGfR42y0u;j>CaW}#4
zC*)`MVWnmC4paY|it5-=))nM{trg-)t?Ik#k}K)!xFJ3}JL!IK%;)OvK5sZflz4{Y
zmpR_CpnKC*3X(*=2Zjg92bsFb!I42tD>YoGP}}7vqD8`#eB@5euTcRTywhI!(=d0|
zX*AQTi-ePf+05H!Yty%fMO>ClZ~rWaN*6Vme^xj1>GaJH{*gH11WdT<66RQTA+$}n
zyzf3$ABc*N+85OJ10~<G3mJ2a{sd4=a&F$>x}WhpwU4(Pqis?X<vN)cT<@uS%N;qN
zYrpRiO-ZhFfXGvar?hl5Y00>fcgB!$9gf0X!I32!Jd0NXz$vVpOe8CUs?;}A-XL_R
zS8gG8DID)J2-(OY&4L+KIVJ-&3c9Ar#il=AQR_mU^O@dun!$iO+v|JpM(DcujF)`d
zb?@+3&a-@(3ZYP{2R+R$M2=SHh5gl|wJ?kS2#UsdS{e~=a8!oual}ujw}`GWd?aTH
zS(?*hoS(gh&b^&n#>90iF3wl6DCn?NjxMD_a%R?cXRtIVM?_HnG&OKx%5~ky3)cPv
zsllBZLL`Jm)I7gw_hcW>-3UZebXmC@92W_YT)JrfdfvL<zl`vT?sfo9zJOS9Rtq0W
z=E~geKkXZ~`&{QbkGD?c^QQga297>*aI<b4>kK_5P4~H0@BexGd|}<<xo=nxq{ss#
z%=Y}x$f4<}$ku$J3%jFU(^W<NuF$B2@{iFd?0DFg`C%A816S|ts3wwiSG}vNRMEV>
zeQ5c`eqr1oId!D%)BS{&G9l!M7saT87~{rp^W$`_Hrwz9%i6+XC)St6Ac4Sqx6Ayp
z49PGm38b=R>p9JnR>kMA<=yTs+N(fjkz-pSG<;G}@(9Gme?>LImM@nV32Os;5)FUf
z@v6w5V0`2MbPSA<@r{>VATJhJ0o5lxZ?wZUStZ}xp*P*_7KliJm!wKVz7;6-iQQ@T
z$9gA9E?@WAffJ4GLFuvgB){6yEMN|TKc57+jaA0h#>U9d?<UU)hy@g@xW`(K&$0*e
z>U?`~nl@S#3X6B#6=mlg4Pxs7;eW+&{Fz<)@nKteE^SU1viIaO^)aEAIcHmS-EwOC
zKN>v@YY1&E*VH<q99H)Lofs#t2upkK!cIGYI95F<*^XOBU)kc5ekt0(-MVh2ji@j2
zH>v{^oVPd3XTDXXE4E=c1~sdp36dcB&+izNx&Ap{`b$;3EeFj@e*&+qrVtEtWBg%~
z6V&Jla7)FJRj08iNI^gL%jTJJTq*nqy74{#(Ef|}d&6381|wvPTnVbPq1lqT@uNXm
z$MFV&sz#mNdi!I&8x)B@F+>*y3!a^q_j6-hIska`I}fm^8^K;$!Ca57iLlT*NRGr0
z1)#Q8P4t_aEI+;Dnm*dc-xZf-XI-HD>OuRK=4##%b5ms7+YmYY-C@l49}UR&1hn`5
z?H&4zgc3<{2VHDuwOowJsOMa|t@5qJ&YcZ&pVYXlvR+-Ri^+Wx6G0|$zHExx=KZ%r
zl~1op{4tJR`gLu(Vqf}G^=SXh7FR-pr}tf&f=_5zhcd&r#adZb-)D!Zcz5(a`LZR!
z*NV^b>sL2gy4Ga45_;Ccv0ZB|*)xShqet8@N7si?7hJ=GEi3gSD>$ZoHa42IJ-|O}
zrqOpWM&r5j8EHTBGqtS7aVPOfuYSM0Q^#KiB8-K>8^tX>fUbx$rrieg>_!Y%qBQ9@
zV?*(3!ahQ-MN+3D?1eLGHxB=wu@u$}{rJE)wWYt6kCGr;H?3T@lJFl6nA?IL`VVhh
zI5jELVZ_aJLwu)MOB2a7<=fh}PQ~(I^nGYlZLy{q=e>n&$;+oub&)H9PBq34y&wth
zZdTc)th$C3%%g^{2zQ#)i_DBDlcVx{on)N+0-7w>zMETgK|X|V9~3lsI2iCYDNR)O
z=ZAtBzOUPo6e-0-L@2rL3+dygpm-5#MO*i|N4f+i9zB9A7WXYtTb>&AhDXsMGVCoP
zzq(2VgG|Ere9^6^vP>xjol9l#L6u$e&Gk2_wx=W|7AHXLghR3uMDLZl&`pNt|Gry@
z_qkPb1uDHUK)8ST!-Ncn0T{HUgPDv<U;vjP-kL>i7Z@7jx1@sagEMY+Qq~PR-L5Gq
zvxqnKmonr`X{)u{mZBM2*N_^rS!fQ@b+f{Q-D{9ne_(5(RQ&y<-jdToXp+Q-Y<EZY
ztdenONpe|o&tV-l?)kaE*NfnVBGE0aNGFOisz#i(O8@KyHZT9-DiH_=^5{)Hbpci-
zeR-n!j=}j>1F*cZ)-HW-c#QUa!N!YzfD#h?&+u96BIy!b7yEMNX`e2yq>Sh4be3z1
zy8*2S!l=*_SlXt4saxyPr|7%+bJPEz&~RCczPvLLcg9nAJ$3OLgdu3h185aLLyeFx
z89rEwzg%%v70g0l*fZG3qwNH5Dg9#wM^v-Ckob>AXz)rm%y>x)YO*`if*3JBz+KXq
zm+5OvLjQsGthR|pqgycxyr}mDPsWmCO8l3)R^fV2DdD)w0Jn)WW<<8TFFb;3cT&Nk
z!;9`cGD2f&Z9Cx=e9#e&XHPDUH%#t``gGwLj%zsR`=&H>7AJnyCmXu}r9;F^L^??>
zSZtXN*J=F71sC%2t_l_h?7J2}Cu1WVh#1<k?Y}{mLlKuuGWe6tE$-ZSX9m|utg(Og
zDVM*lN#_m56o|M^v%K(J9V$YZ!mltGdq$|+G$FP~P+$sEhfvx%gBVCoO_WkPO2(Ej
zw81)dM#ppt1ftUs0$8!Fm_zz%$INn2@o|5FJ=W5pL|gVyL~p$zw|Nk#fd4>DtXhGP
zDb+8AhK{K`o|)`_U-G^(O={JQC7fmPSD$TfMCXe$Fn@`@fyNgoK%O^Lhc5mmJCA%!
z^=3#g*RM{DxeJ#hAJ>71E4>PX3H_8p`*trjEhO(7(7&BzyqK-4-`w<}`d76vlN0~Z
z<gR66lX%;9w?fG2uixjBzGaf$9H)lKs@8dkfZvltsPL}5?OeQdwXZF7L%zn}hzIXE
zTp~hi+MzQqc7y95;}xsF;|n9M|0M48JTw&^EPIlPI9nLnyfrptvWp(A4}ct{xX)2;
zIFFudb`DQh6|p>BaSDx#d^0{15PbA~l*OQ>;tV&6Ilk3~8J?0jlIEh0ovG~@&P-|a
z@z2ko8)qCS;#;GTWnw-BkkG)0U3LXqNlVAN1O?lcS!PTI*gLX(I^F$R4`xFyOnpYF
zig7W$hxo1NgI2^+S}j8aTdwaDc7GCfvJ30Yzlp|_RT-tDhjl@z5@4NNauFT9n>Y;h
zpsWc>*|U$ipKA>u$dKy)(QqwHmuI{DuyC&F18_81jD4KJL&>)8X0V-2izytpvff}8
zb=?9~;?M(*Qg*p}1LcYN$1eW~F|l*=-lkxC$IbpcPpz7tvH!}DSIw8=Yw>jmxUoJ3
z1Bgs|pV>J4s$q1+z481E*$`8yfvrAeU4YC^yqVVCcTE$dmIZA84J&jkoc<P4UH`Pn
z+0N`uim|rRAv+>aL+l}IQON@Hv<-KpcB#t!sql!XqtcP6`=<{ie||GH<G6&GM}21q
zeMfBY_fwp=0;66CsxF!<FaIvfAfzQk=2&f)$d)oN??zE^IP>FxeMI)$?qCP|9%nJ9
z$BJAklc&<5jk~<9wWVmr?DD^Twy*C~2Ed@wH8=JjlWb+2bu!<DSVG8kw%)uyhFor9
z9OVLVAto52{cLgUM~kG;4fWx%kBPjH-&V>shKcg{e|@o!U6e%rmgTo-g<RPg9CfvE
zl+2nk-$-bVQ@PfKEF+HG{%3C=U-)FCqv>m@@L7^N%N11r#6czv<y$1H`n5;k-_s=i
zl)D8lmYO_UZXIX}J#E?Ogsc-c{RsXn`(}htV=vE|YqXF}OU4H4p%=2^$HoIP-wlL+
z)p7qE_h`bAf80Mx6ox;Tf0k&t$FwkAliA`q<ddGVE<^3GQL2@#$GceyjFfr#OXqzk
zrxUt>l+ABIH%JWWougSL6g;XEHvGop`D#d-=Z1N=M=sIBSP`O#(6+X4X#3)nv)lc+
z|8JCTFxJli5@>h&){hVF(p)3EF<!IT`Dz9Er_CyrHUmfzmv_QWH8Yh%;ttfAtT4d1
z9cz`C$Na_IIL$Y<?00R!RXR*FMB3~gO=^%U0yPn|=0YR+**aIga08<l;)<Xt=Lki4
z8gE#+29V!4dv2^wmry4CNuNu-?(fb+Xt(HB&lC$mvCz!TeNg+WM^xNFjlHz*Cr^0$
zitW$_TA^s^Dy%2=$k+x+f1akves>MraVA|Fw2ca^qm~(Mx38<yDE(aCDJ|?aZ&6RT
z04&9@qgE#Te>7zL*M+D>8|zlTRnr6d$^iT*<rvzvK_(?#nS~9BKK{|Bp0IA7PO6Qv
z7t0`!CWCfmmc1P*+z+Rnx*V4#7PCe`!{50G8<`B$5clg!>V}q2`0Og&t3GA~CwR}?
zf|!=LCiuQnT{Ok_5Ei@wIp@x8PM^VUR=5~il5*Bx>;1z~^aL9){CGL}b`fah2jh-Z
z;!vA8jf{O&?2X-wp(TIf)?vx1Y<ukokn?hr$(|sc?$Xc_$)Y2&BPOizcVwQhzUyKB
z-ug?RfL@nh<kMZU7$b#uUez=dj<l&8TwN!48vk*@eT4oUxxR|Z8y^`>Fwok6#7S?g
zy#?h<O!0Q#vw>)6eq1+?Jd}x*i>`8jq<eehX15ECzU=HkgeVOME|}EaBZZH$FwOWv
zjD+F?7vSz=mENGBPE}zNg4aiuV#7b9>wuTFcv=Hr5thh4pBSncR-(ek--HNuYA}w#
zs$_y@=J3~D4)!mX0mf0>_3n!S?x})chf`sf_f{IgafE=JCDGhrD4IUqMUsj)nS=X<
z=^}E>{@N0Xr4VJ3Y3u~o_uy46$Jagu*6&k#`U(<P1k<YXd`jyzgWo~z%V8TpUH6Y|
zd6dprahy|wjIpL<X4I#mM!TTxKD8(>*N`ovn5pCww&B{!I}x!aqmoN9Z;_PA8)%L|
zgyEqZGBs*-9;i4R&Yco#@2WE@4;QwZM<yikr}Di@5_)lve{FAp3!&kx+0gXD(}JIu
zRcXjj3{_ZoY3F>>a3MbsIBF=!(1jK-gtTiE8`&rGib+0LINL8$1&3$f;13i4x)=W9
zQV0jvRW176E~xv-Ve!RVoOaY1$=-q>M7JezG-Bvi&{6dP;KdhQa2Sc=wL@xNK-8|!
zKscht`&HHwl6(okzJ&Tmw(9AQq-*4iW+`{=F#(!l)+7#l?9`D#XShZ7$>8OmI_|Zk
z+T`NpGGZ~MXbMx__?lTE;jJyw*s2lAacSMuz4V|@;lxm4oImelU@=YQ<;*1DF)o1k
zbQ9Gfct+2@xkJtWQ0}_>k1#QAP0Op`i_yndi=XvtlSMlvx-{#Q=Cya-p}HWU$UXZU
z6hf0ZSs0?T_5hWwGC_0C1t$PVvJXy3UoTMKFL{7uzoW_1iRx~79!R!Z^*?gzwRt$(
zpNowQiAiPGFq`(Deo9YY4{MC63gtW*b;9Fb7gVHW5AaUZev59-AmugO0Du2;GQ$1?
zI8nG9=DQqMn%>uB0!cwBy40U&(t~qZD8*{d*=tMtQZv9h=T`K^yN-3Un`J>S;nM#<
z<Q<SlTIsfg4ouVX66<-$9V)-2QXX`2)fG#H8(1t{9OKr08(qD!RMd-|6$&Qbnk#<T
z)))v}NNlwKL78mgQ?aH$mQ-YxDp0u=<0ZDH$|EG7L!9;=)T}HC`x^2vCtMF({IU|c
z+ifPJ<+^-L&qId3^|NV^^NtxtE|MeV%2Htpr%t8#q@Qy~D|4ALKvft`u1~AovMg}J
zD|{1CrEvf)d43>DPK&QYdEOdQc9#;#+D!tFU6U@Sx{iWUvL{x>!X9cZH3Uk|CAmHh
zGN<1Q<XJ$J$uW^8CMMNIdfhDoh3xzh@BHGl5C3M_vgJB(b=94cJlmKntgg|9tZRK#
znY6DPao(5mTr!KCWU#&L1gO_C2tw}1fr3>(b8W+izjqu8YD-Rk2=9s0?u&<Q<~f`;
z1|mW{)f+&}>6Q9jEqTD#@eb_q|1q?M2pD^@3{4^+KnWj@Ui!+i5(wtu;)|*tWuT{h
z$w~JqHGvL{tu*-E$%$ld-H9w7`w1eN6mNY8>YPA<HgQ>4B6w9R5b;!W37Wc<xImkU
zg@!K1RfqL<ryg1e(4x$Zh;!vz+L6KZB3+A0KePIx=gMr4oC2#S&F;obd0I+c2S1sD
z%=k&3O-#Q|{E7NR$OWr)R5a=uAf=+VZ@$w8n!A-R^fBX(7elW1emW>57#n=F`&hoa
z)P&?R?d+P?{uVU}K&yE5?9`LFzBJ(4DGU+zwSgKCWW>$o7bW~&zYZW?wauPD`Kf(`
zJ#dN9>p+7=cP47QR8y}vY38x!MC+;X{hc?6Yj`&A=5(%E^>@SMB)#JV9kfvfODYTH
z`oL}p-ZR@7?GR#nM}=2jV*&JSz6n%9xV)Yb{H4SZ9aAw9`+;75y=I3FQy-PiIYzIx
z4)@qo>CiB*sQYOW!c&5(?L@5aHyB0Uy~Yi1i6TS33s`%&*Ba}h3dtT)-mLRg`2_Kj
z%~E^#B&&CAD|eDlF#W;6%?ZME2K`PZd>bz!u25x)c6fVYT(u(e4<<0@QrAJ5N>pdO
znWs9BlTG6I=%c6`)DtpY^gKi6)uV6$bMDGG=2Kg4_{NzcmQz2ZXj0=sy7)q1yUO@l
zPoiQ%h4>^FX#|T;X*Up=?Ai=a1|FL&ujZ;!%tUatpfctimau30q`kL3S9O#!Qk%ea
z#LdUV{W+Ancf{LMW7-8+M3`tx$dX7P64;j3#dM|}0W(Fv=+?vA)5X7g$Na`dyWk1+
z69`Xbmg{*N0|kf^t2T1=p~_s&Th`9){k{O|{Dw51%KzBSA9H+W?J;tq#ZXcHwHYGh
z>{MRUgYgh3<2ZO@m6%=~@1pQ5$nnx=jiKta$sui;RFz2pJvc;>7-SYqSUxJEDD4+`
zYUf{VcwamzmH<gCQPoi($Qg>eXC}tIL#Fem6ha!qBDnEr)sV9{Lw|1qla$JKzkSc=
zb`OC<!4re$6}8saOTYFQ-j6P7eC8n%6gV5Pd4_awPwu)gkS}W-bHv`k|Dxq?H?_=k
z$TfE)*Ff^qj6Z0cB|nsXPeNfm;CGvm43^M5-uiSx7i)!?L2h(EnE-4xNqoS*93d#n
zP3F6ELloUDIW%eOrWOl(l>C~71~boh{TQ&wBztf}2YN)lFSmSM5*;GJTwkpOe3=_s
z4u1r-l-vv{gDi1zk`#wV`#8$fWiO!QBSok(>56iKlsEn+n?lsKJ^s<hoFu3H(uQ6R
z-@f-MsfB@2^DM$uT!fXAz@Hcrw``FbQF6@*b(lA!@3|iCv)22%j+%e`y{{wgBad2;
z%|{E7BrY(fial@LLI*=gP|YrplFW^{el%bCtSqz$WUoTuw3$}1r<Muk7v!?oLNjc5
zZP^^Y+peCFdi?L4D<fWEqc=w(rCtw)Ru_}5Kzjm(d_i@LUoBezAz|2vUhZ4x?odsJ
zGe|ZgrtGEQZU=C&=9+N2!`yQ1m7$=^enBCF+L$F($#+Gt5dLonAS$BxkHk*if?t}+
z4h%y0AgX-nr&a?da>XXSd4!FkG|Jiac$~|yQsH5X<K%4Qho(I<cuSq^Z4yW9h@ztv
zTw#zKEBrIuu?G#a-7do-sy>jw5onl=nC<UU!30uq_h68Bk%WH(6!w`ypq2jCBcFcN
zXiu(msa_p#QUB3=<2(;{Mwh4Dp#Aa^AvF_0j`}M$`!A%e$DNNcE;B*frBuQt6mAY=
z`8T8)YK_8`3Z>VzToam93F-v71XrD@^^!K<WC2Bb7wvY5%c|k`T=D=wgAQPNqmwqN
zIGgfph=wPy<yZ>rkgg{oZRD(T{-be~8$KI|cSU?#+#N6vr}?5XHH`UOSENYVFI+ed
zq5^uB1W%0;SSoKoZ2^<x<u+3UVKJlOsJMD#&$?t&*4^mC7xRs-RlN+s-td6>lF3r1
z=e$sg0h3eY?noGAI_9o2Yqr_FqgrR{Sm4h8Qno<j(|NdAm!|;^bx<;;<5GBUf9oZy
zW(ptv!9A!N%Tk|XOE!AsqxFM>ZdmquC+Mc2+0ur={%qhx9_^ojB*3b~ewYRwPqm&U
z8^zh{0RPQ?N?S4gUh1dwVYMqc@xJBQtgBY3Kuf=liKVbO2KddLkp6UHY0Um{*}$wy
zIQ9zPrQ^%gSa$KPv_WZmPsqGCyjk?n3#XRCFki$pw0B>AJu)S%v&k!nr|yoq%QC1>
zt<^~}f$<-W%_>7PVU;;OeLipWcz9%baYWD=00V5E;>yR+T2J`{O&vU5-nS)rF>HC&
zMA8FICZ(n3>s^9zes;Rc>;CHItA$}#M1gm$_@xmb(lm`yF<`A@Bxl!7u8Dro#UoyY
zx=A{4Y_5g_;YJOl%&l4MGp9UKaBIsSW+2p;(nKd%5}DAAi<|R@I_LvfA}qoK^2nen
zcI1-1B)j`06_1zlLe%}TUkz`Ks+Oo(=?$_wB>>7WXXXZ{|8aEQ(QLkdA2+I6vsAPw
z+R{Oa+M6zWkHn^}Rf(-u%+hLW7Nx}AWKbi-uG)JhS|s+az32D2f6rh3$vMusPq?q^
z^Lf8tuf)Q+V9mYBt{0+wfJZQYbuTElPr$Gs5$A6H{UO=+_8cx@WTCAiK>-ppX*XBe
zSHyvuR7*8Qaw4^u#X!rnSYKq&FNo2qB~#k2_H_lOB%&#P_ru_$>@AD-)b3jc*u*4$
zPr+z;D8s*{Qmf81tL2B4c_|x*C>H)>Zn;w^G)HuR#v1R^Eai`ZfBle{>GV{v8;LgI
zH-g*-VJsZ-03DpHIFvsDU2jZj6+4Xd8IB$`b+t4C7JqCMDtqcpgc{y0K5ZxiE8_(?
zZ}dBNBw~oxDsT4QFHmh^C}#14?5Y{3aY+3d*inb!q~q#vJd&LCGSecxsj&4-jrV<k
zk)d+>9e0?bEx5$XM9AU++dg*AOY}#iiG#9Wl8I*+=|WIPRapeilQ)t)?U-Eqrd@!e
zXetsY0r(TpK18>Q@89o$Q0~7mf2k!U|M1$kT#?2)f!^q>1kucLzk)0A#${z7=1R3(
z&Kj{qyhvDGP4A&Ef7T{)f8&KHXL2WXDfrLlMvzxW$!71%FGQi>fr8NAHJW?%i~mt*
z9-LHK=b%L<ga*_`8m9F>wtN4=0FJ7h@^aX-5&z3gaP%j0__+w}1!|o?<YC`K#Hoq$
z9VMO4wCF6@S8f*dGEC<}dnUp$QbA<TGVY}=A%73McR3)aOfSW6(jo|ED#WZ6(iE!d
zf(Z=|+>zChZy)(w^&mIIYLz`)rG@PeQjmmyc(B-COlFN#$1Vx^ii%*Y2ld~5nY9p<
zCv@PYl3y(^0gY7eOquP?ztBt2+s3X^-nLzTDP@s64%N~AW*^Tu$;>qxsjk-+)A%IG
zT7E)vEa>R6bxXC6YKetyUc+8ls|RRAb#c$HSQ5B8^UEFzT{#X68{!-=@mn;(NAuS8
z^UT-(qi}NC;FmFzT3{JzGTILLdNmD^5Sw%m*E-FvR_64xV;7TZ$?1t8(Pb2Ds_*nJ
z;VYmT<Gl&W4SK4j)E#1}8-J^3udi>|==!y`9wu~gzLR;x*zTv+Hh4`dcgycauPfp)
z-?@{l7oX{<yJLT_a3y~4)z9<Rx>#{*`}vsrTIg4)9}E%WH87Lp{v@x8SC_1ELGVvI
zF&Go`fD@NSHEol4E1Q>pbA*jkXVAOCqr?K$5|NDwvV-AMwRQbY>nBO68|@^^Ao}+E
z)oaj`pfyeIMWl;nk+<k~tOWHtpcB|osg~>%IyK@VDU$H`NnUy36?xyFt&xnL{UX6+
zj>54IV%mypQ(`VFa5?3AWClM<Tga<q1Xo>SFSzJoO`74o#aK)q76e|Fm;H~Tf#&@w
zd-H?ns7k@S<9cG%(&*FgZncL@S$;=d{D#*Q#Z4Z@^n}-d*R8d-w>nim8t6PUDKB#$
z6~5b4!}_T2_6!zr<A<269cF<2k`(MSS@Xvm!(nSR`J7@rA2uN;@nntiAwCVOAk<0X
zTYnW5^|C6JC2vBP>#Px-Gh%mkHM=stAx?8)mWnH8W?T*aa8dh!``uTDqbRL7oT-2n
z7O3f#ZfD}VIrFMLPRCP@4Yz!d4VgF7iMYa?2lbnli?FtUNbQGl!JbWHP4=GF3`FGS
z6!6)}MsJ?i^I_up<8R_fTdG?K%!o^lJ!xpz?v~VhzbnAK%6lTH*cum?J@HA!Rg_#e
z$pxE5^P#>YzUja?<xth(c<kSMCmsfpc=NplpZghDWErga4lF}O`W?5xzgal=mQSZ3
zYGlVabKcQ%%m#LpU8w;iR6iBclPmKZlnS%O#E-HWKlpGpz8bfvCtL&e4uX}2W3-a3
z(pbHlt_L92&boqYYMQStKhqnXWF9=zsKGM2-)g+FRhxGw7po$?)@qU6xYl>;>&0oV
zU5r2liP4s<y?rriGO=nwM3J0b8;ICHX&khs;Hj~nYpmU)aac*LLGwpQ%Es?~P;B0l
z-<t?RmWZ#po#>~M@yL478xn#fl*prr(iy}k=%qdldkyIi1RNt|haEx%=nq{jJpU>K
zC($eK4wM&dp9m#HGzwuMb;uH_OzfZ0#h1z%xBraR?O#FKy+8hgT0}q#?aHOgB*iia
ztg1oJOKgqm*@oz=w)*9d8%unKYZeDo1w%DGCB*?BtZ_E&sF^zSa_79Hch9h^ic)7a
zqf*ZmrZ_DuND=eH7vZ^MFxv<-=%>)3wIZPiCSl^0SkuB#*DX<ZnNvbwAiiV@G(b6r
zss!bPiZm()TQw0li(L=ITAK^*K8k%A7h?Y2cH+xU%>^=2Bl10(lHVO2KT__>zgKv2
z!b9*}GmPK$+DD0=_b2LK`B^%Q#3$YHxZK3#p4VnyPip??UBdo?Okjim*(MjJp9`yL
z=3lub=89S6TIx4Vt$C076dn)+dnRopMdzdBZ0HcmwS){(TSOu<#wZnJ%o|<v`Z-o$
z(VFH}ch7o)9#g9lC2P_sqk_Qb$}j!_;q&C$e&I1CCQm(W;(rv~OzF>^Db3pkj$}!?
zFCTPl5)uG!$l67Ve1$G7s;oD%A8Fnm)JjZa&D<wKY|I0lrItQkLGJ;7;iWAd(=aS&
zgfkNkCJEMNw0O$Ndx$pgs~&nbrMJU`iz`Bl2SD2^xbah}2p9C%csw6JYQbX~XWa0%
zmM?x1B!m`?4xd0Cld6KI5krX&o#@pvFOw2|M<pvgn?(i1BYWyPrruV@t6B(b^*<Ns
zh1#7$pny#L?*9zpjVR4$*?Z({dg$7%EidyrOD*v09W$0?j@1LKi@ayGih-ECBJX|H
zYaJ%^mYFpi33K{tnrr(%SZOR5vb#DFo#R5?b<TVurGObCPr~yS3n;eO-!V7<dN9V7
zhxwXkZ}|yR8^Bqp_cpJWWI^=3eNA5ECHd!@d#~qkPau=7wB!rLRW%RwRn8R+rl~tn
z+w<~m6)o3E6DD^UCMrT<e@(pqox^i)>Fs6YfcklBJ<e?Re0lxYDEfiz+=uTA|Bw1W
z?LMB&B>1S%8yC2sBir!b`FReg0|~D7kbf{C+yhWUrMiWSdb6aIwo5;S6lmfvwozDT
zS6Ci;zZ&eO@(HQ?opO4`eKU@K#{zf^!DJ;%2%|6`cglgJH>7D`fkD!ZG5SGm4=ny`
zX}3cBs)jQLQ#6vQkka6h3A<=`4Ge~6OB0h~n@;jLwsdgO3TAWyZk3BgSBuJRND-Yq
z2s%#DCle*nGms6}KI}JvupktdH|cICVxso5&yURIQdm5L-e<WFqu%P$O(UH=2m1=~
zCx?QFWDeMgn|8eO!Q%x|1~#cbNPaf0U@TA>HlJWZE4eZh#4J_eTid_J5!cW7Ys1<F
z%~;bOyAjr?X8p0v=D4xnU#Yacv@-ej8m{VhMkmSALDG6I7ax;6-uFzfuFI~o5eB3C
zx!U}AkfYQ|yzr5T#E0y^nQ+_R6L2*qRk};ufM$y$bvoykg4l0)u^{kGAA7RBFpCBk
z{sP{--QR+*eAda(oZ((qP@ZV`BFi_kxxmGIjoJ2&@Mnqxbkx{@1rP9G{u)_rumA1Q
zNDbax+qK~}uVOq51|8Y*5LU=EDgE4yZ46w>0U+k~|0w9Oe_~G`AIk7bDZ<Nzn67Ss
z#bbvd8~cPu6`K=4f8_xD+kRmIS`e-rzZhS$*x7IC_|>hc-M{K)B>7NP$V-v&nyY>6
ztEB6<*Ba5JV5Z~08xCJN!YS9=1OFkzPgLr8aN~Y@R&d$e_-k}dMI~wZ#3Hg>`xs|a
zL?v??I6d0K+N@ZpI)|b9)*HP%TV%_2R$CIBEhcv5ulsWax;O}Nt`6_?+FRTf_}0NE
zvV>xSa;LVpVzZd0HDhq}*O}yB2W?>k0VCp<2MOeNMK30GuK~JFz4>X7|7lP=w(G=Z
zX(4EI!v=huAoIGZ*8Z-5bF@k77pSjV-V4i~POdpS6TE|^?J+#dCK>1bm6psQnPkm>
zq<9)~P0bKMj7HW2JY<J=iU71DerhR~)VW_m9eyrBvBtKUo1k2v&uaXn;D@B78sVvI
zHVvQ|FsvpNiLiCeGZVitOjsn%2Cepp4kF&iR|wN^y`$V&crS_IigpL!tey$DCoVcP
zr9Nbi@yu};)$@PwX@%?a;cOnNBbgQo<GnsLjCUe?{6Ld_j%^9%-R$%Apiy0@>>`g{
z5SFLkNbBopmKClW&iSNwPnzf8<?M7+@UX*|T^Bv=J*y5h75Rmf>TRB1d)AQz%OzOf
z-oqNOsA16~tR4f|pj8~2oKqy>R<Cddwft7E+y`dKOq%7kxVTez9)mv}f8!{f(iRee
z_2-Q^?_!|T5V!;gr7K6;c9xk*C;~@~Eo%5&U&2LGW_pn{jYAY>6*ImV=C@WA2ef1r
zHvH{k%qb>Cfs*6ORqy}gO~?J9om_u8q>(YoO0d<b*(;p612G*;nS3T&TG=Y6G)B|P
z3g^)5H8Mq4W)DA6H`v{VqbonZ>qa5GQ%uLNL_>?{@MU72OOAQB|FWp(qSarF=Huy&
zAK`g$gAZLC6bmLH3w5UipEO^DGVwj87xytfYG3kJKQwvTTlrtdx1Ptbh{&Jc02Bu2
zCELO8p(`H0g9Lj}kzbcOHA^IStRq4t_~W4{9sk&filwo%GeU+`PJJVl$(nO_39mW!
zVZkPxN^AaN?Ngb?<~!Qph^@m9mlsOhud!ZH_kur;MQRe_12SSttN1TP1~5V^iB0gR
zuM<j3rLBHD`wz>4y5&|(c?vO|f+nIgCkK*m!5yyqC;FxDwcOEKhl7Y#i-_Q)w>L9u
z3$LOq-BytavVxeqSc^EGBvuYFV;t*_p=%MV%7g%R&!>Q-VNAQ{hk|DClG^E7>PPvY
zoz_8uU5`{n_e<u#iIT!2%b@puSsFdlZhqR&r8?z>1lZgV<hL6E5K7|?P-Dro``*b&
zv-Z%mG&P_donJ-5%L||r_~L~(>euu@AoY!QHA89vNGOMbp^>kv<x>etABEhRy?wMi
ztg*Sz9rnwm7y0g5`umf(M(8(W^*Usl?I~Z38rD3#5QZwYLExBs@a@-VyIyp*8~Xc1
z*Q^@aB&+=5ll-ADEi4fF*IrE8T439DB=ri+<WYfHgo5!0i4b3q96$!8Z`aQ9m7tUh
z>`q?S*Lm39QqtLlmrV1P+Dy&X`4f|Zds$Uu4s0I_heXA9&qi&zf{!Gs^eMHWlY04|
z8Hu4nzw;q@PIO0MsYPd9sMLEyb9-(IY>-pr>WmA=z%|fIH3<AeRmcHK2eugHy)jBI
zT~d>=Q(S9^_8KWnTsk9MN#Mz9)rY-&L~efq^9?XE+01C=H~#vJ^+mgdWH2Gz8zowT
zu0}zT)aGLq>6A#~htA|Y<)@fQ*{3a`zm&<?+qO|oLDNq+%+<X3%=^P{X-8isanE0Q
z@L|Tt<&d%M>Xp;cg77{XPcahsEd!9G1<!bFnHdJ}Z4uksUTe<WnAYsW8w|MzG1}hm
zE1`#|x&4p~wiGgoSKWdozUmn{^H6WlpBj=a=1>S}u;lzR`Zr}h=LnNx<mL`H((^dX
zO1T5M$yYXH$!_?fOu=>}YwmWkR)otGXG#97W~e{N%Zjq#w?{`XY(+KTP8;#$MD1b4
za`_2BsJ?!y*EurJ<N9P~Rb*`xV~(FXbwLovneb&VVz5)^c~(E1oAu^bk-TRb$|M6D
z#lNMisnRT2<`>wD$UwfLW8WF;FyeLxL-I~H-$u(5RSh@9Vj}#N<d(d6@5X-=GsqC;
z|0pWj+!Xh$*nXvLFg74h#1FGBVsK?2b~1E{s$=_%f6kukA$i1;56^Xi!vZ14jm3VB
z^R(d#A_Xs8Z48sze+zWS^6Q$OIexG@>WsVAI=A%^Nx*H?u_619sIvEJBD*!zULU7f
z#r^i%7?BcI>r;-%jaQw5Cg-Bvyd{(a;Gic>gfu=Hj?!Jtn}GyXAD)QPuHUG>95mIO
zk+UyHvFS!hn$X}&0||SS=KVmQ;s+-d6Gh%iN!D(u<<A&Md$yplW9UtB`|fja>Zg}d
zI=<F7K1$sE(4Fn*L52F2&rrN!@)pMk+rlqZW<z770DcOVA=GK3rF+BTS4*bH`<@?z
z=?Crb)huP}EYgOVFAIBv4`0k*tfFGXKe{zpp0-n5(T~NSwiQL?wEv2g>qnx>$S;aX
zsP^;+Cj5a_94_S!1BxTrSdv*_qX-?N{pxBRz$Lx(u&}=W=4w{xGqftJkK;|QlZ40j
zo`LpFe$&tT=jMt#x&$sVh`IJM_cW%>GLHW#RQC;<{gzwzxL&;{w?TKl&Arr2w59Cs
zxXXE?>hoy2BW9LYg*PtqBo~4uK3?ST$5AVKSyld7j93<h<*Gqe)UZzQtoo575gk10
z)TM4pn^~KXb%XjMtD@Gve3X0O;SE=QeWh)V(k(rkhhsOcSs%0>2ak1oLQ=w*SAj3d
zI(ICoKeC4??O-y|Jo+O<W*?g>WO(wj&DS<Ja*0!C{LvV+>~mRu<QYR?*MAg(8dHYm
zJc-*i)<L45>Xq2u3ym})G_dC{3-#!oy`twIHt7DAX&9{P5d9}-Vv!AZa|_~4@um7>
zuV4}+4=mIlc~)a*Pu>&1JFZ%b%xh%esJIr}gGA@SHq}l`?_y{FM4pVdSLe<hw0(5N
zK4ghW1OM&gAN5zxfIG>3{=u5=F;SZ;?&encO)<bIXN*A<brR|BvAbWqT#kxL{#@c@
zzEk;1@Ysl;kv|rESEp#|3F>9^CdE$GID2KPkU@pJ+}uye3{_RNZll=G5^GSOA%i@w
ze+Zv4ti^Gws_<rUU&@qR_?9*G9Gt^tGruNntIU$OaT6S%yF=~tPhZK_t)X^)8^e6d
zswrphb8A@-cADp4nOM>nyJoQH2toy7cP}OeY>-(29()H^(fmn^T3!=H<+jSgPVb;l
zOD{ENqC>dPUj1=NZ#x^)JIdK3ABp<tASX7lQ==5uLh2gB&QE8}Sgy~FbW%szwEm79
zg%`t}HkwHmo^uiUQ%5M+?`Dt8H?CFvW7yr19Ub9$z5+vNd=sQ!qg|b~^&uITQ>5M1
z^ZXYmmD?;#r%^EPPXq?>A4TZ70lzd$Ue_crpcz);*<tWaE(aedwk$Qt4tX<IYBvM*
z2aj##D2m;rlsi74znVb|50jTGVBL43QohemnJ7LepTx6TjJ+?7(;1s<suK9n>19@v
z{JOi=wD@wRG|h4u{KxA152uqK_!aoiVK<HxmQ489Guf~g%ohh%BA;E)D{@XFL0$2<
zxzwoDxfi#^qSMEwX&Y3eC(lElUBB3xJ00;0zXfFu%Xcmn`Sl~g)Gz)Szkm$CVakZw
zgf;i*#0JsCrg5b4k@_f6X1olL)dD@ggOsxU<J&(UHaZN5E{0JoHJoBL^Q&qV9`nU!
z*VP(~N}PR_5~i`A5?6j{DXeA=9+W*m1;2K!Em?||)+`+sBdWG4^xznc1fd0yR*g=S
zSRO?|UBHqvh;W>wp)mKnRbT=S!|`S`=b#cq53=7OwZhkpA10+WYI>3cDqcLMzP}iG
zO1Lf^Ks+;dU5_j!%Gc}jUC(vc(u<fnTEYRd_0A(FXkVFuHeEIGto|(~D7m!L^;UL?
z0urc5sQP|a6xjI<TmE~2>hozL_`W**p8a{E>Mjob)I3hJkr8h_miA8`w6Sr4+LxMy
zy?bLYfhmS;7zISDL<^~-*wgSv(wr{<%(*vo3WmTZC`F3uI8hxGl0q-t@}iCAjer6X
zq+Aa;OJJ|ROo}qzDFN;wnqg0a7|r4qCu<rd?&}@Efe1immN}xVH=<DR6N<Ijpu)Cy
zHLF0}jbguF_m_W&$htw&17jYK)n{FM2Hm5wC3LgC36#b8^owIZ@42OA%@xj*v#N6(
zCe^8ayvr{dqDix9Odng{t-(P*yE#Wz>QFL7JEdt=)5vdLY&wC$$`s$dzEQlT#|pBH
z*RlD5d+zP==~oV=;qFXkk)(=7LV~L8I5L@rEfmQnhzB^J#iLwaMvqR|C?D%Ptrghj
z>t=+|`D~F!Dfykf=PD8DoRwZ8GHxTN<W`4o=JO$+*`}Ym3^2-gY1AyXQvh)Z@fR^w
zwnu)Kf)sjBjj-R7`d^y3jNP9ZqWOp>P75xT0AWqd_Fz|k8v`KOkk`4GJZNc0bS=Lp
ztx`Es{lrv_Yq=Cy>#q}`sj&z`{d@7pl~+;yy>APW8fA)o1OH<Bua-ExmOm=uGepx^
zXGtnavQf&3opRQ=7Oftj3;<0Df_;!Yhon}$*=TTW0<1}X`E61Or(wqbRpu^X(qCFw
zv>(`h7;;Dl&Ay6$`)y%)6m&)Oa)$dvX>7HEk3n}X2y9@Al?d_w+7VK3Ho_uJySHT?
zJzeYL8bz#p1Tpx=1QuZE#!ocitTk)ABTzAW#y?m~xPa<;CyZN#>vO?#6|{Tf2{ynA
zGc96c-WD!d<W?tn;s=xpyVfMCReo?hT7}|r{e)TsjH`~@#9sjg_B_<%@Aq7lETLWf
zDI~<0A!QBy>DcAl00Pvds>5C#N?w<ppJ;|Q&tzrsN{|A6Ni}heiSR`kRNlXPJJgag
z&ip)|aNE|Ms?^W1y{_z&_m%pQGP}xJql6t%P~Thu9$fT3__XGO!b7rLc)QmV5wsdh
znEYptHoJv+azn`B=tZ7HRs?|Z@#dE6S$jm-x)1jgBDC<wM`!whPMkc>3<EXlBJKf}
zN5-z*%dLmCj1~ZJH?oXp%cL>+auF}4-HyLZ!Z?ja)d_pcjtgPzE!B6`&oAbf^wy0R
zdW(2c(Lc0m7I}k9>aSSx+KgaRv2>osjC+vkR)|Ir`lk9fE#86`eNaqO5aYAq|HYdP
z%vB{{XrCl`n$b2I`AnS~Qo&98qA9kl6KA=Mxa$(+Oc-+~Qu&|9!pExQjcAIe>b;(^
zo2!zlys_Zvsa5=)fFn$oe_Ccx$>oElr<2ZCYa*5ymLI-0`zb4dg!76eh>+Flu{y9E
zySX}D4MqAN3?@rk=BI&S0ouHcrbpq<wh9haRd@WU-CdpXZURYAlFWPw>6p+A3f%r3
zo$ZJ<AHjB}e`JU=Fz_2<mSaq(BR&QqAh#qKdA+QrTkDr%=lBvEcpEn?gGSj00cyqc
z_Nx}xB-Q623?yZ-d|z=v!#^KWl&!g|cS~*Q#qw>wLO4;PW^ez_&2yPDnVnZ)ezCBI
z$bX4-GV`_VD=_!6SQ_c}QR&3pm^?COg|XS4noE~16VPpcjhx&&HBqRR<EyoK9{nUr
z{C-X8#0u_uV7=}^_{_MNDj)xy({Zgia7g8|+i~Adq3>3J6;j#;6A7)Q)6@M;6^6FA
z^_}-N8$PT`UrNFl$t$wY=B<yO6&1i*{rVAiqy_&<#^l#^<!2HAQBBQQs;4_VSGbc|
zVEs^T>HS58RLEaNaL?skCnt}lIabgG8WpzR>)5EOE519<hqVrj#du|A9$>OOR27*X
z={QNU4r}Iv)&Mfn;ZeWqGcF5-BOb!&%D=1$F@{p&%3Q{r&drOvmiO(yyZk<dx*5}1
zR&f4MC3i(1tXt~h0(%lfQl#eZo?d{hcjqj>gjLTEBgP8@c`p*qy>=#cnBod?)#?3b
zB+Zo*kGqF6%iACzR33Jex?+$g7)MeY7+KR}-5W0ot{SY;aK}3@lbhl%P|9yNj1HY#
zICl=-oj=cBQ|r-raXYYCa8^~D-w{8+T)t`}O|s;trQR=v@2RELR@wGUh#n}$A1;rS
z+hSc*pNl+7KEzZT0!@NaCs&`*NyD}fUx`JR-pDX1c@IO?X<;~@p(bm*g}2}il={3y
zxA+%{oS#qScDrA&yABXYdY9aye=&7EFR$<R03*vtDTjy2U0lmgQt`a2hhEoy0fYfx
z(D&<uM}4Ns1Fgllm#SoO^{c7`31@-G`T(Qq<g*5S_Szh{C%eus1{052E6d5cr*}E`
zx{Tb6tk@F+Y8*^<(b2VL@%vjJ&o$VRyHa~IaYP+~jKHpetpY#ZkGK{$Txu8q4QtO{
z8@2`vMN}dsmyl@=(w&N!R8*UGq?UjpDo8OdHNo$I?L?wkkidTvL|y!*-=y+&S;7>;
z2OGWNRs6(N=#Tmz8jnw<AlI0ta;~@}U&n=Ys|BegE%)ATrGw?u1sA8{5;|kw;%_5L
zImfnhFKykfGoDovo8q?zaNHLO&fi2k;B(Cs@DCfWqulv8y2=8NP3FA!68P@U{(Q-~
zTt?rqvQlI?&{-{Ifi+6NVm{<C$zIKgSMlrHPuMiHdzi8&*!`=R8>S_oUVtFTnjXra
ze+bjqlJsJ~GbCt{8X1jT^xUxewRMxzl9}TSWSlg0ew3lm6d?&eXk^*n*z&*RZ9pB;
z1V0eL%3EL&`}9J^c4F%NGS>O|(*yL4gO8GrD;fV7$JY_QMB=CfFG3N<F19r|_*GTQ
zQ47Y2Vg?*;a@8;>nm~kDkNJniU-uQfNP<CX=neg&boSI5MOU=$ypfAceP&*E7>972
zP+a|H<H%j2G$i;fr3L}(GXHSv)u)obK>t7~$+n-mR7KHL%-xqEGNp&cP6yTLiidV%
z0?JrjR`vt$^$Z~A4b}gni2RA;z41i3(=r)5x2QK)ddqqz=igi{$|M?*@%`sWNrf_i
zu#aLp4;b#~TYfZ=mT3K9IYCu<LJ}sD29OYAF_Son`;$_H{#PSnV;kX29*~8(Vj5Bh
zc^}n!9AaJ?ZTDJ@*F4F>g+lHA=FC27S9CJ6oR1FsK#Q_g@=nqNwMWb9G)1wnr!6Jh
zmwQFL<`m_wg&65x{np*XA}I@=1NCbQ&e1c#Iw{zGP!9_(k9T*vwK?0Y`Hh>u$x;D8
zsG-9ws7h@HYu#}fM`@1=DDF!k2r0%`l{?-Zw-);;lpZ|6uNtWq$!?wM<2Q8UMvs9R
zVliMp6~AJ6^LVJi*jL_U@e=sG#eHUb6M3#qeJ#P(Xq{|IzkHIGPBv9q{bIjdtkhg6
zpNjA=rUQhek|}Q+3?m3tIYbv1Ego(?yZJk&{`L~?{lUe|gdFgiTQ&Q}oF*UR;<R4<
zs_<)kiT7EWSZf#woLXx&;$0eU7WA{N_Lt;<PO^2w=(VJ@rDBw-OIb~$YUjItjTajd
z$A@R5_ufn+Jxk0faC94J(aSxok>QI4LYlY3R%HIwe;)s0#nfnX&{8NL=Z%Za15Nqx
z2YBk-izz}c2dnxH{^}75x@KqoN!y=3F*6s_eTE5>*)SqIJ<>^?6Xz$}Xz)2loI0Qt
z$PT_tN!d3vYiTB<4*)9kYy<un<Q(C1630U@pJ%{x2MO9;RUj_xFRvQM^~kQTl5y0x
zPcLyVX09fBUgBula=UcYsxhER{iF8hDk-*t;_>EEBOLq>E>{3w)CKvV0M0RgAZgdN
zDp7kS9zByvyM2V7DBSp3zmKWEdb?-ma)eljeYE-3Ihu1qo}=>C;z2P-jgO?Q`@Z?x
z&U&1bxZIpXf>K1u#AH~Jn*pHY72N0$EQnPa>RmUJ{Nwv%vU5r87xlWy2GBZHHjlV{
zSW1De{&u-#Z~i4x+T(lhH6TkY{wBi11jq0dESn-WmvY0UctVjNDOOqIF@@#h)x?Us
zSkl$skIeDdO@}&HZ94S*&W;eX<5-sD&r<T3z&dSn@LP*VzLe9uwmPSU6T%mlX(f<e
z<E${citzO9ps#0tK*MX=hK%7@39X<=qKs3k_rvmY6+VeiUf;mBx;DbMJUZkYYBvnu
zNNmI&41;oj()!8kIm~K#a$>r%BWyAbk=Y;m*?rxLWn%eGjSvj=AVKmDUH#h#nVSr-
z^yY>?h(uzU*<v1qV!`O{W`iIZmK#T`IEQM%D4Ate&4Zc#qlnArMlrKlY>jEfhSlAz
zlFh^{3iF*c7r^~q@EvB#5jehsXwMqqCN2N+3@O3~VQ`OL+o^NedoChQrQ&VSnT8qm
z`*n?UZ>uLMc{N+clLEV|8J6k>B%3~GIX!mSEag*7lmBFYJWD^h$qmnm1@KV|3|BcY
z%^a$ngl(lQ^WPTSqEo-NG5zd+W#{TI3=IPEOJ~0+%gwxJzv!Z{m7KJ4;AZ$KOW}|!
z&!wg6m8e!BWi&QgmD1CsrwMHvI&X8xa3F=w#XG9xcQ1iI&99psdD1jaF-y9syj-Zh
z%?ik`q4pMiUM54!<$qS>!qwROfYYWDY%x$n9J4^(x({$y^M|%l9Xhm_UqTL+&4ig{
z=c9cNv0Qa*MoY5mT!+=Qj5YLwHn79E2x}&y(nCuX+XAObn}#g^VHw^SFEZ0HpYI&l
zp6D%docJ!3YW#;wjc+MI5e^iYj^=qIssh-zMLz#KvTpqmWs(M~YL&T*Kz3M9#`h%D
zj&rh|Bw~+P&-pLH1Hm%^ABhdVcN-hL`w)idMMG!fqeQ1=(@xt$^Ot$=wxgeT2xddh
zX+bpYxa34zrP?_zTvb)vQANM}#{76DO(5l&SupvQlAy)(_2a8H|41|%vGa!7PkzW;
zvMn##1UC_ujq>_V3NIaV<a&?>kzD^#(A>SqJLjHbW|%ye+MX=MmY7*%7fI(XDYN#p
zi$<9iOK&=zO_mi$>NN)7_*9oord1{9U5$n1@pa3ApF3!K8W(7flof*jS@+-a5<PFD
zfe))U+C-;52Nfi@V`=DUr(tWP2P|cbY~2a`-l=cFE+qm|k<vzZX4A0IvCVuY4zCyI
z*vMy(e^;}A7hZGQS-x7JQW(nqtZHRdD$X2o0xP9Psh`f}b(Qd){mmBgm{gsb8Kp|L
zG|u^I>6>FJbj3@IAxqP<dKj7buK17}pIy;jtf6$NVpW~!;4zp3%3wG18o`*CujJ8h
zJ&&I}_BlKH^|VWJ-I}-l`juB69lP=l4bmEQEb7=Hx&vP3HL7>Ut{t~PqdKWQkIGt$
zHN3B1Rkc_2GMD;~;{D?Z)A4&bYxH7YT6#HGofp`5A6oDMQ<C1wTWqmNva0DAXo5CJ
za=<S&P9(HX5e}no&#=|^RI`L|dy+VB{CVINrUEH_?%N`l+<yHCu%YrKtu@h(PwS1}
zeeGExGZ3`i-XOvdPuiGxG44h)R(<r%S?1cWxZf4YsCJ>vpx)WZ!D`^3iwN~U!d?e|
zKDI?kF2!9vV%!zcgZ;}-XpjYq{|Rq!?OJ$^gKC6s>EHpISs!u+XZP-7ZI&;hz3zMF
zfvXce#KWz=IN!X+NFYzlx1Y}<37h}Z^pq|Z?!mbbsdMeO%kW{2E8V{2+cUlbTmO9D
zlHko>Zsu#xeRln@@<DI(c93(7v(^zA7ueYEWA&fnUc^EFqkv&vVq7i-_wtFcQww7|
z>#qO^4O%;K#Flt<$|2+8Cw|~`ulV5eWJX7UmAR^=2gK9rlzSf5Qa=_lF;B$j01gq5
zdR=()3t+$V`OQWCG&#rnKB?b}9x*s7iIA;<aHK`~*Eb4U)2^C%P?5)u|Jt5)vN3|w
z;w`WKgZV8`dcrBzhRakj>(e){7+(3{vHsk$=Fi#^fiLP9eamp?X`D>pTRgXIN7S{!
z6pGVvYE0lOA^el<n}3iUka<0?+)N$_vo4|G-M8v1VfaS1&h}}Tl8*nU<*%h8lXqu$
zYIFMsp$p@2Cbt*MH-~%YGM0RNIS<;$$;wzq?bhQjAxAL=TG8HJ#gFQE3L91jj7XXJ
z0;-%Uyx%`N!$bXxJT!;Vyl&>&0ppWdnX$3h=Nzan1fRbj|Hn;7QtiX2_PP#HUXXS-
zVI@jQ-TB3JV{LWu==o;z;Mp2UTKV#Ly(xuen}YkF#WuGf3nz#jJtwK?F<KFTETT<A
zGEgVmXa=c_WACy{DCHK#sIMBR=3><1Y%XyWI$WF~M?od=W2q=A-X~+MH3?~7yAIJ$
zT1+06ZwX7g%Nc*LDik56aT&qh&Cg3vK7sixVO5KIi}{Qe2`QJGk12<7>ditAho*fa
zrS-2Q-g%E=)wT!+=oc^7<FL@P_4sF0XTwNA1!*jWqN(HqK4#s=Ms=a=X-jkpEe61d
z$or&*_1XbC{?rFG)FI`v0X0&J2z_Ev;v7J{Os_xu_H=$119mPYUvGa;WTm|QYk_dz
z&T#zU9{+?Mbs?DD616N@5F(q@w3cY5mt-m0|BwaTeRn+^smk>Opv{PcXu8}Flav%Z
zCke`=h}#xu++w5j6vFZj2)_G%0Elg`jrT=6Hky52w8}t*4pqja>fZWb;*v*QZ1cP1
zSwANpx>Q*Ct&Ch~=hQ~`nV%i{PfxEvu?g~WotSW)A?kx}d@zKA160hyKJlkrO5^k2
zo|BNLmxA9)53excuB26mKGh4<zY1C2QbdbE!R9|SAz%0k9Q$gcE>K<gB)c~TA?$iQ
z9r^Jv@!h?nG;(%sn;^H}8{`l!zG=&iSn1+Gjhd?xN&&lV8EeO!@upJ6QeJMrVRXTr
zEP_R#cn7k>P&1f$cFUqJlfCm|GjrOU3><dSmoS}K;37%(*;PY-viUs6DTJYj-`;Bj
z5RqEHhGKfmOiziHd@kGFcg8-Hak0y^+k9dYlR-Vbx#SJ<xz6Pv)BjZb;DM689kHq4
zNIVw)Sk*j{vdOG)XVpl$cM#z=Yq&K;=SkG*bxn*UHt;#j{=&$5_Q!W&%nI8B1Ahg3
z5BUX8^-E|{JZgM9*)gKt3}&77_UH{#452%+0fbT%3ITK1RF*PkG7!5r>OYpRRyw=o
zOq$iI%2&P{O;4KpIO066TRGLc#9a@}*iQ)-=x~YDOJ2l&97`bA&Sf|}t@(;y<dxl{
zWI5vnfur9hscLYsZx2+zHu4(^lsqbrDtb^y-*Kjx!WU=0NQlx&UaG;Z=f}LV2vybx
zr~fp`bcv*w&gLlX((rxibSP5C{LOK@WSq_?Opq|}*`Gnn^Dyz+=be8tK;eSvT$XSS
zq4_=mON*=09C|pUYe&J|z|=rQ8CdE18>X$id3k>K0E8>ZnM~r%@VVr^$WpqyhWwrk
zpSeJYh#cuI>-@MzE?Qg#I;{Uu*eORIC2E(<#F(*34UKf%2rev|Xz}K3g(i2+1@c{2
zReenE<aWxu`{uz^+H>aslab4t7h#tVpRjLv*u&rLhG5K!v{g<-vD`+N*L!z(PI^_m
zas#IW`tTm-3AosY2{EOP2ZBe9agpIHd{=3&R-FrDB>DA~XCV7c<p30S`nb`FzoOMG
zALjHIy9J;!5(VRkwbW;K1Fg2PyZs?EF4`}Y7YoW`XF?XFV0>Q+$#co<*d`VU)5ya4
z_}C&^>U9gbTL6omGf=bE?O63(HmG<qK1)f0{?^4|o5;HL!8Yxo4RYz7WXSCaEjP{}
zk`zSH;sU619IS8JFgP6|UB3JH#6Q3_z(2;`0IAK$T!61TT<`y#S)`+Y(hn;!*;MAd
zyc;OAqxbv5$yq@!*K*}ZuP@l0%kK7I&rFhXhI{72sJ;rJj;Gg`iOsG`;J|xBO+tQ}
z%df_r=WD}Wy<tlW_PY~J8x35Q4$C^4Bgrg{+tOdXW0;PvuIu0vh^oeu%}z7(L$PW&
zA0y0K$DmgNme(GPjguR(-CcnE1J~8X+dQWz9vFZ}RJ}&yy-O>c#PdJYhvBhI{&&_y
z`sBl*M8&9hZOHE7e7aFyPazkUCeKErZUQfO>tdp9JnKT-Gd(g{ei6sv9KK4fT(W)N
z*5y-FV@_t9!i2Rr`ty4RZ*37bKVzQfPd^y{J8sk*y&1#Ldq1lP(NQ@<UJ~b%j(3}I
z=o%cU*P}kxGunY>CnG><asq8No(~~PMd#q5qR*p-i3|2F7H!Do=q)OkiP%7-ROh>o
z|9CfxSFMg(Clz1*LR@CeY4yha>yoH7T!~P)Q6$e?3a@J1wuxxEavX#c^7KwLP6^Dr
zTc34kf(z{3;hJkaiV~iZ-gHj<GHbLuKRq^XWvQZ|C=40vj3*@h-B7L)*$A3bjvF^L
z#eCM<o`@s;`8~aEaW?Y5O~Og5m0`9h<eR(bICY~#2>kKM=fxq2<G2?SzsK=-uV!w-
z(0YvYtrFcOjW=<N{jW)V8>0u<j<JMUkMtn8@ZKA%9+^@nbE4cFPV_M1&LzaG+0Pv|
z)_NzF6O{yDlA&%a<%>*=7IB5JvK<FZ2Vk6L)6k(aPj41mN=H|RasnjwEWBI4SjHlf
z0#PZ+QeN0GRzCWNmWD(wut{-9X)u~U!=osrqC+%OO@&l%DNFAdW<I9+&1n#>Uur_m
zSDlx*LHJaN+B_`z!KiJgSPW|4?J1>XJ6_-`5O)#04AH!r)E}dOi3yLliVc51Tsrhu
ziO^!+1*yyg=M1}<OHYX-M_4QQ56T#=yE&kw2p-(TDl$#GUj5E~*^?nRCil+W(seaF
zb?<U@kaEWR4wK>A>RS@4EOrbFCCa2#bCZ$lPxT|Z&d9*iTykD=q7zuPfbOQu>(b6!
z38W-Tke&-o7gOOJ2tE~JKOarh%GyP1TbV3y!!BZeS*3leCZ`nSs1c{B&Y!tf73*Kj
z;O~xCG`U|>gfF^<%Z-{%t8kW60z2E&IIUDytbW9m@%XQC;5tP))}=2oZguEx+YJjI
z)BOT_#tHnucdrqr`H>F}H;X~lYl?!H4<=ak;nYJ!o>#IZ^m_#pK~q~<)7b&ESR0)5
z!P{-`HA4`*Jac;0%$G9+`S-4|x-bBl#hJi?k~}J*AE-Pg^Nr;81@<jNZ&`;#6vjT(
zEyN**As+223jxsNpOM4Mu7YFUoSm^%Gc9t2>ucJ1E>>Zc=N&-e;kb;TwD;klPy@##
z1-#Yn#(3&@9&w)QcX&M%HJ$hzlR$DGc|F|s?Zh(kbFNbPou?O@@SDav_wOez0TPEN
zV@>?`rCB{Jb+J2qeMw1n13_R8qJf%@f<*qIqL=Z`h(*v}ic)dY<1ZGzTjH6&8&F5>
zM$;Pg{Er%q;f0AmWMO-bUtLxZW7`_8-gQ5sN<}Q*g(O#;&o}${@*c+6MP-zHqfpw9
z3A{zbJIZQ#jmAjom~k`QWHukJuA?6t7hzn{*YyLz=v&7X5JYx4*@Qo<!4v&QF)D`u
zki}kTX>)%IS1MkYX!Suue9x$QdbZF@&jy+F6L<7znIGP6T5XGG=?H>>tO!pOHt5HU
zsNX)km{uTw3!eNz-&mZV-z|1H$yXx0YRlW*?2_df^F)8*%@Ae@-gM&BX++U+JR-G=
zOB|k{s^_0EkwuaAyseS65|sQw;(D=44F~Hw!0pSpeCW#X&+Ns3QJ;-<{eE|l=Q}Fg
zK?{8-T;4;`zvMw$-$8NlZ4^L#71W%|j$1<O$5dUb<cz_W*tWmG)|8hRZ7zlKG(Wyf
z;*x>%OM46ijEPD2k4I9NnQ9Sp^&uXy%8e5lb$L@Hd4xVymv0bR;Bh|b8NH6}YCBhl
zxHce3)~C`h@_HK?sL4_J?3U1)-_X2#jM$yW1A^}@g_(izS^XQTdP?9W{{5PBFUugT
zyV53J^8l1rtZR$=OU4Wwi8{DolP=&&4nk0eL~uI6q{Jai7@Ck#m!?kb#29aMANhKz
z5Hbmbh1$eU>1uh&{YD1WrDZ^IKNNS<6{>`D5Sg8lJB3^`#drtHGAEb(@!c}EbNWR#
zFR{L?x~QL<Qo1mzN&OppnNe=S{mZWNIkBpaBKGCr?R`nZji|six-3`)rZeI;$l9up
z^i**L_X{oWU))}Y>`7aULyjhH&Q)&y>>t{|^n690mBb`YV6;9J(}SFilXd9Ak75P?
zU2(p|X&6hB(c8(BHL$r<G3kG{kIO~RP5z@Ovl#bC#WZCF05TrS%M8BCV(_oh_}a?%
z+~9YO*JKytd=yuv>OC7m8dZL$`0UE;=kGB@g`uI;7E8;=a2}QGh=^)AVP;U8d%Y`5
zy>Vk`rHj<kfQQM(v_kkEl<S^NorlT$qmZZUGDZfshUlgLnJa-6&62##8TX`d#)9l}
z)igeaW0;(`M@1WjeJ{)xGJ{}TKW`&Gev}&W_y_xsq7`T!)pfRQ-JD!O@G~m*b(Xpp
z%UAOJPMh7Lhh-{^WA`(71}xT32D{>8J=XNz*c63SD*u~*vu)cd$}t}s<hkf;WhJ&7
z#D<`>wg%t)2g$45vi~E3(aQ{>JJ$*M8*-?#?buRp;i9>&RYN@2s9YWsdCLr(XI7Eb
z&v-ut{m5OvZJKu*((x4aR4MBcwTqRSQ8i`&@!-Z*5jim?@}A(1B6E|5)C0-lmX0B)
zT2UuPOzo>?61c0=03vy<Dn@IuVdUQywYOOg^g_q08u4o~qBZ?!AMN6I2hH%>BpX%#
z&M%G`BxInuJhRoR)%~%ChjP1bx5gw<GxRde>2Ug5FsEtLx*N5h=he$yCwJzHWGsD5
zi>^iFr&2Syk;A^;wIPQpC7$iZbR}as_B6JTy?z%y`jfYqmmY<t9&md%*IH7w3_;Z+
zI+}@@8=hF##Fxj}lcw)|c|SNwX~Ig!_EVH1lOC(ejh=6m%XAdW>*O@}oE_e|j55}}
z=o~(fd}wl_!Erhk-}RQxAaV{psXn?kqz0t0okeK<m9^=T6`@l57vTQ4WA>XJ*w*5%
zM$^&5z5!EXgy*!#ZD+)u)<rhuy%f7($g;t@;eQl<!$2SvY#*!RbP%4k<||**6anoO
zbR6fn^p?c6aeaVHW@+xR6CEJ~Yq8U9(gNu{<SC*@JIp)iW$L>@+sZoDXzF+yy8kFz
z)RY~Mb-x8aSoz6@H&K`Ln_n?b=l^p+OL1mGv}PaA)DsC<Ce{HmXgA%R!B5`Yw~Rgj
znHE*rnoQ|U%+nB_gVk9NX_#7gOiP_kOIE1Ea~L^XW_g?&jEbD)AHDJVrhcI1cFf=h
z1%J%3>uk+n0a?qzGWym;+tmwY0|<6=gj-RSf)5u&rPIiq_TB&e+iJd#wh>A=s6~3U
zN#d!oxTQo|lzSrz6l+$2l8XFS@VvTI@w3#G6Xk|Wh{HgC)GKbSzGar*p$P!J5&TOo
zE(&*hX4K{QQa&*wZ@;WK@aDEns}23W^mv1oFC!{<Nh__aq4VOC<QfW=iRtgz?Ao1<
zP_7)PoR`IJ;&AVPwu*)8XP)Jvb?cZN>bcHO!IUe#c`NMOu)4X<N|yPa&!jI3m$;5|
z^^3y79Z<uB+Q)1;Xw)T!BtWInx!fMOQ>M<X;Q&k!GA-6Ey>}E)!@NdW#~<eV;hmi4
zb%j`;qr~ntb^3j78jn%EC9tb=;v^2f#&Y7m;9crtR$cuRybRMd^i7l(cmV&_FD}v}
zj_kwZgYXo*N2v1=R~tm0V8p8BuJfwO1AJm7<jg60cta6Z@e6JE>bK+o@K4tK%gM%o
zxLE=f7j5L|neJk{Azr^1n1Cj9fod4W;x{LepMoI($3u0>kRndrLoX&0h<mW;JfKj@
zTB3{drkjj?Ol54pFTqSYSYFjJzMo|3yA=AMh|h>^vp_TZ0krjPQf*-fGl2ZizJBkh
z<U7qtz!=@6s#8z<4|15DW{hDW|BC2AF>B!&egk=ps|vVXpx~y5z>if_LB!_1syaYW
zr6#cM%TQJQbDSI^y6XNTbg_y427rtKG@CvM?xX(nakqjs+QnrV<@Q0)BGN9?G)^c?
zZagSQGw!O!Qr*2gv4q-{375Ie=#2|_Njcu)3y1^H+fK&2D82j0Cp6#=&l%n1?VfsY
z669eesA{PCX#`{EGLzL%RPfSC)+`k2%o(v!m|8QRrRg2JnE51ndE{FiPs<EfN0o&S
z?Q+~m(~oW5C>lb=Rr{_r#=xT4$#}Or_PXE^g<EruHI3IB7)kzfxm9fQo~vUXK1%7A
zqn6YSUfOcM%e*Wl3!9hC^Im5mB3#Vke_AMOZd)e!D1JesGy0HqA&8RxlB<`ZmbP$v
z6o+K%XrrDSI4KX6aO}t-Nw=mx-EJ8G7h@lI7pETr)3y5S;N##8PH>75!+HILF%0iI
znV+MUB_2&Z=@+p38bFX023c^gMc2%pzO5`339Fv_!#0YK+SL|Rt`@-pi$9gyz3S|h
zr41@<itY0G^5Xf(XL1mMGFz2y*n->n>#ylrBeg<CotcI+>dZsK0>wEUX^y#8$4M@K
zW{*JwFtUVBTJke+&p{avJE?px^Vc|6vETTBJoJUy(D)N{>1XGsBZ-P`hR4yIQl$IQ
z<IJUgyQ*iH{(~~*EX&N<T2Hl+9H)l%-UJnop33B}nHdUCY-?*g=IGO6V@{CH;;wHJ
z@yO2DH%6^qi{*Tqhmzd_Qa&9lv7OHkLo@sjH+12@nt%g$heh7hQl3S!^-1xQ$url*
z7FWdp!v|y&e&S<1DSR(T-gtO$u%-xM+t`0xO8wX}-4~sF6b!ufV9RgPiyuk8ac#T~
zlB#{A`lEFW$+?Z1k6g~H2M?oO)!B}XEIx}1pPYy|0f<XJqX%VGN)&H_fpn$ImudHP
zNJC`}=q-=xJ8ISOMa!&4eartSFkHHduy-MZOz}9)$T@(&Z`YRXSEylFk$hNe1sv%T
z^=o>!^!8{(W!In4IuA7%RiERh1VCE)SGo;RNgTxqMQ%y{=POhKh1Y5&6FtPm-TjKJ
zELo@SM56MKmcl-fv^4Q<=E%nv{@E{ctQuiGrG$P-+exU*WaJh3-2#0`?W#sy>QYSs
zg_N11AX<2-!|7i;iU+XdWM5a4m@xfgq|ow=sn`>y_tf}3MRRMUO3#pN0ibOSWi6;)
zaL6G+H;k{XQ1!ul<}fe*%A8G`e-a#HId7@wq!=F=_qr1+fb3;=7C4tDq|tC`F;M{v
zyOZmQ{EMhA%|)qay|D+>UWS#R;;$36#ZO=vV@8m$(ZeCCGLT^T8)USXu1%81Gj_I3
zK#u+A`D6r2^n!4E6-rdS6d6HMd0Ukeit`yC`cE3)OI&P)bdC2O9f4HskCl?_HiN@Z
zf4gHFE(G6Q?(jJ3UKF#uR<|1rL>Nmmj0ht{u!1$cH8iH);E}Y;a+xauJW0iD3uk{Y
z>tSlvV5qjs{E`wO>wV4-p?15g_U$jG#kOTj(#S|WEE%RA=}X~vN?3Pd2non4cor(s
zA2V}M5+CdgB)$W*D1SseKnHX8yDk7clK_p$n+!qoR%{HxJ*~BtZ>yjy&p(J&YIV(}
zV5>12G3V^cnqPmwUlT*cRR0HtM!(NV5*=^U?NmfvR|*6szRA>q%}=SWEfcXXuZxNl
z7rV^$+48m$aT^44rHGHAMT<AY{7_{alhKichB%f4HR_RyG#__ArBt@3oI_PG%+%Ej
z!SRSdUi9&vY0Yqzg+oJKYA8hCopMuG1~e8Zdu&k_XiU7RH;grlJIB?@{=I!*8=G``
zb6ac1`op&G%UJMZ;<9n-^V+6wfqkT3il6-b(hZC9eH8=QsUQ;>I`<ue2b3!v;<weF
zMf%TMOy96%n6i>ND;oBjF~Ny!@d|zv_FsMk@+6Lto*FOv?s%6j1LsHPjm=PaRQAWy
zyJeY+Noqbphztb<m!Vmgjpcb*I?q9llf({i(|LYS_O{~KtMKGn<np=!`CLSwzV)+p
z@8B_XomSrZk@R5_|B0PD@_kxmGB(OAO89!o!gi~v)W|3GRb&Cb3$HGKt<g76)y#*<
zw-I9B)Nr7VZUzu!mSKp*M@A<E*SEwv7E&M?XVlGW-f+!bV?*0Cb=2Kj5Ca+3>tXn7
z5Si60?_o4Kc@h&UdVu~paT+-yy2ND9ZV%vqAm4eV1egD+70@#g2Lv0FIyYF%CE2>~
zO9tKJD%XxT3H}#|W6B6m?}|6GUCK56O{{BK?heb3bUXANtl?1Z$$3UF_TbK&(qM>J
zy|v_~J8gcT$mDfsY}D_M%S5^?P1i4NS%>G^zcE;~Fo=qdWq6;Yo)w=}qm<|ZKU4Q>
zQn%A!IMRCjHRMyYXfkQVmt8r=k4vz)&#wM#zs&0bm**4$)BD2my9j3WTkDYa-_HbP
zh4C~$<=M@Gxykm}p1-l9)mmXpLm`wl`EQEP;@v8guHs%TjmV<8g~}WZTmJrG{8`-X
zd<X&tdB9Ti;$@qmif{@2bEjx*1^(nSnuoBKdHwk1>|g{!<Z$q-YfiO#sBc}}OoDP9
zxU)$-nD|3(Uwnm!fid6diod6236;Ovc+IkvIGPUJ*VCWbSsJLm?fC0L67zo?on=6i
z@7sn^2_+4Nq$nUAlG0$%9UC=38a5c+sr&#XrD2pb3`UO*k&e+D-6g_6y7Arr{r-G-
zK0VKUU*~Zimx%oe#<M$)jL}C>acRT5GOe98KmmLAT1#{v<#Wz>x6jwD7&_}mx_MIe
zjlIF0t$cXf-5+ie#y~G%KG+(WylJ3wG!v@iG;USJ%<tQIrSG?~+%htk(xc1yN5)J}
zOh~YSX}-{X=#xoiIP$0y?It{MWY^|-^=nJ0t;cm*RmboJrc-z<{A(p!_9@UBFTzx5
zGS4mQJFf<#vz~m2Rd_!PaD}J;X^oEzeJ?1eS(k0=D<k8Opi4Z6!^x9@tVV=010g77
z<Pn}EA*hcz+Dc?k5E5HU{q%#d5I8qmv|qgc<B@cyI{2-t0<ZbozmWunGgCu7N(}sa
zmO0Jv`}n|;TW0Fdj~5bPJ$wf!R;;*;2&_$5b~!b4j9Z+aDC?aMI@r}x6roE2)VDue
zE1r|#ewZFG@=ny5b>d3_cUQ|qdDo-G@A8P@KA3_@6n7I^7msZ{-FH^J<tZND&+?(`
zad*iDjVSV%Yc;Zkgk)mBs-IA>Yf1b?_by|b7nz4`JEi{-{QrfMChiT0=+VahHnzoy
z<}hRHj{dizkm7l!HnG#J$Td&x7EC8Z(iVaCVf&Bxm9<_RBvmcOoZ;G|M;m7~ej?Qk
zU!;6aQi^P0=yG-|FL2!fn<cdCR;DJ9{5?ZnvmQM)v<CXp@rlZ8|MEt=S|U_F@mQ?d
zIc_xtM?8hq(Mt5Ay_xOK(E+zUN5kj#EuFMTGJ%UJ3g@PCclbgA0p+kO(Asc)<K%R$
zDZV<LQ7Y)7M;QTdpWab#oca`X(G>_~IquOh&ZPrKTGfido%vkARK0l_+6rJgA{WWN
z*u10I*QQ;|fPM(`6Ur6BWPTO;BQd|^>2FpIzoA)}RtGv_q1_%|{_4X=1(r|cT@S#T
zrH7>N%Snd&F2a;hYF5hQPJ8~8h3n3It{txDqJw|TA5-0&GH3b9^hNT+Bq)2zBr5q~
zvX{5e6kyhg<a+qN_f=dRc)z<Xa5tmv(6c-2%TupAREP!Nc@jL`slMJX^gQ_LGzlT*
z`Tjo^UU*nf-|N8!)axcs$11CT1ouTQ)?OqF<(jK*w@9%h%s@0YD-<2>ERJ?!bDtCE
z9mg){xP8M~B4aQj-i&tt2$*pCUIFc=Y>;_Z4W9pyV`H(F9X{JTTE`T2`;Z)T=@SbW
zow>6+w%DjXb9<{C*C#PTAxW3o{-@0QNhMSNQJ;*YUO_Wy6J0sVsoD;n&g?C*(Eg>U
z%jTkBL5Knz>uMx8uo=^rh|?AATfAeu8n<Ak{YNk)6v%Zke-4pbbKZx0r(w=Weuf7k
zlnuGpMHb>zS)&EdPJd8Hagpwu6-;^a#r-WrT^cM!6@j*yMRf!@HC{3a0}Cwl)Qb@P
z%l@G~Dj(?TCwN2n9T>KXf+;!gYjP{k=&Kzw>P>{O8?1|q8XsN!BPhdPep8=0eSLBx
z!y`D;!GV0hfR=Dss*5_-AMcX~v&Jm|)26IM8C|9ZTMVcPR?&FT#K)vlDf`ETU#2uz
z`FJ=XF&Ul&Z$)3M*5WsRX4GjDMs`9Z#UjloLYOvlbkvJsBIqfC&9(5$RO(PTC&3Zg
zBw|xt|M5*9ad~%G$^GtNA-uu)=1&Nab1L7xA1Ot9cfiE;LzHSCXB0tzr5Mf;(~?u2
znM&)Toa=Iirwh`(26;)fAyjCmG|(y^b6!grihiimj}#kwFuA{#qtm<m_CsXQ9|HK2
zb1Iwu{I?ynK{)$z^ivsx{XEpd?|M?+ESfjv`WwO2ca@{Vb4WO@td5bi0pIXLuq%e)
zM@RUm^lQPQQ2mHhEGC0II#S;SUl20tXCt5lG;bpbX`cD}hEN7)J>@~TaDD$$u3{-8
ztjjO<-5gK?u#K+SVCqoKb!msVca{zIB}?aQ>+^@+YsIfG4;gk&uaYR@oY%6Es>u<!
zo*GPnXgG7l%KpI2T&l;S;4MJc5$OBzFU4rIj*yrmKzm<?q0j@XmUf<<JUUk%%o3M&
z4I0T#cl6DPfHXRnW(e~?`EQZ5;aZfu7#-)jYv7!u<nqo^8PRThf;0ZYB`j1g#4UZT
zf7FYZEke+M*O`q!f3rnz)=x;BBK>J$?CHAV*&wT5y#+ZHJOY|A*4&{fqE}YekMD-(
zG&5GyZj9C#3^!Y|gk;B>^v!dbAT}IKp~(2l)e`%w<h*!K3wKzwydE#FkQOZYk2&E#
z0voEwH!42Pr^*H69&*C$4$q$AxnyhR-N@V5huaZ1JeBIzCJr>mT@QG^G2J7K9E97i
z$FP0_Je5m+UIkcN7fet~$dHYD5TfCZHD`_g>mA}@9_~j@Zx&<!T~Q9J>f>vIiGhl)
zL<{YAiZr;Vh!a%H2y0GwfH3@H;wzq&IY}9Q%zCsXy~Do)<Xi>%6<4w`za?dhMO-Vk
z79%?+PnK)vOY|8O%p0s2{6T-Wz6m5aWT)r;L8HjTpf%M~(mk6Ol^kQ+2^s~AoE2-o
z{^qTGHmiSutg{P??kc*S%%2juK+(GB6W^WjjGit&o>$50;PdwKi){W>RdQ;g-x9%g
zA6iomP{`dJ&n(qSQA$*y1NF+$hiZGFfUrpaSj^ZTFFt3F=R>mQ=OHT~%|aMIui3MU
z>6}4-!GyL!uj5BaMAGNTukuH((eQN<XcuD~WBNidEz@0c&D?Y%^B~4$#yu=M;vGe!
zV*d_^@$*FEyc{Fz!fjd=Ri+z{s)qf8FWt<A9tvglcHFwUQwww4d`p;|+aF$E9!i{-
zcK;cCC<Cd87_$6%$+F;g4Yb#(q~$didU48uW_2`!Kx1r|-$e&xJ9P!lB$1*kt^w58
zdARG28kF+fDdzx4x1SqxNVm{e!6VC1wrFua(xY^)q&izWY5CtiHyOxj`P1g6+4WRy
z@z0}rYXX0~V_$2qf`?_gV$a<qg`3BoPfND7qCCj!Zrr@~b~zI7;i&{H4<D&mcwx^u
zBLy27d@$eSfTC*M=8vW6{&J|$q(l)HBR|Tji3_5r%INpxR(_ZRM1x6g<Rjiqq#57V
zs>ZHC)9$WqmB9WksJ~9P-IE^=q+@t1&$HSOZ=kr`D9C>-EzESfWOFtD2zs<4;oIFA
zTg@1lVToLG!-ubH%$lcL*u!reT^@Y|(Z?pw?yeaB5l9x=k=D&t%IYO$&4{0goLtB8
zYYD(7($&<dO3;>NU-$EFp-vfs`Bdf;2sR-qUfwjlMt?oeN910^ROlSzy6@qzcfhk<
zebk3$;xoi6MyI`usK_nU!?S~R(H8fV_y~8y)SCDXmkKUNugTw^xN(l|x(Fh3zl<T%
zHaCd$G=r4o$5X;`n&RLVI5PUQKu72JeLdGmQLG@GKzlxe`qR}EWX2Jmt}V0`0x(S}
zj>J=`5iaYob&B~*A$&YM{H|*CxkmRZ-atGoST<Zwe>{+#@(uAQ=Auy9r1~IF#=JBF
zQMK#^xX<OxOZ*^FI6%vJigGey$b*;gtvUSYNU=Cy@A1Wq(+C9rw-79uBf-C4@{e>H
z*6Z|x|7mFM-;_mqnD}pffth{_0@GJRQ#NkriW!(@I?M33bJCKpZPa09R$8_T#SE^u
zb6sIy@E;pG2A3D8Po8kca9afu2mC3xCGe@j6T+oN7;1aYF&m8dO{w5v9Ju$bg{;rY
zxVWN#N!#H1J-_$;S&52Jbv%~?c8Xb8vP^xs-sUmqxthLa7SRS#S+q}Z(rPX6J44zv
zlG3t-=BzRCWHgvo=l6Uk0YYY~)4=>{PO@2u;qp;~3mar$HXtVFd*fN<hvq|SvEA>d
ztFlZO${Y^#5e8@86H^<$J}tUxMISxtPeP*m@RaO;UJ;>q^XAHcq0DFXdGhz*mgd$<
z!^8iU-%w0<_^ltbD!9zYs#usA)GME{hB<rqWtSd4N?K8gTt5}~`qcChGrG&~nOUH#
ztGCCjRFA@K=1~c6pLY<H*~z_#)*w)$?L)7dYWdsH9KGhI-59?u8>x4Mr)Ukz1i`_L
zEAKM0qVPM?K_G)9h{^K}yYRxBmo-;Yt~<CEa+Qj75zGq|FNhPJY4Eo0ZH!1Wt=E;F
zV$(dls@^0xyx|p!ypgUbP3%esN9MX`bsKG0dNm=!;C=<lH^E8a<3l!cF+n9^D-fp=
z3F_576WX%9(5jy$8xb3$eg;hODVM1~Sa(%6ijxlv;KB_r^@HsgT;a7W^^-X*7n5mk
z9rc&q&7a3E<5yR&Ujdz}p1>gXizkJNeCLW>g{IB%7y2PEK5ViHWGh==0D>Qn6jrQA
zST6|9Wk=69NfF6+2w3Tse-56z8h;|>vo+V~2samI2#$JzqLBq}|2<H)eozTeVc`#m
zYKZCRQqd|hz~m6i=8!qj-qC;BDvYDP2bq&!jCoFKu=MpW^0Q`QHcH-Z!Bu}_u}{md
z#VvZVOj1$wr7*gn<lLX$uL?qtAlC*@5g*v+rtL3AzlnX(I!0ttS~E{1U)G#8orwMV
zoAx)h2*@q`e$`>QE0R#c#7WNcqeL<6hZUw;rE${R6%HPI6~jKKfnj(9ksd#gCzsZR
z722Z3LKr0qub2J8R^2rg<%$2Oa{eIUxu?>c_$e=h<1>DCTDUy(8MtzQ1L1EY1aN^M
zXzsX`3bFNC6`y`eCvwQ+NmpalgE=_6LLkV3t@Q;Dvo~GoK7&Vu#9A5b(p9<4FlI)S
zeX>EWP)uyjWnuE1q64S>00NXk@XzA3Z1Gs;qkM=My|ts*b}=@mb&T+DzfKM9T8b=#
z3`G2jaL-R~{6D+<A-snTBBxgPLgI&K#BtlZza=MeTSvEleoc!woz+1qySyRUqSw4b
zi%RE^7{s@<`>t?Q+f6Gec?{K?r((z1I9*@ZZUVw}5}+zwr`&-wj7j$;s=uDx`b3^<
zP{QVcc>c+0=NJ_Ei_-1qif!3o*~UKWb^$T<AO9&vKTPShKL!{r?8KP4mMHqd_4Eoj
z@0yd^rf+DxEY&+98H%l(SvBE+TlY^^+a66d3@V*N{WW^`u;k{uZ^QxBMb*b4dyqW&
zq2IHxv}!_T)pA2#vs2hV0>O5Mlrx0JF~6Y6qsO>WupW(L-?3+}t7ezS!71Rt{9drl
z{6B)PiSH0lcGI`-M%K!y(k0H;3*E5YivQv9%hkV$-GgIF&;OdQQZ^h)nT1nG_GC}6
zgBlw@z2{_7bj5b)zQkCMw+cEGrpt~xkwd=FUI{X}gQvq!n3y{yc_nS*>;PE>YIRD?
z3~d7kYTh1fX-9acSimCKMm1qCuB^*us^kx5*49*^wKV|{+%g7oKKtDCsX9Rgr^1vu
zm)cJ<WMXU{u256b$9fUtTCg#ywHUtcvB;*uO~aRChQaR8yBXCR$YG<T#MfhdRzL-U
zUoYkS=jy{4gtI*l@<bGVU5Y;d=<u^H#%kD^FVaqvLMUWx8xucYPJgWPVu#~{xf7%3
zpZe&Ko9$1orseYI>bws3X+`DICIm@+0T{xRVt^8w6*h0^EX*2qUJ>4>|BQWFPSo{r
z5nc}to$nrR-^;|tCn^HDeyR5E7do=3pN{H)V@}gz6-jbWD)N?2YH&ds`cu3K`h!Q1
zjS6?g!?B|nf2Bml7HEr?0Ac=C+;ve+g{Ti!Y$G6_y-n;4ZAKHTKxEvS+Gj@GiV9y-
zW15gRAbvdp&Aw@7`kPwkz$|_9{$lfO*o_srmy5GPa8|?VHlu+;;oa1n)7GdY*N^X>
zOu@*2@xw%X)E4;*>InjcxHZ1qmN)>b`)&3o@jtSf%!WqIix1@JFtXVYj<%KuJB(7O
zmvFSu{3AHvLN?-=+DZcX`=e_-hVM@uD2alseQLXdXJXp7f|p*z+vRMVsg+-ZuNyZy
z#(h7No-Pe5=58La_n`{|%QCt3EGg}^=UkilP?=4|rmo_2`Jj3>i!}r;p$qU6W-2^_
z0uk|c>&Os39sPUDbGUq_GnJuoQSBJ^Rk%30^QZH#4kKx@l1uE^(E|di7pK{;uO(Zv
z9(EQQP~jxAdx}>Sg-B$ROCK&zRC`1mvzqxCe1I|!!5hmmfy_KpEzU2NP?dH)w!ex{
zeEQfScXav5)Xm^Of-k9m?giU3@l)@<2!k;n!kzzCCc%zHSc_$L7do6Pp9KGUcdg}9
zsB<%Oa#CLdG7I75-jgH#ekW1+D5gQNbc*zR^>0A{(mbY?*n_7ra7+fmAUe;Xe#?D3
z%)S!RwX0`hQsBBzizaNadghp5%1<n~Yx<F!qC{5_U*OA31hoA5Rd11YX;I_81h-Y6
zaA-H}DzyD$Kji%CY!u)o_VhzYu%uc^>#i6pV+D_;mTbmWBS${zKzoes=u5j3WYm-(
zSXJ<g6i6(WIi;ueC>zxyQ|_hr+gj*hv-OlvDXtHtvL)z9@0i`u1CDU((0z^vl3Loq
zYvlt+XC<$7HV<`$Y6GT5S4!`v6wXOE6O{2#Gyz0-O<lNKbNuIuBhCYZRS9Yv_OJD$
zXpQ51=HYH)cS!{0xenA>jU*&XMg2iZ;jdvGg#?0ZW$%XkfS%fQ7x-<|^*3n;s#W{o
zpt^pyMTp@Blf=9_(VPL2<xtBRo>8CnFyGlP%LQd3dvXK+q#ku1h_ID{)KnY$S}t;u
zfwN59@V}}pJP1_5?Q^GDN0-XF*dh&#dy}1sQ66w~$B&`u?w-og6KXNQz>$*;jzU!{
zl&7hr$?Ej=ugNCkV3-mn*&NSJeg&^)(;H?h)4}K7OQXiePzky!_b{2VqH0;N^89gZ
z6XGM2O=}sOi9zUrDP>~v!&$nD_njH`0W0u79=uZgexz95(7HNXvFmr1KDSWF_)X@(
z=PJ+B@|NfN-<10t?dB>JXNe-Dt|}ojDc20U-pMP(vZrBzTXlVGIHnfht-uyx6&sam
z?4LE}I>Ax1#j-77Xi@s<EMKPS@T0l|LAO~1uAtZZlS?4<hnEUl=XE=ZJlo_0UUsxq
zF<P=|N^XCGDw)<IZR9bboY{%A5;vRLAKFy3GRD*VBprTGquAaGRRdcR<2aW>S2>mp
zlZfklW@9Z5XVt3z!UkUI3iI$g(0$*tm;^lIv>Dx5D#CZ}J}z@_Uv#K8eqRh1N-=wg
z{y4Tz`(Ufsx$u~Z>_)cjPgs3VFt6i$EM73+wT7&oYg$VyZ~M)gyV<_rR1)n|mJvKB
zN~WmgYRq#YG3ayIRw{O+DFF~$l4LG?i?6X5_1F`6#iVYoQ}>u+Y^GW5H-p=j;MKUk
zHk!tMI@6L*G=;oCci_h8^o~OLbB3n2QY5Fj-br!8kT!SHA2~_uIHLwJci__cFc@xC
z6C0inB43)?Q#ry5^>ONVy!xS=m#Ik6B*Z@gl(}}?e*kWvF5ZU(dIdg}Kzw)&nwu@9
z-7l%ne6Tx_WYY{Jo#p{koyFJ;9bIFyx<HOm4F^wje?~+)dhGd+Awb`saU-ISf&6L0
zu9^A=FJpXnfkCGP`1MW*T&{r==T-=Nv7YOJT5|}1D>y=;sLV3f+jA|1aR}wAb5M|=
z8s++f1i!Z)zm2L$o1jCKT3|?lfJ4evO@VnDNkA=tRv<eXyT70B^jbpm4?!)gI1XDJ
z6TYCYhSTk%v+4gqR*(2#eQ4Yu{1qpQ)0vo3mbh$#7h+zUwh(C>Z-FOl(#}Mm!)5xg
zDXRYk{e9}M9}-OCKy2i$+9Aqafj#>{l028;uj@oT(Q~DDb~E#XNO7Ujd7=PBp{+Q3
zb~Sa_YkzMio@F~)zx2Ade|=X6M!|u;(l3j)X4EDeKV8-YuH#i^s|M8Z)xe*(kdOw#
zd9T<Hl&1w1F?3xIKEsDVAg5xfHU;CbZrA$fh_!SI!w&LgmeLi{Y>9R-ka}*>9R$4J
zmF=FXts?hS=)uvZi`Ue%ZQJ1R!o{YfO4#D>R0<fC`qj9LAU`$Fn;S%u(qM+mZi{Hp
zrI(l%OKClgW)eM*2QOHCn*69?7(G2vgo2cjSv$Q<G~jb#pr)J_!O8iw;@91pYg~_C
zVQng&7dp)5ED_uz@{#SEK(K6noXZPX`b1|Osxl#fLcix>AT#3M-x62O*HsJ7@jF?L
z=y(vIacFt;pwO_+*ex0zEcTvi*!_a^+G<tqryNxP$7J{S&+Z7JEt8!j8jksi#<Fz_
z4P(h#>@S;%@+pMFvjJH=^nPY+Ee7w4%2h+Zdl>IjofM;GbzrWBET?xWe>_k4g|m0+
zBkj+Uic=k07P*cu#FNoiwurg%>Bt}9YweIiBm8zVl-=xPF1^lX_}9pt)`5UJ4%s%u
zW_Z%H;5Oqmwl{9|>zvF{f^h=~?PTyUujq|4NMcIuqwa)<&)%f;R$T?7z>h0keCxoY
ze15>|FdNzouMy|)6}oxO;cdp<Hy^uzC45e@guINfXF`LE+>d0TPcM6IN&VnE0dH-Z
z-(?<yd*R^@N$Z7v=HZS85SO>0Cjk?Z96leHOa*R<l-Zsb{-Wb6aE!LYwq!`TjXS;s
z@|nn+l$>)(R4VXMz1FXbR%1%wSAgJ=O!2iEzP+Up)8fwtM_%h~cy)jcyVS_sEWi;=
z(a4BZpK#n&s6#Jc?qg{9^G)I)`L8QxMl|ssY+*oYRe581A@$If-ParA+l(8*LqUz6
zo|;+6(7Axq)C2sKnYqWDIA}tSG(h369Ss)FHreMU1kYo?_<VQ37<ziY!@veLsskUu
zw!U>JHyU#Gc4F#jE#^(&WZey2YHEv&-uDwZ#IRD?E=3&wBPek&MsDSDoPAA^iodJ2
zC47<26bRLw?sA=(Pq(|L&*e1iI*AHO4w_-EY3uk}=OeAOIxmhpm3kKSJu*1ZIogSU
z(hx5}h(*ot^Ajf)%q!~(Jyj{7b6{v!vM^UtbkT1rz-Md5qBfv$gC@03$>hbjSdXDB
zHk#~5De57vpQmnlPN!EYI-J+EsAj>r^Mcy1HFxo5ihl&R*>?aWqBGbnY&-E@<H$Xn
zxzo*?zqccyuGD94HTj-#jPw^bAc1*H8BCtsyUxu@vQUWgl3$L-Z)CNLmguyxvUKyB
zwA|adm*3`JT}SRB88&;NlleUP+hu)vZi}Iytkx(-4OXU-vZimiJ69vIU{QBwrgj#i
z8@>zCwb3Vw{f~`i)HX?e3JBk?uAHj=Suy9jm7o;$HK63i;Ua6T=&-o^fEGGG_pSue
z@T~Oo!z-SD1hF{guuolYqOC#t+}OVbF!eNAa{t36WcPe3V1KCkSiry~;o7mWPc(On
z)o@dfxSEZq;(4`9jDup|0A4rWCl^RZ(MSowMW;s)X9OTenYhU$#402wUUx6n+?ytt
ztVo-GZ}0<Pu964^!15~0I8FP5ECZox@mAk3&(Y0%v0BF4qWRbD;uSfd=q@pQ1N-MR
zD}{6Gkzh_rlU>uSR_-87s`fo&g-SsmIuaNM^*7*=A|8E(@85oa2jGr_Q7V0!UkV;R
z(Rdb67NTGrLj-J$_S)W&2e3vr7zh0UofQer%hSAaP9-jk*hC!Gk0*jGz|mv(iyXYi
zT6GWQDc7+DaR8G@ziND;yHP;Z0|KGVn1*M8NeL2NP76yePK${bXD|eLkL{1pGA7hP
zbVc`_&WSUVKEFRHG~>K)9HgOG?8)8FP>ml4y|OiJJwZNXFF&~gmS|#@m%fngu;8<N
z2J?E5x*Oh%#`8h52^#bt=!ZQ5)f#5<d9J?P`dH|mxD1q@?rwTG3#vHKnmX5J&$j3I
z`l({1-!!mr3ov?AL5ball)kKxfj0eJ95bw0mb=E3dkfp|*6p<RW0)lXWn|QY_`>Y3
zSR11TfkmXx#^6)Q>30-{zdcfU81}kAG-WQY8u57-AI%3;Cw@C=Ex6+whv1Tx#MSiZ
zRy>R=M=Nn;TD`qFe6#nD>{yFyHZ;Se2h<s$p96JjS~M5fn>`&biNUr|ym%XcUx8*@
z{<@jsMt|K6SPch9SK7z;@{@pXFbrA^?a)^hMyV2g*m!3dcddkLy5oEU+`UxtU~Yvp
z;ow>$_KP#{EvL!PjvlpS<TtlHDC`A_JORVpz`t;?mM%J`lbt{Jwdeb+yKB*9N2PJz
zw2l8{(D;Ed+dxww+iMps5scf+HWC=S8PI->XKfTx?K$LB>syG1Vz^^-<gQWAm|C~t
zMaS#8^Ew(XKzH5Bv5XYBX^!AEh9vDR=fP5`nu|%hEQ)j(&0#!dThq80cIyzWirX&K
zH#(~{vv=qUelDAI&mvQ3y%_r={=YzOmQDQJ6hT)3NR=x6*7B?_tgXFnmTmDIk)f3s
zF8-noA=bEvH~7#sd9lTlAV2PtXpQ72=ru@;PL%bde~u2?KNmcX_`4QA^FoPFxy1yV
z2wQHsnx0~QEPkl>4*oeqRD}O|3G1%LJ-Z9-VDCcSuKsI0RZ^?|yzLD*Y#058#uh|(
zzoy9CZ_JLovHX(3-Lio6I!1K_;I2GKlj~T!F0FkZ#<p?$N8!@m93DyF$^EyfXuojr
zXf1@Pak8~j{f==$n-KG2U28#XD|b<z!Vx9q^wyJIgFrx}P6yBYoaDu|9c9-Epyi3k
zj&<JO)Dh??P!s6KukW0e{8647z=rVl>2-Dt{o!s^r$?G(ji2DtHlm}{;1Bv01$-RF
zEh3*Ser<rj=;U+6{YSJD0z(Y`C%Nq~f4}jY4`Uvs1kR9P$4M@Uqu=)oHjCXGZV?*A
zFb80R_JdRkJd=_hb=D)FEQFs(=?q>^uE%ZmLPQ;PFEb>M<!N4sDGa>4<ud(pQnWv~
zGQk~-*DI9;>kX8-;v>3IXXDWZtH?2}R!TRjznDapVMk&iS9lh?z)`ID!7dQ^F|vH&
zp5ENtsE?}?dE(IM5IxP#3H21XUN|WUxUpy4aA`Y4nuY7?eYZ3B1LQ@h+`?%0s_Nl-
zN1+;d*U(T_hjBpLFWIvmvLPVv&&;)&n+_pO@hj|bk2e>7<Z7o3G?(*mon9)Kbo%<r
znXg8U{}e{MsFY?vc1l+~zvHIH(Ek;TNO|SEVvrlacR{t56PfzGuV6N*UT)X90IZm!
zcSl!VtSf@q7>V?-GX_@~NN`&DMn3%BqAU+IQyKknU=F%N<G&eMXKkKN6+U_6xIN%`
zm;LObz;zcB&r;B%yTNw)3zOnP^vM&*)yM;emam;NUDqKem6q*okt5WG;@n~}?K!H4
z(DlOCLljR=X7C%+_qNMB?k9PPnK4v{4~}47eR=(>Ab@nRxjEr5^1)BJLKCP%WxFk3
zWRqRNUMByN)O5<mX1+|1{U4jIT<HWmXW7$nhdFW7MrNFjLHkHmuZeV|m8q}TreQ3P
z;k%C`KyVDLweqL6+ypC@P2cj<CW>AYfQ>Mt6$#QPG>YN%FT@52xGwzcrF0~U#W{d7
z*?4V9J@@VzHbP9_0`t9#H4t)DCd|o%z=u&3VRZPU|EbAq2gu2@P2&_16E6qhh6hvA
z!Fg2;b1Ge=-IIHVlo&9mK<r>Sa^g+OmM%Oy`h-Guhwb+-6!3unZI4T{S{+}^>CDcg
zZ&YlK@d@56J-%j8JcO`eGSM=)OrrW#&Gvy=$25AJq7e{LV&rF8Z|jm0!4c?X)kx|e
z9T_0xzg64nz*ukW(DIe$&GFGWfH<_YDAyU*f8Ah}1IF?j&){ODy3(1RmN^ZST3PAV
zgy9>Y?|+^lqhY48^7u5joJFAzkb$T3F~8Zb#;w(!9xFwzvBXAR(q1+(s~dEgqd4{i
z9ehi`K0}>ww@!~fbnE0gt&!IG_j~7Ujezsv*Vd)Q)oLZb8l*p6f!qs<U+95lxeXza
zP`#tRRk1X<j|rD0=WdHjKsccXZDnZyiKSKelP|8O<~`1{Bg@)T8Qt{cW{P=Wy47eW
zBc%W;WAD#5aw2c?&<w1xxnVxlT_XSqqjS$%ZgB*ui&tRmvH}W7Iw%;q<G8vAth;_i
z&+qcqV`VCc|FAEjo$<FjWa-ru?qrBI_4O9aWQMX}x>_^Ck9C~~8wuOifDJh@4Dj%}
z=Iz^Ah2$@sRSq-Hub)*lT(nir`!N4Hmkkd0C{<5=uCxApn&7YBQPh+jiA<>+WOTkv
z?;;=Sz8XhDR~JViOuOB(^l>_E&viVo)|}QzWy|c-@lV=bk~V*R`)dI|lbqL)-Tnsl
zdyWfW+B2>l5Z}fEUU`xl$Z_!O?=9rEjAAg52XQWaw2l}M+VD77N^pR4^u5z0HTt&4
zjXZ`W=1hD}fmO3PuIKI;?q27#81t~}=WaWz<_ZB4Il>Bqe%jkd2TjsCXLutaf5>T8
z?Aok$%4u|FV$+dMdKV1ESfryRCf2+A@i*-!oucT)mr!2j{dM>s&)4iMTC3oF<il%;
zkGqCy?g)~{gEXqJgA4%otK-N9LOp(JUSB3?`h4z;UgPy_iFeAbZaAf@a#f0WkL`o;
z*ggCMy-n1cm)3cj`VW>uxL$C58#N9{`aPxnAUJ}qG&d2lmOIDwkHFQ<_ez@x#ZbB%
zUh-5mnLdP~E@|_!)aM7Gc}jLSp1Pe=Bp-?-ND<>XZz^T|j%S`rz(@iUw2ZY}*A>QD
zs37*;fxb{+hZyG(T9Iv`1E>G{V;04IA_j+Xq!dem0v<DqF;se0njhcnRRZvu8b3Uc
zKbxsZwq)d&Jl3D)S(e4XgpWA;SUVuzRq*N{y85BcHPmXOS<_b2%wdZ+WU|&~<k42$
ztE}FX0hW;EmeF-n#q3@32ZYC9{>*tXi-u`~q6Jute&&;1V<k7bS;N7J=<OWM)28A9
zmQt#?)I#^>XKYaoa)0?xCrwIb*qr+SW6>KPl8jDlF#Uj~wz7o7X2K@g!Ni4Wi`Ayb
z$~s^iEIIU$M=L;yd!^(zsC6$uqEyVE&KiF!C)u`0TK6u31lKV&qV+gn%MgwAsp)~4
zYqp^>4}g1pauIL(V(0<zZA@B`o!e@hh+*DxrjbkEVbNtD4APX?po7z}oge*srC*@p
zrXtjI-JuEvJKa428h;v%b=8Q~o*5|L8{)g{f{9I%M;Fswa<^!p+;s+C^|~9g*}09v
zht6a137<W~*OT(;#vvy<j*0Ybv7hdvem|N@QpLg9T(V<c>!KOku+4VdTO6}7FUz_q
z%?{kY*%sypq)FqK*%{);pos&9YmP3n9AY@o@%lFhnn6j+Z?48aZHW*@G|ivl^-3~w
zFHZ3=FnHU$dRK>c*zbS;u-p+a^h?{H7BCHrpvS<0OiWnEZdbFrNzBBzwRb=|*c-+A
zxW#t|*gy~M_u4L$daBjbI?VMGKemMrn>eEcGq_TAaXP#$B~{E)<+*_l-T_UBFOCzJ
z>oBS5NoHD;1sMnzEc%1QLLh}=IRm!LET9(F`Ge&3<Xb)6E<v6|qd~l4a${xze{7&h
zbkb}5gic;d-azd2l-!$8sn6f`vgd0-@)3|7jpL6U|2um8BUn2t+Alg*lP7D$^}MFm
ze&98s*P!Qb8^Y9_lCFs_d)>&M2n?yjtLD2~iC>Mq$%|KZhP>5&Ww4bO@3X5G&4_=L
z6u*Qsm+=dv8)AQN2xpuEDNlY6pqi+6Do<Obw#Iv{D`}m6ii_sF2DgPJrz=a4Or9d#
zO!`}s#joh)72;B1K1KUwWkrKy<?SYj>Yx3&9%U(Rc{IFIpT>kTxb)7ou=$<63(?qN
zS>}%9FATo-v$u-kI^mg<05w~yo8RoeyB4lpH5H`XvCV=jrdkf{ZyLof(@60>qlav}
zO*Zk5<K74V%1Rpn(z-}ZRp0Kpo}#^YR?<=oznk7K7n<4CZRjfcFM6?CA1?!O*Gi83
zy*BH$o;#XbEHK=}t%2_gQbjXZKi8>)xx8tUMHO1D2D0~ACn`>pOMdYw%9|M>RNQ#h
z5P~<f8s`iqqDd^~x?&}#%SB`1&3^YNTAx4Vy!5gb%2CahDF5%thch3`qmfOUXhF33
zD}JTKiX4|d9%pv<xE6zH<);67$@!ba89oDk2hXWaD2G~GR~T$ELHVzHp8a`_^~%-f
zuJ^79g-hrzWGn=S%hK!y@nS@7#d;@)Q#LWg?ja^cnvw>){|NRf|5A13khF5yIXgR3
zUqvZ3ukGs-6<AGIxSGSwA4z+c_%QTcNei3MpxW5@SNko-Le0yVNm8{BiBqAx<66#b
zZy$*<HT4GaU~@H}<ZUfx$UYNBMwX`aMzbtDZNht78@%^gR?WZW5Z%06^LOH();bId
zwR?S9kq>>Wl>QiIpBdNskc9B0vQNJ=tOoEU;q8k|gM+x}y79y3tNSghWo5-RrmDSb
zJr%6i;|IDw;$YAX4U)S@2P?NUBd^wEV7vNFUA{w*GD5d!msK~r=t^CYyritzv!*}^
zY@85CX)7qldM^2AjDAhx=g!eh)9FxrsL%Y>OqQtpgsc(S0C@h~VdK>atJfpuu%O%v
z=&E_wKZ0;CgtW|Rpsd7{tEcpP75BTt0pvG<&Gtn1_+K~zg8R25kiU?1$F|8`PMhkW
zlhMSWCwE=+W}k_&rlVl+yHx`Ak=e+PR4$tLEKcvYCrq5QG5@-P7(3Lv%+B|nvwpiD
zcEh83I9Nu=pyoXSrGbiNB)G|x^_GJ4yc4Ic==B{YRTPY9a&(2GSp{Xoa~z5v*$y$@
zI`Dc-GJcMI`vOz#Om%p#$BN;UUr%lI-lG`Tn!;FuK5k{%Q(LLP-GpBFrsWxON)u1X
zB*rBNN<|JeKVm=fb}G|VHErAhZiS`OdW-&08dhEh;TL&x2XfO|*u*jSfIhunc#$;d
z2C!IsfkVCdD#V?<0!;fyfWh^NV948(>m14$_hIW!xD*L-K${H=;ND}^+vf$<UD^T3
zy7BmQvT})ycNF&{GUIxYDy52TyBYpR)vG3C+F*#OUvLZ*O>MKcF*|4Y$D!c|+QU!b
zT5W0??(j?`k<v%>w<M#qZKk>K7`$31stLoE(L81F<cp9mfYC|!cC)i0acK6pdo$__
z_~Y+Ea>vFwXkQ(}KY|M-Y3l&V!C#_!(&S1@>g&gqesF+ICty>KT7>MY(>hjUb2-rN
zn7{z>ECq~x8g{Bw7*_hi?vdljfn7TNcS*f~>K*N!w*%hA@7Nrf-450#Djm;|iOmuP
z<$EBXvy)=QdCADEBvD}}9ij|hQcuw`DE~exf`PupnM<<>F6_WSL5e<4mD7XwOts39
z7NkBYqUB#hNXV(;>!V9EqJ{TBf9|aVig}IV>at|?WQi{duf6f8*%Ejsh<f5^;MLcq
z<qf8vD>yBy-N=jWwZ870CpYrV3CCazM|*sRTUk!LY}A;+<UfMH+8v?aFXYzPQ`e9}
zRP0^d_~xBM3^9OTyF&Sd7ny@-SU>FeR`7`KLuTaJ&ikBN-AGw46r`r)a<+SM*gB@k
zhA;X!TMBEI(>-5n?ENG;N*ZKSXNCTCXe6Xlh$zuc+wGH6MHRoX87OEEqhB;U1(<6k
zK6R4eGV3wQ5Ulrh@hzX<k@Jn53=v!S%fo(sCo8D&HLxzH6?`Xh{q+VjfXERlQZlY(
zaMskYdpfsA?+$2saA1dRrT~T-=f2#YYuXK>GjxsIA-%*xv>+H^I^%PRHs$6{wABJ9
zRHzGag7%=Vdu`F@*O^X=r^P_J3s?rmfGXbviQiD!L)E^1mP%BoNfAkI-X+Msw#O~Y
zRbplahSs7Ei`~aCVh?E1d`-0TOFycEqeQ8P4PgGjdecz$l7%Y1NCM@*xAwsK)02Xo
ziA-e6jsF;-{>f!3asJ}5Wp*@)8Rxw}Go|Ymz|^hMd<a3d<nyf(kVZR7#N!^jQlS8K
zk5iXOON3d(QV4xneWt{ZfFbzvNCG^%U9@uACxXBi*FLix>LRY_k>RJMtO;q)E|W#v
z@Juc+2m%*lz<1oH`SdFH^;|ae1>@hVWK*fi8%0PANI<F?`)G0c@iD6D1?n+`7L$k|
z6bzH<A|ZWN-oG(Ffi$icOY9<Tw|$5Gy+`v={rHccfew`AYI_F^O7T{1i5uoAn4C>f
z-`+%8EhIKg*c??SMo&9Izwt$Xo^}zRhaU&Z9*o2&(pH^RbZE512h?l7={l_lW1*MC
z5v44h3bwpB_3op|TUU9y8KRRwqB=5h@W{hfUZTeddQ67qdv~e00AS+g>4Rq|DN=$_
ziPFYO^S!<&B{ze$zK*Zb!X+$r@QTM|@N}=JZQzC}rqh;B^H|o=PL$eXF;8^m2E*F)
zNu{BXM|R)+2;+D?`L=$+_B)L?(=7VqUZ)C4C^p~0Ym(j+jZ^JvV%2DJRZnnE1w68B
zEP@H~2OVIuzBRKi<JG<X*Hg}y@}umC%?3M><u}pWajKTi)i9N}{M2jdTNpX<iErNg
z4tWe54&F{@1;>;NK+o?rFq~TB)x>?T!0MG7Bu?=lRg`oKhT{0U;ucuK299Yv*Q1Mr
z1uvD{bg7`VBVba$L7)Q-OB(gJL+|Uco!~%4&U+^VvbNiWZ(UEsPkV%vPQ?oz%r@`~
zYfGo?C~l1|S{y{i+?_2vb@;pr@`FfAyxb}U&!uhc98K$yTZ;P!LTSzv_@V<)C&fL*
zZqZ@~9E)_NwiDL&z5}AA#UI+-L*(!yLEH!aM~JEsK;5h!A_u8B)_isYt$_pPqYtmJ
zP?OS2ML}KSr=4tz#IkiA78!Ii(85|2XoO%LX=d_x{FM6Vs&&}<^{>cX{&JSYk{d*%
zgWy0WWl~6I=lW6-|9*Dki_^)jf)7U<b^v6FeY0_w)0uY99PjW$Iw~Aaosw!uBm;`P
z8wee^E*8OzJHe%`wti6R7<iHvHvSHz@Yv0pFK-hs1Nlj-b;g+A=3_%7m}~p}UWuS1
ziX=`dKbnd?Ons%Gm!t8;;um)gtS)5Li}?=483vanhBiihsJ-rXzXB#t;;g_A!`_Go
zR}Iw$r+8gWMHMW(>5pwGL@7qcl*P}Z3~Ya`aFOh2{&cItpLtg9SD&7WN?s^lHVS^3
z?Hhz5>f^0}1VT@KLp4VfmN(sqWshH+OqeRVP_5j)Gr2F&Cu-U%=l9j{=L{2ctcxh!
z%JyY2l}O^^ZB~K2_jZMYp%O*Wh4Ha2>k~(k78f3OwZD7=;nt1O))YTNN1(ak-fk6?
z@^_4u75Iw!oHZlU?1(QyXP-ps?PuPbs&DLlOk^h|23=zL&Xiuad!*PL{p<IcA?1>j
ze9ly9-Q;CK$zsns-dE1>;a2*?-jzLGP=b_kv+Og)sPOaae{waNFp(Z=menfMYmnF6
zysTJn%3_@F*?v;j(J)b|KCYO{UMv6T=US5m@7JH12WIWb!PhGLbn4M2_UpT<XZ(Hd
z$Pf9FbOJNQdPiUC^7TI@UbuhmsblnH@!3NQ{&xkD2R~jq$;fbL>}kYYY?Bw<FW?+}
z>@cxR#V=xDk4G+^V)&SWCJaff**(rrMJ+a)eoE|U?+F?5RPy_K15~?3Ef-P`TyWcO
ze97#Fd-2YXR0_7&QBVCJU~7icX+L!$hU5FkZVH_4&i&HK^bM;TV{BvkM<z!nDpX1y
zZhQ!{VUjoA#@sY|f4cPTgG8zoMbNo;dln9}2A(RoJhY^aYhDyK8ugjb*o^pfjde_2
zvQ~Mqpxj0zHel>MBu2s5TVFk6X4=m>N2<CeNcYmu&$hIy)IwkR4VC8#{&Dr50>w_b
zrLDXL6Mrcpi?QG<M8+P!zIj_KgXWw2*$nMG&()g~uia=1bjt-BWKU|9y-F+t|D@!w
z;#rahwKOovYke_#<`45G^QJzDs|&iCFTGlHq&zvprp0fB&Ff_!T+38MIS@W9eOf$-
zx2M>4JPx8Y>01-{DoOa9!%>C1%xR-?Skhpxp-Y0UoaHU~o!P>J3(IDpgeJv_;<GGn
zA%n%}Lr~tF8Pfm0BBIGwSMpQzbQWYTdRq4e_t|)f>yUan`&D~Gme;_uy*Wi$?Kc$e
z4GFw)1C{ALgJ$DJfhok90P(u7fDil+(SrGSZFNqe)(b&6P`fM^we7ZB`bUDzt_J4#
zM3>CN@WQyJDqQ`%sGi#C;C)O3rI#4zR80$4|4AwQ9sPQtn97%u2W>iVrT8ky2@7uD
zUdhGCrGWHX!EyVkL6}kCpA)a$z77b{@K-0?P=!zNz@rh@V_>0n^@e`gzVHcB!icfQ
z9;xQlqgvlrReHYuqcB#uK`NF<xv!Mc;EC1T%z-+afN|Ee)dq%PKunCpc=F?GukHkG
zWb`CxHQhyP6X@3CD?Kq+J)fvlR}Ds{t_rYvu9YJj6aV5nPQxUQ_e^E83K;$)*IvPb
zr~T;VJfHmb${#k9G${K0x7@3V6~0-ACK>vKmweY&MLD%kntFIk?|rQJSWCoaS7>PT
z0mQ1QyVEHR>@VaMhVP1J8XE`*6PM%OFY@PGJ2!U`>fhNYHNL1DFn08fNj}Mba`<fL
zIk^?;`+lAKjoM-`=>gIi3nY<1Ti(e34KU-JFaoqjmBghJ4q@M(UeT{fQ>=64wq&Sd
z;u1?u65GZ0Ne-zXJ+cP2eh@fP3oiF`;dDH7E{?m@aQRWH#IGxOa}DgkB!4u)%Jlt7
zk#Csf<c7trcJ<<10;R6KGwHxmL*4wB#oaU=RILeIbvxGEN<43WtrRfezJ))za&DwV
ztla&_J_jR4J&DM6d70{c%$!y^Oa<UNrac!uOTD3%T6=d$(jCLoMKkfuNrcDp7!e;=
zH>jw8jVQT62bio$JWzI@YySdw&5Mh#IMB(onF=kU<dGZ;I`)QOnxniYu_bdo^);4J
zBKxw?Hzt`^Kog%#i!a%bWs}#0CuP_(i}LYK_B<sOD=i%nJ8#G@jI2p{ERVn1rB0vU
z0ETb2#<jmp=o-Z38#YabRv}#z&c+axJuK?eF{orcS|?J%R*BIuqh(1REH#(iYaXJl
z08O8xe?x00fyT@ByC(VBd8pW61C61{t#W?)B?TEJ<3FeVfrxEpZ>0M)O)l^!ZU2zI
zic~V$%yG+aYOuBb^GM4m68w$=PaAvZDoT<zCAJLY-!2Yp6J`6?A)!alX>b0};=05P
z;*gSB$yJh5QIo8Zp*juRK-b9rM8Sb$##<g+C5{U2ux>II8^1}qr#Dm8rStGlii(}3
ztj?*_{|PB9;7gtwh3%&T=_6eB1hM7(G-Z7tZQm^X(Bi)9`vesGoY)uv66LCRAdF-U
zsEg_$FjTm7)xg0m2kJACO|J6Kc!bF(jfKkaAo~pX1^35o<Quw{nTfpQe*`iPK82%N
zu%w>~uOSNfJ?Gr%K@hyGO298gY;k6a=UD%HwXwS9E!j7^`OL^<&AE~q09e1O-VZLn
z7$u0w#dD&9+t>=;&w&mLp^!nR4o+wMGPDVYbA~EnFZjAs(3!jGi`2I%&%ZpR>qJ^U
zRe~1w&%00F&h3f#N0$(;KVWGzW#5Gcv_|4i=$;gOts?ST!6Db1lIo?L#Fpq-)cf5V
z@RB0b&!;<k5^_uB0I#|j)uxv(x=>CIM|8Q|%CDOR{17>00OmDz-;rv&`op8*U872v
z>Iy$}LR5fQv^$22R|PkhXHGx^uHU-l*?_z&wGV9Nc}?wGIcq8zJ#m{j+<TAiv{LOK
z!GGk4gh9=vWL<OOONeIgi%@{<uxmH#h(;4xCVfo)T+`_4QVUNG<%?WHfkG>lBBI3C
z%5tisDf1crdPI4kqT5V=_+;bctRGj&v_@jg=eRX4!*&mw0R}(MU7a_)OPI3P>EMm{
zI7r1<7@}5qixXF3co{)lS;w2>>xxs49U9h^zg@w7sNvpZ*N94?&1Erb1J$ih+BJd|
z2t)7gYuKAy2pBd;2WF@>2mB+L6xQ?jknqR-Kg!V02g<C3b|JE;`S}~RL%!_?YViiQ
zh&lWv%&zj8DS-I(XQ~F4!kWNx|7>ekSu>KQEBZLs?F-l2CFp{qoXxEnPJ2Lh!E>Gt
zZ_(Q+m$P~7x;@I<-bfT&XdIm}6Y;eelOs}-e5sl<DbgdAdp;tgwZXH;M|Gw+Sqx|M
z+3?&~k*NG3%ZyKw0OPtGcfAcMj@JE$`%)b*Fc>W=$qdG1!RjyLkRT5qfz{xzt^R?4
z_T(b<-6ezfWtZ`0xA}LSK`$}9Hv^)Et2x(2Oustbolu4Q!IBU!Ya)4rJqap)P0N1)
ztWq=ER4Qf@Ke<?(-PG>|<#gp<YOD#&n5@x6&hJfh0i&>Qt%LicgHT<IL1<a2YD@l;
zQl$G-7_Zzr^BP_IvHp%fe*;H?3++Z%GaQ~n6{%K%ycApkLe@%jUn3W8f<n(cuwYAV
zD}3GRd+5!s<>;A1R=XG65}y#?YPViM3m4sAC-E<}8v?@I|3Xz0lS8fi4cMxLv=-yX
zs%xN2aycy#^G=<=?obhT_3sW0oc9~}F*`tB?eu3XZNg~fxk<}6y)wHhv8+F7o&BP;
zfQCtx=(q!2h%n$9^~ugms4iXaX!6y6zL`fdh!<355Ol}jLg3t&sQT}h{te{LrCThb
z51L9=zn73O)Ssdanltu~CrT2yXa$zGci!*)ivPI}5wqB`KB_xSD|4O``$ypQ{aN<|
z{oz6@s}R}>A#Tw}(a}>ap?y8tkA0Wl{wOI(wA|2D&5wS7e)rN2MnWB~${^&43=~x}
zzKoz75nQTqa+_qqk)y{*Cu5@hLsm+zi<^pGO={E&fA>KszHiO0t5rUy?5Zw8q>5YQ
z6sq0daJ}23^nycetU2Ta3RT#`oMK5h_Aw4NWG~rn@{2Ek&sad;Z)VnIxC^_&ugiZG
z+S>FDw|XEZI16;VMmPHcFh7QosFAtRaCt-_l`4uArrt0dxbvA+y@sdP?cr~XzRK1<
z2OFMOqDnFH<(mXNRsJhBK5+8G9$TGWq|GKVx~gjirgi!@vc<fFNXdXNYErJw=qETO
zg(ZGo$`O8|E;R6l?erQWrCMF3FU~hD3i?RK0oF&pJeuadhj*zIkeRvnFAMY|&++!1
z?{;2C9jeyC0GkiJcPR31r=LYM)fzPuZ~4%Y>^6Z-3xTEW7K`wAFJZoVINc$B6_<7s
zf)*u5{qI!{&q(%9n95{mQ?>oP%3N_VK_3!#)YTjLv#7Z^h7L5L$R#5qv6S7|WRNXT
zKI1dVMEVy0ieR({T52W(+<30V1}W5BzX4jxyh~ZOa-%XHu=SijJ}ErbN|iU>m{4u>
zoX5C5mj)Wl9dfZ9P+zivALL|K$RH}#JD3m;)fzYBXWegt+Wy$Xq?=gyAIcnR^M?5z
zeCW5f<|q__Ku~kpR_3Yr487Ng%?`x8f2g0b`qB60=S;ha;%oU&8Rm3!5-cSN%EA5u
z2P6QqMWwSTiZMTUHF?Bv*JEzV;D#UK{Uc(L-%+DQ!B!?FrdI2o3FZWUyI_+qW$3yn
zsW#0k=qb8B*pc?is((H@k*Msc3`w2b34V-ZuTSgpnp9%c-M$_X&VfPCth(Bt)g@}y
z`j?e8Bm*^18?BiO{q=eDV)%_<KH|eNM%c4&*Yy&kQK<uGLux(DJK*kag103HW&zFL
z;%gIiF4nHjoaxLAwb&KPaNrkj6y1UA5m2XIEmS~|6{l*;#w>H5jmWkZ8Krk=2VmT$
z7h*2$>$B}?^AkC0#tmBP_yryP?WemiX7s*-n_1Z3TAyYQecpJ5NlSXWxK&U+M7zzg
zquswO<`0aO+1as59_W#JGOT2)wh=#ijO3gOD3*IuV$r}AO_{|#?1GI430o2ob<CDN
z`3Q}lh}VuZU1om+)#V2r#}#pRyS+`-R`2m6O{UtfjlAbnQY*IA3yJ9)o)wjza5VFk
z#ei&VchGiR9wL!{l(qyk-8I2ZKMD1nVOD|N*4Mx^Oh-#(jI{^7pDgy1vh;LdUj0fw
z+dVKpJV%H?&@?6WrD88sxibwkg>!rb0=}f=Q58R{AhLGweif_eh3$4^DD<8*e!&NW
zW-2xb8$z=c9~wP(yT-<}UPilPQ`d;T`Eyq0y`PwqH54!{M%p?UsEOz#JN=C5+A*jU
zjj6<O6~!#}=G5WP!VdHB9EaJ+nax*ovKCYHN5RqAl+6h}s72OuzvOH%5%aFTr=D3=
zpOcb1$()79r#D@+IJ6z7I}eixq*+f_`u_s^HU!CY;V;D)JXK(+F6&#d_u`O$_3PGc
z=JS*?9m;!(^8Gi!38HwC@nMI{iP$=-j>q1;FHh257~?8gN9kPAl&@xbc%@0po)2>T
zaPC~5l>Y!IdRA_=;>{z%wq|`>Qns*&e-zFK>DTnHCGnT-nee;7>zn<0NuQmu+S@#g
z*!196Y-!F)_B=_}rybets|ap`jPu4Z^{+VnpucE6C-#5vd(O=r*NCpslEY6VZf89i
z`N&__+P*{hlm7q(Fw`|=n%BUZAkyIjZ}x_u!dY?t$^t<A&3vV)c-vX=?zujvu4)$=
zmF%aP?j$bQ^cnl3>0Q+D_G!rTF&M=t@<7({Cb@g!J#yz%)b1Yc_VQ+)X-VbXyPxIs
zu3yCs9@HZ_J!@j*;uUorE1&T+B-yL_SEF}p9s@F35jCDyr@c|OK-^=R($mkGW*51p
z2$nY8j%zbWkbRu_JvgqX!#2xxVq-srbIbz5Mq7&LydK47+T1U_V$PzJmhEG8EMRo1
zI)q`ja(S++#4v@qGAQ|vTIMg*tlaQD>dd;_=4Ed#;~gta%yU`r%uQ3iwPtPVDJ<nF
z`@Gf<b7iSMmfmCW$JV>u2F%SLOq${J*pJ%+uORwYt9WkQLuwr0G4-wKnAF&U^TWSr
zh>#bawaIH54aL6V(GaC^yR~}OnHt=Taf;Q{ygv=?o?UW3x&<~U%E<Wd#Wp@n#hp(b
z*0H8j9k(OpuXORh!x`?SX&{dzjP1`A<9;g8?sT%y%Nw9mgT-;nbE-1tM=dgzOyp*W
zq}?4hX7cca;g3p#aaQA{WNu3&gf<Tq8k4sQe>!xZzD)ErR5LGAK+Dmh%D*>SXOLrD
zoEn49Bc(b!qRoTSff2N2CBdmi8FPwBK7L196Yk=G7`#=FF-J<THytU9^Dbxt{_#fY
zs}$b3H87MPS^%4e85GrB)Uo~CQpU6Z)m_vp1@B6~Ei-7K3AYCO)a7Ee4(fjHKcxUn
z7&Sj4s;d>Jh9-b43>tF$t%_n})}6cM_o^rtD^FalHk>|g)Dlc24NirM_NZfwRLaGD
zDoow_mRBo|M>RBA<8kX%cKoV+Dp=(CRQga<O*>gO*FmQK`F0l|=kI%0GX#RQwEKds
zn$j;_LPv9BnAa+6weZUYL~QZzUT<f3<-Mz~@cr@Ji23|0bo!$c)C7z6d;^uOuL_~K
zyhy*FTE?|lO_<J05t`{d10UMFMGSGNF2?|3u73!j?#UjjW8t{vvx%H@jGij`cwsi<
zWqQYlei~14r^4?1vU*oRZ}72K?_}4BDBaxlBU?kkTf<S617|(>tPM-U#g(v`VdyLB
zc>WmMFW!xh;ZFYmg=c5S`ZAwhDLCjaY|-&Pt>L-%;y5HD*=wI~1!`AKl+&hv_g23C
z{{V$$X04F|pQ_aH{5iBvw(5S1iowyormGPy$LHGo9q|qJHdr#}`^OclXZt+(lHxo%
zEKGXwUwuR1TX=Tj1>^mSww@BVPvBGk00F7`zji;ZO!!*M_FmMkqy}r2+usBK0IgcK
zf3h?#xt@7t{4rl{Cxv2<q(Y}(4#us*oP8+vC{Gb7bv`)LzhvzuK3*o3zuG7KE2+}|
z0A_6#)k2FH!S35KYw9J@B3_C=Dwq2%HR!b67<5AR((Ziyr~C-gtiI)pXX=&E*?c6k
zwN(PzGDrJ3uV!667pSQg%@?t&jX3r*l}7rWU2pK#-G1ySKj0N|-{I^K?#xQtzuB)x
z7QuQN3|o8FiV2g?+~8vHL=i3oc&l;vP7%?6O6+#~3XK>JxT^Md7qlp7z8s3*X#<h%
zUACKOv8sg_BC+jmBVL47y}j~|j4RNm9#v#>N?Mw+#;eepom*{Dj_EyV{{U%UTGNz^
zqR-_+C~B^=hVM(~4_ZnrE{R$&-Rd^C4c>*9bvw>ldWF&a!LziDkJhy;Q|3d(XU4mm
zf%4YHoFP?DVcxMt8ga5i2^{2BWz@y`)tRK(8)v;`yq;%bzglEP1b;74pQT!Z?t|;v
zs>LDQgI3}BV?*A9MqQ1L*5nNLHJ>&+^X@9?m@Ig8`F*O39m5DdUwTYPd{LP?ky67C
z+9XJjjm^-Wl^)W#!5sFbSr;LOOAeXhs0kM;xTiDz?}Je&EOC)ehTwTpM!`Tm8@&KF
zNk4SHqN#?<0l4#0$-B!OXEe)rmvR%$O9I`i7C+r#(2lifravHm&%IKOSzItYj+FVN
z^A$<$R3M6CvT?`z)ppb~NI2b|n5^C2c31)hUbVYxRzjmYdJ0woNq5Jv3}JdzKi$gi
zt(>2jeQIfLyu*eSpC;vQxN2i2SqGiYNTrfRQWLSL;$Pk5@_i|iGP<raxKs+(EI^xb
z%gT<`JkDKkrvuc~7Iz5DJmqE`_yaX7Dy)0+{{R810v*OJpz;MOI-ivN@9$K14N*rY
z%kN2y*0N@fGax7B?M%9mvW|H)%X@;T6!1IKW>e-9xNYrBMPnh8A-?eRsbiCJ;eJon
zsqQ{t{c3h$AO8Sak^=0Aff80co|vkv+q=_z;8$tR2TF&`rcgi0)YD^A$G<~ORouBe
z)yZe`6@Eafy)o914UI?cjz;6(wKcFM-^%|0yGay#urt8vNX#2OvFkz*^C?}t4k_D9
zL%o=O6$FZXMGf+e+5Z55_M_CINg2IrFFE#7q{}DY2Aj9##WW$~A1-;vr8S#tW|!{$
z-f676szfs;g<{(%+fpip+wD|}ELqHL?qHNE$g#FeZ_&QBdsMnh>+6e%p@z;mq~G>?
zb;2*Oti@oXJ*h{_(9kOvE%vm;zIgPhE(vW;(p4J|nS)J+77T>&r>Ql1;?<$G(<gP>
z9&$e_ksh1yp2}bNOyPKxnPO#S9^r+4y8Jx8)NgbfDKDhNf)xcp<2Cs;@IK*}$6`I*
zVk9y31Rv>NroRgQDrnal97}OJLnLT7sLNv?O7*ZYQ*w^%`2HfRT9jzYF3{Hgy$-9!
zo)5EW8&!oi+~oO(B<@^bXWqGsJ1x>>Kq9(d6?{U{ZfqmBvA9{Lo2r0E<IQ4eQh6|f
zNF>*Xn^TNxt0U|<TD5AJHtIX8=vn^H37ikT(xYoIdJ|ecc$01iN?76ZI&t-{39iT4
zruvy}Y&PVN%C5r_V><%<Dt5Fi-LX%K0*iyi4n2#LMQJTVecTF?IR0;Hwy_qbs%>ID
zDcM|tHHnw7q-FVt996~3R1FzY8?yARQX!G;rz~N+iYSgV>sMhWhWToGih9NseX3SX
zxZ|xJOKrzWP9e>qNv%V5so}L%?N(ORo6@9?KWb{4key95D}97FVOFiQ7LyzM)w@`?
zHVEdrTSXr)9V*op*uA?)HEp68*-lOcbT&F~mmWzO_N$Kys^kiEEEVydD|pgT6stKJ
z8h&A2tJ1e1lZ@6SrEFXBxUC|ran-bubgE4mM_tNB2tUL<4SB!V<nZ^19@T*sjyFKw
zzwG^M(h{<KO{0xpPdj*?Cf^f!H~3XG)aRuxSy}AJZKM()@0zkXtW6&8%(qn`go<!f
z^%b<Hn%HxXD<fpxQYvrgDgIqqf)PxE&Hgd?>cin5gl>FUsMy@z*jfOxC8DWexZzJM
zIM2O&f8sy*C^o&~d&6nrohA(q%#a1kpEhHkuL8e7d`;l}JI4MO)ie!K1-R3-8(A%9
zj~Mw}VB_)R9@Y58u6R24;lGZxuMS@BH+o&%wygO709ml$H_-4$u@&2frx$kB+2!La
zRB0saj{gA0elFEKL3o;$seLu|%l+n>LVx<-rEs4Z^v~@L4lA_{pO>)XJ#a;CUSj7=
zP<d|HtsDI<b&mncjCV3d-F-!R4^((jlTx|!HlHV!q$t5(;vIh~=zKqL-e+7^O~s+L
znqauXoEn!&pUHF2PfErt>OCSjeCu8XU~BNmwnui-1$15?ypHD7fOlfEyg?XfSwP^{
zrWR)xq|AQKe@dP*S##7?D^zG(VIs7&_45}Pp@PwsrTJBdt#sZM!*dw(4Rbd{#&F7O
zbHTR;-ufumg^#%v&h%$ek=Ea7rgq8cT&BBg_I4$(aoE?XNotR591Q2BasD#V$&bz3
zoOQ)gWE>kt8a_p3Tv~ZjZX62NX4->16WW^g{{U)-u{D`V*q6iBrt)QF!VzAP1Z@?I
z=eJB(G2uNg$&3~ppGxgo9hNVZHPuc&>4iz%*%A0pR8w)a*!9LM(oss=t~mr}>0S`G
z-`V<;I4ywL;=Q-Rx|Djho}V^Yr!^rGOFJkoZk1gD+<Tsaxu1xi1+9ERXDrH_q;byQ
zTI&i6Yvk%P*1H>4^PvF!(@^4#<0lq=eR%WXj-%n5hgOZ6H0axMMS0ATvawbhNcFGM
z9~*cE*Tgn61XT>(PBUK^d{g*WbK!gGT_gR__pV7tl<JJGwLG>q8*x#zELaXXrvS?3
z3a}t^{uOOw-5GDnvotfL{oh~Gn+svHQ(R%XRe0zE*KE+dQY&s;QqTh><Yt*4-NjDO
zvnNqdecrSIG;z_5y=l>~sWyh`Kn+<j*NT=heQFkr{{R>1QLC`)Ko>UB?G+G{bxNmo
z0~J?yx7M4to!`Y$D(>q^wt)<%B|mBY>7-RExu9Y$7N*=b=Bwx0rpL*O21tw~>BrWU
zS(nnI+ydlOnV>_Jiqx4k3|{o*V$=%PED2csJt}qiR~1!w=e<g;xpFzC7O{2)HL+zN
zTyS};Yj2viEm?kFY7w38box)sz#jG1>1*YLZicxX9oNiWdRI|nFnJM4;<_WTndsVW
z)3vJe(EgR}U$b76n#Yci$h(zP9(rWgglTgD2tsg8eP8<<_*d;(wZ@=9Asj9~lvR>Z
z9Bp)ctE1^LMQ0iTj4lOUTLfCUZwHqOjC83yy7P|Pls?2Mzh#f#@~s~(lgS>GfY19h
ze*XZK9!RVHSoEdxMMzxE`#-%m?Eb>E{E<hJJt(k@-?RJDeVlrl((=cpH_4@8nf^fZ
zq-d3UR)>};jSlKXx7eU#3rJN~t+ZRJw2;yk`;oC!>S_sRjeF8f3pG)=v6I%Qk=Htk
zNBmWTsCbq*bB0lhlJiWN`d2@9;hW2i*>1knbGVIK`h?y-j^C@|nGeMH_h8H`oxkuc
z^xgjeAb-NS+mDBOvbI%U6|`&F5hyN)tl9ihjq&@@&1+lyNVbz0nNDlL8^PWnw$8}p
z_NOhbw{N$6Eco;ouDIf3ZBd)1`=3Tb<EwY+ApI)bek`|>q{Hzw^K5#CrEmA(n0^&%
z9}!%`_x!>5S43-5kq$DBr_!+N(Z5MY=T>9Y1fO-s=U+M7_`2bL<*}`6UmaX0Nxxd#
z_j1{hsrW^29_7tz+p?@XPfDj}3bNv-W&v<G`q!L|g6Y0oX0t`dm(*_h)tjjl^d_pZ
zgPe0ol!+txn>BVe&eA?*6-p2Fnf@BkRVFd=75Y$yHhDY$01a1-o7$wCmB8ksjkj+9
z0B5ZrCmCX-bri?rr5<w0)D`~#cA}AuP?2S1jcxZ6``_-=GD)7GQ#YQwNBiH>m{_+;
zS1Q&OqyjbC=ZbVJLaFqk+=g5g8L1&%&FDSpkWQ=@y+<Q_xTV@RW15YaH2^v`Df1t?
zI#c}PyQtk!{{S94R7{~;IiO_HH_e)YM>9#DhxDg5S#$b#rpC*+-hq-U6>N}cn;X)n
z-R+u*hSno-=dDPrOK#p_0P&8MERy`Yuqp`t_E(x~e98T3`Uwin(>>^ob2b3T^rkZg
zRc}gsGkJ3>4yJ%7R?pu3DU%Z2Ge+!n7^tR^ru*MF=|~9w0Og)@%?&57KT2=Qw;bk{
zD9gUmMFX(rZM5Cq)a74VQn=$8Uup(AJZ|a5D>l@}dX_lpvH|H-B0a@dBc(HT?c0GF
zr#H;Frfx0wu}ET;E;1>tmP~i_rC*hQtsVD&6$}PC?deWjwKbKPXEh?O+G1nA@7<)N
z-AhPA3u1-K98+GNzbur(<HsqtwJNJI#%NU|sKq)zy_y(Yv2Q66DcsCQW74-T+sx8*
z=*#C4F&nF&hdc3tMt{k!gHY3D{{Vz*PrnM@NR^bgrB1~~H?iFOC!6fPF|}_n<%18G
z>5BaU{hRbVoonqf+v)cY^JC9QPS*VE;7@`66V*Ij;_F!Sm<-0*T6x^?sT!#ypUS^h
zKV!W~Kj9IU#e(_K%*3c{Hq+4ct?>AVhpOcc##qb^Ja!Y3rpc6iZ=s7FV!}BsyqlI@
z0YJYftt%S=_H-kW?Oo@_%|0usG<jDBb&;dmKe~PEo6_z4t93#$1$nso(!^BeX!i1~
zJTRDLP8|XpK_5z;W7^miabeq0vh8fwoTGEs<{9$(Q;cE0_vua%=9g)}KDAt;atPxT
zo=Dusj%n}5N`rZ=IZ>U)NB6N(G<%m6$o}uWSXC`lqA5z{Pb#FhfUn->rP=dhr&#vY
za*=Wtj58?otAAz^brjWz$k|bmT9I4sT;ml{h@~Q`+HG9oogJ9;6{F;hxTzK$wW>x+
zRw+kmwX!Qx2#bB*KD3d=?N$+$I#VUco6FxH6#~JwPHI@&_<Pi>yH4)ZT&;^%(Ut33
zw%;vh!lxOlaA!EpA|9rSwkg}X9{uVy`G#qCRh-8wtwH^fswKh<LvVS=XRQl4gq`Zg
zi7d_ji!uT8#yr(N<E3HfmmX{&h6J2o{c3h+L|;5E`W+Zu)n+a#oK3i9rjAY3Mh3OB
z1+U9L+mhZJ5889Y4;Iz?MZKi&kNN0g3;B?We;WNfxYVH1>~1eE<y(7Oc-m=Ykl|5O
zWRh#~@8egGo8oW9?-F>EC&|9MK2IGXLAE@9z(D+KzYBd?<YM~}>Fy*5zr*icUXdS{
zV{r*m(Yld=?_4#Ej3H71$gZP7)HiHakn`5N6Fg(U{7qu^I`z~WcBbstIcFhT$yA&H
zUZ3$|+(WIjssIi%#d)!XeBj5cR(d0*I~|XMZ_7N0#bd4QaxwN^IcoEbA5?og_|867
z{`GnkbBpU=GdVu>Nr>UEqgeiN!&UtfOvA{@sdYU!Y=L^_tJq@FZJDQZLx}xJ1IvSh
z^s6>Dth;f>D>G2I@~${F)c9Re_~EN57IwM@rYE;powDP#Zr^D?Yq67Z1IB)}#a&sx
zk2F^iw;jN*PVkNNTE?27kj$w>M$ZKCMv`pqA85fHJu4?n3|&Cz-o2~Ve0kxvS((DF
z59wSTmX-Eh?CHmRQnLu^j*CN&Zqf+aD{kWGt($@|Svov;Z}w{Rt{yj5KQEx`U6Il!
zBlwOJPz=oEax1F%f8yh3brd183W538KYeP}?Jy%7zU^WOzQcAU9FRyIzLn0I(PwQB
zsBV#O;aiA)Vm&KoLA)(!p}zR9IQVzsf3|7j86T6<yNekFcRVn~U56Ka8PrXNjDdg~
z(zxFl{6DqvHlH#}6l&+b%-3NCCXC}J(vgnh6tVZL9of*gRz42+rSL}g!?%w-5V3`h
z<Fxad@~4fh)W;%~e@|NdJJdWGVd6wE-9(HZughOJ{C54CE$rf(Qqh;naJ!6qhs|7)
zv>vBiZzOz!GM}5;p;g=5R)(WxJ(Ti8G-V_Nf)BM&W!kyHG=@nXSQ}caD{ZGMZSO#f
z8;6%sG+U{$<zxAdYELjw1#QJT)X_=1zA0pkRdTy%0`d^uY9?L(0CZ#PP)9etPaKp1
z8ZjL~6!CJ&)}8*dtudkv*w8Y}GX`Atr@%?-YL4O;092l79Wy|S;fsF}p?+k=SB_!p
zQ%2%}i?`*?KW#>=88P0aQ@Vf{-zs9AlS4*ZxT9?_6>_yCbjo~mgg9i>>V8~(-}h=2
z;-$yE09lE>IISyL4(w*KE%WPFt=LB=IX_Aaj<-o|m;~jGb~<9KDz<tI*AJ$*%%pRQ
z>MU-Gw_Y>T+P90)lhSl}{?)XFVg-JI{>6G1li=mK+IP(<B%f;h`|w|xv@S^m{{WL;
zsQ&=6?}yHl@b5_8r5WQ{58^XfR(d2le;Ms+s;i1Tx|e9A+g#RXNfyyb#ZAYhCznNs
zat*ZKEY+Vbo5e0ska=dL;-uS5+f0$0X}^U@OSDoQ$Ts^?x7w%LDB3AlWq&MsQySbi
zYPYmev{7IvwU}<y-?5?UwV||Ae7~g@13UXZHtM3IZ7mzMYHg!!O@PS$$x+coGy4uE
z>Wlf-)(^ECYOH|gTS1Asr9$2tfxqRF{#DcM`c!*;g)0Hb&EYF%>jF2_RT;bqXFuMF
ze+91Tls9TmBzjOjW6h@cYhwQZ^6nS?+M{vsrp``^$@H&PXjQwKYg?;#p<QfCcijGL
z;5&K`!kW$Ib*6cZ8)bU>_Ndvm*I_mjYM!Q<_i?wkN`D@;8n*5_4|+^wwyIyJ6?tST
zq-68fsavoeD#r1Q(gG%L?w`(?8wAdK^`W+^=ApxS&;^D0Mrj-7T+@|HCrXQ*xbHv@
z^A63uv%B%8r6%FGH7em~0!^WMP^=r;qusaennp+3N{*FaT$Kz<$prgVW%<3m=|sf+
z?3x*SD-E?QNNeQ>Jhev5T9=a-KYE$Ekghm3<?Hi)lq5d%)ljuGZJGv1tTB<m_ohB~
z^ds7z9>!?H0uXhnFO9NEw%o{5)xgj5pa>vVPKKo8Z<pGVI!(iVYJm|q%HGtGsM>vd
zRE)~WyBt$hcF;b9o?_<*k?JX)V55bH3V%Md!ObQBIj3^GgUvA-q>J8=09J1F_*s|b
z$II(N@kp{2<FzydyS8=an9?&-BVyfYxI1%AfG%0F(wnsU)19!{6y;sZnm{M;?d?tx
zvmtG~3Tph_>0@@_ZaD8i0<Q0qBXuJ<T<~e?p;Nwpz21?OGK1;Tn3$VstCs0b-RsBJ
zn;*;8nInGx08T!%_`lY3Q9Etv)7GVr?`JdtI<c3j0+D}q*gu^tarvAeIrOHp25vc8
z0I>{$IXM3Sckr)q@VACz@dv>DOsL;1JLN?^whY&WMYW_IvtL$x7qk0+!JTCY2Xt2|
z7og7HTF!eNcx8FD<aU3tmx#PS;r{>w_=er|_icWE+?%Dx`nEte=RaEd%l1uM6!A8!
zvdDr~fdp~36dzo3^skCPWxF`rL)D8e33%{70y_Q`^vA)QC};SAAMc$VVj*@h1&(k9
zRVe#P$_CP2%LyqTY<T%}{{Z+zZ;XOAb&qmmmQ(FsWv0xtLA^*T$Ln3x7KoZ;hUAYa
zSCc8@AB}mYjdU(N!x##NL$!NjwO%t0skG6SUx@aJqkWh!y)*$QL65CE8*?DdS@L%D
zuP2oE6qSlt+t!g=cGd1G!^g&IHBGIWRv5<Sqh;Gud1jT9bu&I@LmYK#IP+05o%=^x
zvlkU{C9%Fw%hQ^ZZ?#UUAE2j0Zg*hS6L5uneqZ-$(#m|k1w|Ar)p8hwi-G#n6J%<4
zHZxY2=BU;Xt(tVo(bb|rPFFeYRfW+f88r>WrJ!QG6H(c#NhM_|a-DIdrP;X|B{S63
zr*Y+GondZWa78WExbh{(;Z%&Y=G47z;KmQFO9Y?1$9kn=+bu&#LWLPQs5Cp%eQ_+A
z%6Xi9D!hs1z1li>+vUYAH5ZK2s*W;1yOEawh8X6&`$4vz!g+1w81m&IdCo<89Ci7p
z{{RIU{iJ>%d@J~&4~4uf;u$n;Hf=Hs**?mn&N4ibgY%J`e80-7H2LMYy|SYhb-DT{
z2`ZR9X_wm7hJ#}*=B0TJg{)tGYljGf@iq8ir~cW086dYd$?*qRF%R#`tbaUK2DR~f
z#lAjPJ|XcIp{s-ZH&-s|eyUC@tx<~VaL*Ll_fhwU?TPzvSa{mj<G`K<yYqC1J2fp;
za-!vP%xV6?^*BD&@)WnP=0`U6V%X;xz#Xw(U1z9yQ{cNCoPth&I_|8l%+{fc0tZUo
z=_@14o0E5u-RSoobU5CB8n<bD*7woJyA{i5_l0BN)2{VTF8vFVr1hqy+~;vRUyB-O
zxYFRdOcE25UNd#MRTyy9@AE_bz2TQYH{oyr>s~9aY42>$8zBt8rA_W>9gcQc!zgZy
zI#)~KZF{0iCKT5pd%8I4D?6sPD9eI9=(x{Xxbm)59Z%<2H<si!<v91FL)E<KQ*m7S
zRhy|3ZoqOqYNEo|F||00jldec;Z0er<mIvTs3%>~=PXa9WXP=(WF53fe5bwX5Zm~Q
z#YAqd<F!$=xV!LHs1<`OC9pn~=Uxi&!{|52r*Ky3j+O2Hb&rfR$m9--FdTO|s*iGd
z>TSymJ|xrp=)jP-Q(kYc=x=8sU9E;V1P&{6MbvJ*J$dCzmw@#htFY9*!>d@ALan%R
zJE``oTJA%-Jl4k4q+G7k?^v3}#FkPtfZ!h0-Y%Mv$JC##avmbF7YIP)j-XXa>LxMx
zj60i#xP-_!$EP*T-$V;=4sp}IYj)m28Z<5PgX>XR>8*DrzyUs|t#e+-Rf}E~*B?!I
z!te)5_5T14_>nG@JFw=wHula<E=1lj^sTK2T$<ltKb35@Gk)v2^wx`g`<=t5y=wW3
zyE{kHxL*YLg)gndZp+Pj4ZX0FECFNf-jTd^F||!rT^$xw>IQhPHt~nXt82*S)26ly
za0V7!{{Z!?(R@v%-rcBJZ6=Tv$I6DhZ(R5@;_Z6z-qXXkvk}pjFc0cQc-Xl;k?7EI
zvbo?sI`F(&ypqonA2LQHkU8eOmr~JYv5<nS6nC$;v_IKD;%>acGoYI!KgV+_kL8-)
zll}?E<9`xHKGiOpYCGgJe?eK&tEQ#R!*tKjhFz<Y4nBsU=DwKt-TMuENBBz#z5dSr
z+0ztd?rftcXVq)x2`%J?MoA-NaBx+I59QnOuDHoaz=Ltyih8hWV=FE`Q-j|In;v~d
zP|HcrT5prTp%lepwB?g+0wQ}>10s`gPPKH%9zH4=BNPD~GH&mRnPn|i{JeCcHZ%cX
zWxXi!&fManRx4Ba-!3Qu6<xITl&`Hw6f)+c#(E6U1c@-I<C3YYdwUv@Nm>Srp|?|&
zgH=_NZ+f0OcPN)0^yT?#5ZkI%`KT3#j%mupy{VC~KHpkaZ%Raxs~_RwuiCcd=dDs$
z%C2hlv#QW8RCGEVfm<2;>!*iq;Ocob#aaBVl{l`aPq^C}bv2@PB1rb{gEp<KYZexb
zw=%NCp!dnee!hG+(c{uQGi3sFiZB=O7_ZBJ*x$pS@RR&X((;?*w*C3^;BjBA8cn#j
z)8bGug~ewkq{mFV<(fRRPokjMk0a8Oo#mQs02G@=EffIVw7W$u6k6a%ww=WrXxmGb
z0L`>*r|qL{OI?6w+BVavcG9x~Iw{^+q}xT94ckWAcG@=4W&?K8Q})vBqRa+usJydN
zQt?HY5pAYboDW)$ifMNOFDUy_<sWL3bsMTc{$?bYbnQs3#!W)ax`xWGGhG5E-QPUY
zFU|LjGjit`H2CBH0A$cIRbl0E)~53Z-Q=3B8x6Rpd4X5u>+e7hZ<%pS%|A5{m^$F*
zffFljkCrr>shB%)OGrpD@?xdkQLoKQxNXe{E<b3ehT5zFZfTpT18(eUu2AjA8Ka7B
z@0X#?F6uxg?~<dX-STuaigz4Th>=%yIxZ+xD^kV>rBNeS`>oU3o8*w$&*4wo#(KRs
zp1;LF!E(I!qKYZpXm9ZK`Dr<&WmQ~fnoZOUk8jJ@=BF+PdR?@hU%eZE_~WHj+fIxe
z1JgJIk=~kHYhch#$N6TNwx!?fYCquJ+L=Jnp49F(;+xN?{OUF(XaHp`NEoNeLCrpt
zHss6o_Mir;g{Q|E^%R?ERx3b--SXm@&hPeVXr>_JILJR*i*fDG(v811Y1tlyev}9o
z2ac;wjYTnkchFRe^7H&u0WKSEcQX&~bfjP1^r_jlt~&d6sF{_U7@!B*?DG!rPo^vD
z{{Y$1R&Ry6kj7UMqWwzsuaKd_wrlJU*>#p*0p8mY@~rQJ>{p7Jd?!{7{`XS9?5}h#
zG<{as5X`1F1a-$h<6lhtC-7@)o+=k6d7<(^!$jo{NzZEeZ}xPy8ZX36e??}@arv<f
z@%LW>zQXud;%GcS;$al+vKxj~Q}3KtP8QL`rz1RbsPHn4J^uhh=(}GE#dW8>wDR0s
zBd*30N>3HZL7<-w>XO*0W{FC&vvt9*ThRhdZXIyIjPYDXt#q1>wJf0s^4l24JRDbM
zK1xxs;^h&XCwH}n+T(69Q??qC);gF+y1sP#G|c;ue7-6l;m#{d$jab~i|qdZH?>bG
zr#W9|%bHKluUo;raB5Go`%~q)9L8#{?m*_Jg*WscomyL6rv++Ddvcvts=>^H3wHFY
zF~m1#6*MTz=~Cfi>sMkHx*>}Q>?&oqRq0QKH+1XiQ|+llU6nkOw>0S#70DH8qub9_
z_N9(BWah5S(KWf0_Goy=6+WG&;y?)G)YlV1r$T1CksyE###s*41Jsk#@vjN^f%|!U
zCHQr8))4Cd0Pv9Ct8Ht2CO<I~>n1|}Myfb@OH_hbn88`I-tM&%qOqEFVQvmSPzkK9
zN5#;&3l*zMJMQE9*PZxJ<JW?|D#n+(e~7JOPzEh7q!BU`>KNj?8+mk#t*Z{9X!18j
zNz|X`QS9X%EHzS;l9w@lYwPV<1-c^J+mrK1obg>2ho-HKx~hK&{c9gjjr9Cu_mLlP
zMQ>T%$0DdwLmtFbrLD`UQt{=E#;U&6#YnCisrlb2VsHmq!|@l6^bZPIwbrq%*lBjj
z_oTYGL{a#YS}3=&INOTqb?~rkVEc6Z>+sw5{P>$6#BbV@#n%?oEcO-+9+_<Dav5M4
zVm-XV*!DH{kL^kS00kV-w0{ih9vt{*b^WcaT_Zx;!h#{UR_e)~OMVskU#whxw|vqw
zJd8=)-EvPPpHtNP*IXmX8@<matzT&_Rk4z+(6bdh8Y9#m2H^2gNZ$2QNdot%xg5RD
zrrzLOg<wuacb*^APm&e(;MbFBw}ux3B!gZ4lXd%)VYCXaVjOIa^2*_3d~%|x>bFc~
zR|h!$^+jtj4WGO{>W-y1&5ES^nDU;-;LUYgNdbumzolUK%SaO>*Ae{Ny6vui!+MV0
zH|2aEPW7#(c#rKcJTH<nkD7?fLhab|TdP9fDPNkRZ7E@hmLE)4W2sr9%o#u=`c&5X
zKieTAZb<AZu1ktX6tT@{i!@o-25VzY)c(~8Qdc9WJ!`VF((TsOt)`7e<-*sVYnn9v
zAHI1b8Ht8TT;h|u*ikmi9r;`jtyR=5mDe?$Ype@}A)5ep?^ofNPMrQ`tC<&LAIsXn
zX9RRL(0F&o7djm4vS!#SjP`2HxjK5)QEoc8P`|BFHa)XR@#Uw8BiSJaPi%jUYgybK
zJ%ADy2e38aIzqjj*O>|0zgp~I@ifuMft&HIBP*jBM?>2zEsmvXHVw>w&MS!VHkW_l
z>yI+(23u)A<*qWdtE2d+*y-tT9|Ifky>nX9d{ol4ofcb5qZ<vFXygErYf0>38yH>^
z_;CI`kPSE|noa`b^ZHXfN$}5D@Ghrpx3;em$Z#96H5bF5iZ<U4yhyXFXHAi^S3IxM
zy|dyU#tje0ULQn_k>m}%9OEi~I_KrN(50cux4?GsSBF-3<UVj2=DaJ#o)^9FPNZ$_
zqjrz@ByIt(r2GT%ZapsDA(fN>HkBr!@xF&+@iNv`fpXu%RAjHxwApA&w2u?;Z-{)B
z8)O^Tu;#v}_$%Vu4MO4wF9J!<Fxan_HE)M{?|^1$F0AG@*vXMs8-BQ{v=1Cfq~1p?
zGZ>^-<O)hxD_9?_UK{ZCk9}&AM{RDy*J!SqKMC0;(y=Cc@n0DH0{x?w;%KaIXHmFu
zl7A}uGeg&-)U4Kc<lGo!8s?ooSE12Lv}d8IZ=u>lw`@u=+XAjFgeu&oN6>$TG8xuP
zVS&%RSytWK6_+W}X6im2vee;dr?-w&Wx))!@<;6f{{RH%@DIgrv}zW1r$p7I<XqUU
z(YxJyiu)yGR_{$RhTygbwI-Ce5%^>AhyDr2@i*bbw>Mg4w0<SiBk3i!R#p0`{{R~K
zgIe&$sp0#hYpChhRu=%{%eaY<{-VF8JF7curC{m^J&rm0@m_i2?|_~R@vMy|tD)Fk
z$Xh$5fl2)FS=w>g+?zCiIw4CO07>+uZ<J(^KBNl$#PP5E6Pw{jiW+$agAI(+e;v3f
zx9Bro1>-;XBz}puNpwFD-`eBu#V`^t=ZX}i)eoupaI$SVImR=L8kje)KT7(O$Di;{
zKZ(8(yqPSl<JRC)@|cKHN8&|z&x*bP{6X+Lk*fGw{@DlowDR{a=~{CoxN{?d^8S?;
z(BF58l4X+GPcBq=-#FUJL8hSKbmyldns>OyZKmz156ed3=|BpsRdG&PJh6~PGNaq`
zdr$>{<-IA}PBT!dm8m@Aw;kvj9#kzoWZO`*p!(DIP&P=cR;P`sw(4@RS_N`+V&8hF
z!2a;6@$}6|vKIakK#OX{qp9E~s>Z`-TDfk@&58ilvdf$b>2$xms^4^eb<Jt{R>@Oc
z?}2<hd*bbS%TCj+-sa}vwyG#UpVFQ5yCz58{{Zk#Z-?_q{h_bT5snuk-X<fc9Z2@C
z(AZOIs`eG(zq4<GW%zgS{{TtTUJ!;TM(hr!y?Z+9O1+w}jIXOIG37>%J?Sc<>Jhhk
zl`i9xqhsq;bXhes%QUMhH}MKw(*eF&rqr~!T3|=znmn^kW()62NCtUkk1W%-QM!N|
z<(h3j%oQH$0Cq>E8)>_#HyQw4)SGE3IiLliiYNg$QM#K&E1`uZ>RN3L*g=y;6jE=n
z5N<SXH7?p;FX>Ze2lF9JRwJ)JN^+glyf59YwE;hiH7c_&@fwbocVF|GU_DzDD{iI+
z%?iO+GzpS<l(`<27V2H6)`6A-2B!IfqE_49k1%=~1U+QhaZBb6nC7DY0G61g+dx>g
zpO-X<$8KsLGv&o9G2QJz$$Z0|-CB*qZfZ4Q{6DQK{L#<?d53N>PgN?}q+B&RL*=$U
z_tJq5SsA-i)<#;1<SWvh<_ZZPxUb!&WYT$q%f~yZH&bN~T(u@mD*fyZew3K`b4W;W
zP~#xecAccr9^JU+mvHjOw@OH%X?FbB$fTx!7HYuhJRNVKNquy6*=}QjZJ5ROcL;pJ
zj-=<%)(p(AJezn!(=^49tR+8);FFB=-jRZ;`_nV!v$*J{{`EDoqZ#6wxE<;3^L>3P
zQD|W1pSU+bX{x10Y3qfih9_*$zcS*F;h}z70ChH%TXiKdFmINMMs3sv%><30F*Lhq
zRUxrJ0`6g+Dbz2@xTM|2Xadr=twpr%&svj@T4T4E*gr4(v;ib+E9!6AUgA&eolSQ(
z#hN{<+mVX-V%|29f~w@2`g8V9+a<oM2J?~SwmzBX^`Wf|u)oDc?*v^fkL?Amplzl}
zq{qHc9DZiLiPCjQw4V~_*B1Z_WV?%TJdAGS*US3Koj3N7kE|&3!!cpN`H%uXO8SRL
zv4d39ZPkbF)yC!Q268H)9%_*~E7bNlYSXp(Q_%bO!&-)#beN_~h(v83K@Ka-(r3EX
z@0_WPT4pQmxy>(wG+2JaZGgWj^e4B{x@}WPTX^A<X4VcAanyIOOBIJvi`MM;T&Fm#
zimeE^S}cmvKbJK+!d0=}ooIogQV!)kz%=QkQ6a&`J!|LMvHDCS+;5O6zh?SW<MO4!
z#!mz4Doe3hR4+kR6b7kQ5Z$QqL}UUO6YOcC8o3JE?ri3t_HBoj&$z2m$GBv3^rexK
z-WHN~Sm05UBQVG3T18xkexgLG)n8E4P2_l43OBV-z43R4bhuhut$$3liGK6Tc^ZNF
zn&b74+H>JQhV9;ML*f>J8ZZ3wyjc(AxizDwF2t&li@ncC(<~a!9kMZJ^(L=*CmF|o
zPM-DgXT;z5DtE)53C8yR7Sr#%TXp<0pY&^Q#gEPZ0FKwopAx^|sb3U4R}7Zg9C~ks
zZQJGFWdV8d{{USapXF5=4%+B*%Mlysu70BYM)<wpzk~K`b>i<3*=W(X+>Uqs>-`XX
z$MUa>zBPZrQnXz+&awD=p&d$PU4v8AGaMlLta;0RIIqp06nN`h@#e8EtFP*pn!ceg
z-?_Y)$sznt^u=Yy<+4UeBc=(#_Q|c{;ZIbq&N<boC2f)FKM_A@PaphNzPY*aABgX-
z{PUA<=dxR%{_4b}E<Ux)=#n|OURdPh4cH%$9M?S$nIixc^IcAdX#0~ZwENQbjlIrh
zoveCRnIUUfq&&N4C!C*5(>2X!P_q$B4cyT>cS#w4tw*Ahv;+4?K~eaU7n<KV&(gH#
zeaxIu)c9NWhxmixMn=>8SE@wo({;l#{{Y`4n(y?_+w0>MikRP9wz?gjYe>oe0MKja
zR99h^Q&O$FgSd<e&)Q#eMJX+r>E95)Y(I-$D2d{m##gq`w^;1RfPdqFSB~Cz!&LDU
z#!X95xxJMA=D3}lx7RtP)-1f=x(#ycy58fQflJy;xbJ<-mr`an000L;j-O6z21yv5
zq#E2=7D!n2%|~Z#9mJS*sB*X*(?G>{cB(T#?AY4X>b9N~j&%&FxSrjq{{Zl>nnv6r
zY<(*SQ6qI1Bg-IGjg*nny8RyD`DzIu3g4UI{F`zi#b-~Yn-$td(@jlFV3Zo%=`@?;
zJUapAn|R;6`Ktp-xO;PeYh@VSKg<nUmWMlux#8=8w)+bHHBNm&?DYoPPQp6-)vG&q
zjz5%V=|tK~%{K6a`*x|haYdVp;?b)P-nFHp>k>sMbsu~0X0yJ@wY;|kEn0;IsTm)8
zt}8_@#7ZXZ^EJXOtMZXoBa=<m^j5dH+4hLn12xXesYt802Nlm;O*Ar@BFN8QTEA$F
z)NQGa;td>Ww~S#wDfH=0()9+Ad2L!IM${K%-F<5o>rXawZjFv=O+8JiC6?ckFg-hc
zDZ|USTxPP9Tni7CIoxx*KJ{HLj8C~6D^#W&w>n5{-WSMYk51Ki$s;N&DEWF~u(Z8+
zfr$?F_pSR|=%ieZYgHpB9_LLCr?*Qn;OA}tt`2D~ZEqwYn1<+0bui{g!C#o4N^@u_
zeHhwu=7~s3DB<pJB$)i`mhRoE?vdgPiy0Mp5sulezCVYO+$k}rVm;_zg*Venffhd+
z!`cooeG4{zDvISm-L^tL_q}~5`#$S7o@#xe)DzcnubplF5NgV)u^6}3yI%_UGsIdP
zs?e;24|7!-py~EL;qf+)p?Hfzk|`sNw<oaOUl;sS_#1tr-Az80H^LhtH9UTG=~@P{
z;&|Z!$z=<G8I5#PUtC?<l(|JD_4#vKDaJ&&$@D%+(sk`m!#AnsiJ+P@xHfy&)gKH0
z0Bg&wD%Klo!oVuvOSqHyR|DfO4B6^}Mi%S1^=x3*i|X1f=9hX+v@Qo_-bnsby`)pz
zN`reVGxVFnzqMACsO>XMk>lV00Iyz^rTlHuH4nLU8-Kv7@?%Za^=LU%bU#Y(d@th9
z5kWB8(yE$%+L^3rR$gc7L_ZQOB*xj4=cgj2Umn{x{InmZ_}9gr3Gqg+BvJjKzTwlF
z?^nc=tU2Cu^{D5AEkb=yra$32w*LU0i~3Rj015TC{PZ8Ce8hZHI{yF}{OY8>Ex7FL
zKN=LN^&5=Ot4EKl+!AJ9K9sNHdkHrKB7lAC<S9I3c8b6=^I7*EIlo?#H|I~9pK@(e
zq4pfUA=0jg%+av_0BZ)YHOns#YZmPs=>)+401JRce2I7QCs&1Vj>qw=slF`hQGCKi
z&(@34>cf_-x*nVHTlQM`b>n4rxY6yr!Z>)qR?pN|#eWumVjEo=Lq3Ie%Mts~xENoh
zdzH_|?P?XlcQwWMgW~q9c^j<J`MbJ}zx{e#)H*VKole$1Bk?!GJ!8Um6D7n2&~z$E
zu28EhV5<uHW5*hm#<-<~1sL@N99N5K9t*#jw=4_=bIMR#Gp#~))f|7#Mn!M8gY^Ym
z`AmM5S-dfRxc>k~bDv6{X!jo^yBWOWG!AP)v>WtX8)y8fi=$qSw$cx%r_B*^zQrn+
zBh*yC*>_mal*8~e#?x+9vE)hlP^6gb!!*8NrE43d{{WUR^}(fUNtJS9f6ATFHe{<J
zs^W!_maCX#Ua`3RDnT2ZaCxcbUBb+_8vgFJEn{809JVu5EuflwDnla^=(Xt|4gMHw
zUNh8ft*z~5iCy;Z*lBx6ts4s!yer}R&k$ZiX=entR}&m0aV|-()34Z{_E3|+_DSL$
ze%${6YECySF2CxR+m5x<{>q=S&xU>-={EP~$!66gAH1Gr$<MGoE9;93NUb9a2moUo
z)T_@|ur%+e=xov$l>it629xZmR%zQsRglc))5l{=_5!^PXl)cx%7$h&w7rcOxc>RA
zeZNYNX!NDZ24~u?eJNwqZddOjn$q&q`?TLFeJNRiPy11q?<G6?Sv^Rp+SwbmL-q*U
z?<o8z_l7xj7@PR&Rk%j}G5OV2(_`$l6KQYTB9}CfXlVyl=xE#c);Y7CMq_-BttQ);
z*KmK#(aqfs#=`vFsH}gsUB~<-zug9&+Pvx32K<FSb<{?ye6(h>S6Uy$hw`VcsRMr=
zKDAuP5q*AYDhThOJtPLC6DxEyz$tdq{!`HAlBUQt8A-OC_|U^|dUm*mQBo>~?@Q&M
RN;yyw?WLvLM%rM1|JlZyBgy~(

literal 0
HcmV?d00001

diff --git a/index.Rmd b/index.Rmd
index b2e7ff0..7c139e5 100644
--- a/index.Rmd
+++ b/index.Rmd
@@ -95,6 +95,7 @@ document.body.classList.add("two-columns")
 * [Katarzyna Woźnica](the-team.html#katarzyna-woźnica), PhD student
 * [Vladimir Zaigrajew](the-team.html#vladimir-zaigrajew), PhD student
 * [Artur Żółkowski](the-team.html#artur-żółkowski), BSc student
+* [Paulina Kaczyńska](the-team.html#paulina-kaczyńska), MSc student
 
 
 #### Collaborators {-}
@@ -249,6 +250,15 @@ GitHub: [Hryniewska](https://github.com/Hryniewska)
 LinkedIn: [weronikahryniewska](https://www.linkedin.com/in/weronikahryniewska)
 
 
+### Paulina Kaczyńska{-}
+
+<img src="images/paulina_kaczyńska.jpg" >
+
+I am working towards a Master’s degree in Machine Learning at University of Warsaw. I am interested in Natural Language Processing and ML applications in social sciences
+
+GitHub: [Kaczyniec](https://github.com/Kaczyniec)
+
+
 ### Piotr Komorowski{-}
 
 <img src="images/pkomorowski.jpg" >

From bf80867fa953fe701c382f5b1893f0695bb447a7 Mon Sep 17 00:00:00 2001
From: Mustafa CAVUS <45355486+mcavs@users.noreply.github.com>
Date: Fri, 13 Oct 2023 18:59:54 +0300
Subject: [PATCH 4/7] updating my description (#101)

* Add files via upload

* Update index.Rmd
---
 index.Rmd | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/index.Rmd b/index.Rmd
index 7c139e5..79ea267 100644
--- a/index.Rmd
+++ b/index.Rmd
@@ -182,7 +182,7 @@ Website: [hbaniecki.com](https://hbaniecki.com/)
 
 <img src="images/mcavus.jpg" >
 
-I work as an assistant professor at [Warsaw University of Technology](http://www.mini.pw.edu.pl/) and the [Eskisehir Technical University](https://www.eskisehir.edu.tr/). I joined the MI² DataLab as a post-doc researcher in 2021. I work on explainable artificial intelligence and AutoML.
+I work as an assistant professor at the [Eskisehir Technical University](https://www.eskisehir.edu.tr/). I joined the MI² DataLab as a post-doc researcher in 2021. I work on glocal explanations and imbalanced learning.
 
 Google Scholar: [I63d1WIAAAAJ&hl](https://scholar.google.com.tr/citations?user=I63d1WIAAAAJ&hl)
 

From da33bb5879e773254c671ee0b18793d1c0d7b557 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Maciej=20Chrab=C4=85szcz?=
 <39712646+maciejchrabaszcz@users.noreply.github.com>
Date: Fri, 13 Oct 2023 18:00:13 +0200
Subject: [PATCH 5/7] Updated maciej chrabaszcz description (#103)

---
 docs/the-team.html | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/docs/the-team.html b/docs/the-team.html
index f8193af..a9286f7 100644
--- a/docs/the-team.html
+++ b/docs/the-team.html
@@ -204,7 +204,7 @@ <h4>Members<a href="the-team.html#members" class="anchor-section" aria-label="An
 <li><a href="the-team.html#przemysław-biecek">Przemysław Biecek</a>, PhD, DSc (Team Leader)</li>
 <li><a href="the-team.html#hubert-baniecki">Hubert Baniecki</a>, PhD student</li>
 <li><a href="the-team.html#mustafa-cavus">Mustafa Cavus</a>, PhD</li>
-<li><a href="the-team.html#maciej-chrabąszcz">Maciej Chrabąszcz</a>, MSc student</li>
+<li><a href="the-team.html#maciej-chrabąszcz">Maciej Chrabąszcz</a>, PhD student</li>
 <li><a href="the-team.html#mateusz-grzyb">Mateusz Grzyb</a>, MSc student</li>
 <li><a href="the-team.html#stanisław-giziński">Stanisław Giziński</a>, MSc student</li>
 <li><a href="the-team.html#weronika-hryniewska-guzik">Weronika Hryniewska-Guzik</a>, PhD student</li>
@@ -317,7 +317,7 @@ <h3>Julian Sienkiewicz<a href="the-team.html#julian-sienkiewicz" class="anchor-s
 <div id="maciej-chrabąszcz" class="section level3 unnumbered hasAnchor">
 <h3>Maciej Chrabąszcz<a href="the-team.html#maciej-chrabąszcz" class="anchor-section" aria-label="Anchor link to header"></a></h3>
 <p><img src="images/maciej_chrabaszcz.jpg"></p>
-<p>Master’s student in mathematical statisctics at Warsaw University of Technology. Interested in deep learning on text and images, explainable and responsible AI.</p>
+<p>I am pursuing a PhD in Computer Science at Warsaw University of Technology, where I also obtained my MSc in Mathematical Statistics in 2023. My main research interests lie in the fields of responsible and explainable machine learning, with a focus on Red Teaming foundation models.</p>
 <p>GitHub: <a href="https://github.com/maciejchrabaszcz">maciejchrabaszcz</a></p>
 <p>LinkedIn: <a href="https://www.linkedin.com/in/maciej-chrabaszcz/">maciej-chrabaszcz</a></p>
 </div>

From b3829248386286742707f7eb026847e2217f5803 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Przemys=C5=82aw=20Biecek?=
 <pbiecek@users.noreply.github.com>
Date: Fri, 13 Oct 2023 18:11:23 +0200
Subject: [PATCH 6/7] Update index.Rmd

alphabetical order
---
 index.Rmd | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/index.Rmd b/index.Rmd
index 79ea267..ad83322 100644
--- a/index.Rmd
+++ b/index.Rmd
@@ -76,6 +76,7 @@ document.body.classList.add("two-columns")
 * [Mateusz Grzyb](the-team.html#mateusz-grzyb), MSc student
 * [Stanisław Giziński](the-team.html#stanisław-giziński), MSc student
 * [Weronika Hryniewska-Guzik](the-team.html#weronika-hryniewska-guzik), PhD student
+* [Paulina Kaczyńska](the-team.html#paulina-kaczyńska), MSc student
 * [Piotr Komorowski](the-team.html#piotr-komorowski), MSc student
 * [Mateusz Krzyziński](the-team.html#mateusz-krzyziński), MSc student
 * [Tymoteusz Kwieciński](the-team.html#tymoteusz-kwiecinski), BSc student
@@ -95,7 +96,6 @@ document.body.classList.add("two-columns")
 * [Katarzyna Woźnica](the-team.html#katarzyna-woźnica), PhD student
 * [Vladimir Zaigrajew](the-team.html#vladimir-zaigrajew), PhD student
 * [Artur Żółkowski](the-team.html#artur-żółkowski), BSc student
-* [Paulina Kaczyńska](the-team.html#paulina-kaczyńska), MSc student
 
 
 #### Collaborators {-}

From fb511e1baceb647fa637bfec2a9648f49742e045 Mon Sep 17 00:00:00 2001
From: runner <runner@Mac-1697212601104.local>
Date: Fri, 13 Oct 2023 16:14:47 +0000
Subject: [PATCH 7/7] Render book

---
 "docs/images/paulina_kaczy\305\204ska.jpg" | Bin 0 -> 168530 bytes
 docs/search_index.json                     |   2 +-
 docs/the-team.html                         |  13 ++++++++++---
 3 files changed, 11 insertions(+), 4 deletions(-)
 create mode 100644 "docs/images/paulina_kaczy\305\204ska.jpg"

diff --git "a/docs/images/paulina_kaczy\305\204ska.jpg" "b/docs/images/paulina_kaczy\305\204ska.jpg"
new file mode 100644
index 0000000000000000000000000000000000000000..3d4966642e837a0cef36147917561f68fdacc9fd
GIT binary patch
literal 168530
zcmb@t2Ut^Ex<4EQL_kE4UR97@q<2se5HKJup`-K`TBL=hsGw8<>C$@%NC}}MUAjPM
zfq?X0Lht_L%$%8X&dizT-tW5`*0XnZU2j?MddqM99RK+Pa8pfLRT+SXhX;6$`vd%(
z1SkUVuUz@{kJ|`v|Af~F2?+=YNr;H9Ub{|m{W>WLDJj_v%9~_2C~lCF-lVxnaf^zY
zn)*69EgcON9VHbt)vuS};p3hmAS5OvB&H%GC8PRZZ$G~RD6SC#2m$zbcL7%@@bD?{
zezpNv0RTKgTyKAM_#YeI6<i;$5?v!EA;sN*xCyv|hmU`S0RLC7ad-RSt^)`t2r2If
zJ-m8L>jlwWS1RGxQ5o0R9+fpyYY*<Qi<r9w5|hx-($O>A<KX1t<`ESWmync_R(!0a
ztnx%vO-EM`1lBh&w6L_Ywz0LdhrV?8@bvPAy?OgCDER${km#6Cv2pPUiAkAR**Up+
z`2~gL6_to8WOYq#OKV$uN9VV$?xA7S$mrPk1bTLEeqnKGd1ZACySul4aCmfla{7xd
zJRF_BZGR{1Pjpe>=(<8cfKNd5i!Qt?p16ijK|pv%=qlwyEut5$x9$qRzDD&ZDx<8K
zm`y}`huYk2kc5U^boL(h7ioW^?0-gB;QuGe{z=%s=t2Xo<Ky9khfe`e09?prhL|8M
zZ)ApW0e)*ald78OS3<ac*SMz*((!vlPk(<7cN`D7C*uy_r_y>r_~$0DrpATmF*K+I
zpwjw7!zba&eEqxIweggww14|Hot6^SHLlDM0IprdJ$MCo^Q%9$oXH2oN&;v#+%54C
zeCDTl!(6|fBAH*w3?cdBt$%*>|KD#~2&0{+_V+7Jx&I_ez3LU^ANce~{#@_&Sc3?l
z%ZYIP1N_!>RzIEoBJZD#$1o)4cdY*VD+ni*(IzOTbi&_+@~1r>>W^Ig#pOS9_Loqe
z;N+N%<M6ZlxEI{S_2Jd;KmGe-?#Y76XZT1RJC8veRcXLU_1af7i5eJSdM@tW1d+Js
z`F?-@|MfBVKBjUFU`3_9nNEPy$x6VdO%xiZlsFyxvkLw5xAiY(mg7kLC;td1HGaq1
zUw!n9v)lhq`iWNaSOXaTn|l3SN0^m3)(rnCq5ndvB2MEH0{_9|Uwr<DHgk_(3E`F?
zH+;ZPm0V!=>K}~1Nr)pa`^r|9I|ojIJ=TDgI9+f;tN95h{Y~@za02^+WAtys;s1J%
zx1XGk6X+G3bT!hi>L9ftSrEWIsyf@r2c)&h>Q&Ky8f*U=Fd`nq`I*J9e(PHfPCecK
zDF&Q=0AzYZ(Q%23NdO!T|85O%=g5D1_D_49kr;CNZ2)jfJi9XZyTS6?p!vuCI!+P<
zzrKcR95_zv05Zc+5NSaeAUee6b86)bLgiu&5HE^(Gn%;<e}+0w0{Ej1ofEj%_g;zu
zK54%)3`ek0;*=}#F!y#IuT!p5PQ=F>7$DB3Xrq29T(TWE0H<~)m7l)<GEulbbJP6e
zdw-zz|K^;cZzVuW*|rZL<IY%6lO>A#v{UX<*f8ECPD*_O!%qltUF&S1JHm~Kc$R&O
zX`Hn&_8I!6W4~{8w)oN^Tt&BkQ`0|6K&QUqFM0n<zwq||-IDyBR-Enn`v|$^eop``
z{vEG2E*WPO$WWEZcD8?^n%uwa+*ht_jM|jX1s9-?Gl7R#fJUL-ff0n&D@nUyX#4%B
zi%F+~da!bj&kv}VdP;|?n0;_is!(^FlEc^5<+j%~K@A&W9Jp!b_4=t=P(RITTGOM)
zQ09iC(8@C_qF(aU!SIf<1?<B06aDPY+}5Q2VBi@)ytX5RvB1$hjeXFdAlh!^!5Z{y
zZu)d{f84UtixS4K@9skDMqCXc1$+T8B{R79jWVpzo#YLet|!B%b<9=Ty`uQCwWQdF
zlCuh1IxFl^k*4)&DB43#@eCvq_B2>N2(B2w6ml)`+Gd{%jJ1_iD%Q!VwA3jf0KY48
z(d1rZFyY$}bCWN5Zfz*HB;EsQuSN43rXlYH$6Fx-$VPM>wDKna@f0Pt?k6qR+sc`%
zVi-!Bkvrr_FULbuhpv%4Z-;(a*DRzYegEBknd*cr>V!R^cV=75TD9SD#jDjuPBO3c
zKJ;~03Czf4=N1*#XnyPDL)yB2ZBUN)^O<TWTY$G_w}SLl7`arZ8tlvFz)*b$E6W<i
z)O?p+)`3>iYO$;Zb<2|C;%S&056NN~YS?2$-<_b^<eQtEy*p%j#t=rSS{og2*I_44
za+6rSD!g)y?$`X0{-@#k7jpk!nyeW2>&><|MF{gvUp$R7LBxA{BcMs6LLzRWWJ8g2
z`HkZWjl2nsV{z2x1}?m*b=b{pK1Gc}l>+Kg?$AP*!X(krno_;;%E%(;oL<7ZTUv@b
zZCj}FPe9!INr+P-%LGf#Vf*EQAUghRWGNI&G8nI5tJ$j)KSy08;5%~Ou35oaCC9m`
zez6uX8Gxo<`U%)cHA;=nIJZ~Aj${S&=$)?|TAe8l%8xfzHrbmQ?u}sE$bJI&<I;Wt
zh}(#=LjoA&5*o_o@HcrcW>wJOvzBUVY)YMg@8bDt!0i3tOIYc$N(C!Irm0F0F0;vf
zu{KNF@vMBg&#eEm><j#Wo`<-rl`?6Sc}MwYY5}n7Sga*K?)~7s3}5iAfF7mn)lhUw
zQ)*+S_xifA+4;S}p8!9*nVmS5i-W8QB7}^BR6v8u=cefrAzz(a0dzS(0dJEOwnI$2
zf)O&Tm7GU~nw5e_MPYwG4HrDWhmX?QNbu3g%?#lcaruWx@g@uG>g-Ve^h~z-hE4(_
z?`wNozheEPh54hk!9!{P(_sH==DqTdS@>_x4sM!J#aU_GWS35)rCJ+~0#yD-E2CQb
z@#>nyjb*A1mj(y&wp4+#oy}xT*}`iF+ZuiEa5jS7-r&6+b5{v%+YI%~0P@A_X?CC8
zb(iXmI?`AL+zoP(Wsg2`PCJ3A_XfiK`GHoVnf%Q%{R1(_dexIq$)3a&Kel=DNY8Rb
zHWhY|I3^qV#?)+fCS><`tc}%jhz1B8I9*-%7VgkiFEz`0%hzIT-Y1a*F(~0|*TuyY
zP|my7M11U^cnr<#yw1&>Bsk*+usKbt9*zYnad#xEsHWSRe>;h+^YL4Pe)i+_0Gg=}
zQZFPu-;%z*R3F;T%0MX-j2N~sn5ez!#cG&gbBIwjuSMnB&bfZ14GbXJ)OAEiS{a#;
zKTqzEfGrl=XzDBy?M~E^AwYYYG2j=BPdl9->l`)o0e=+fPb&F~2c3+-7)m+cvkfot
z(ktFRQD?yoT;KZI+Zt;dBjtm)6vLl7EE^-IeSd&WvQPM}u2)2gy&uo4KLcX<8gExq
zC|>50ACw1pqHB9hG&iutCF%EBYa&r7w<^QfDw%5sT6vGD(mwAS=3T(~F&!gPW^(~M
zB^nU@I<;4oNZbI3xZrl&E@7*Rb)z2JX4*d;jmmQO;568HDYDLO=W>>O2NrSgy2Pu+
z!06j++CGJrugpzq%kDsXx(pZV30lv!j{IW~Xyh|I-In$^_aieZMKUu>8))xtXy4=a
z@{X{`@;2dIonBc)|8+iPRR#3A?zu#!N}fUSf<N}Otl%KvmOw-ilRfg*Hy4jxhd2il
zvblX!G(^r>kWz{bBOF#{uCl@t2me9gNAODXw%|$|9$54OI0l6X8-MeMu5b&;sxn6=
zZHudgRWtLlDEeDMYXe$k0?3gm@>!e{XOD$NPDjD-;gfwLN=uy#7Z~{1x`}@>U+SHY
zrp7Uc0Vg($t|z02!(d4M<c1Ym4wO<$CZE42$J%krsJoEET>sUrsl$??o=#1I8JmQ_
zp8yOS{P~O*XrPrxzVDW7MrhDS=VgQ&Lq~!0>_Hsx5h!}s>5V6^L3O-8BobPN6@#z+
zP(3dKNzKP^-FMFMolW<uF18&sne5&ix)+dqFn>_0{N2~F?vXLgyPp8GHqo*{tOHAO
zXMXko!hHNVdV20b$$HET0-|#FNsTrcL8uaQ#UDbi{~H(PmkZe^gzt+WZE9eT)>AKd
zzvA$ut?ygN+&uGo@<?e_hYL)vPk9gj_89FPnYy_=L#=t3kyIqj@tbZC$#|bv{0Fgu
zg{YqZFO~0O7IPO`X_A|!;NmGEn|no;(0vhSvGx3)fW->(j%ad96GzNpaOdb3chZd8
z3)uB%&C;9OAqb~j-#jbU_gjba0|EWTopTPDne|Cl!v(f#x&9frNq8sRu7{GZzeIuB
zOc0aOBpvH(<=?bBd{6?CBC5_S67eipd<ZQuW3VTw*$Xg?lJyQ?RaG!tkex*`4e>fT
zYIVQ$=)qn&0TWyzB+mKNw+&~96mNf91@1A9M(7fGHBkx#&U>zdXy>fZW2wAXH@F*L
z9EiGQOzpcLXSu*5llOlD7*Z8NDPc*whvr3NNuX+ZFi~2Lz~g3dSmZf>#l<{`?YO$=
zD7(+==<_7BZa-&l(kUQ^DL=W==kB>lxQ6V^cT^8HkOnuS#m@P=Vm>36wM0|HPqn%O
zH;&SdjqYy_e6B9YI4*NWaxA)<DO{v)cKiekB@ByiG0nx_E`{!Qhj5kh9R~g~%D4zh
z{*R7*TxQ-$;49|8GV-{fFY^^G1m{%#)ohRFsZ|>t0^s%-(%wqPcmEYa0B|9JgcdGl
zjB#ho$HfzWe0g#{0Q1zEq%L8TWnZS)-XH={DsY^JH5_ZDffGP^jcZzNFC^u+cSUh?
zZ5bWMU_qVD9=AlE<~|%$7BOQxZkn?$n&($4aaxX((VPu8o~sh;2~DH9S~tRZ&vWbc
zDHw0ACw~(p3BT2AEvf$<tZrPS7%T7TS=M`;Y=A=8F?66ktKK!e7dI)&*%546DM6#>
z!HV}a4_xH^1K^Iwv|n8(V@Jf4L@*Se-tMby=8XkZz)Ltku<bp&c77tMXR&Ic>#{X9
zBm{qttb4lRB<3SXtY6qp1#zP#NnqvoCJ&Rf@+`6Mx*j7Mfw>)-G-{+*_wgbZo9%hF
znd%^*zV0P>4(C8~jVO#wE3jXXy{uzOYB0$7adCBoUBoO8L$9w1I!)R)BR3>n|1oT>
z%6?XKE>t=^S{*=kfz+?%AkrZvD~p&hJ_Y*;6xq%4v)D0MMX!X{ElAQIiY3=%sw)<c
z)IUg4iZMjXx@oc-QN@HwRPpxCRy6KH9A$q3q}?tZcbyiK&v_b?^Hp4`hT9d;!HKPB
zK4T9p9IKElzOJ;j@r5=yx>~|f!?vl)u?}S61^nDH*%MvRZ4-tm9^c_Yp3-R>U&m80
zvz;GHl57cK>-9}j1;gG^u_x?7IzB>PI>l<^L~9P2qT6Mmg&ry$6eg;7SDP)W%$nRU
z=2wW<P3`6eLcgrS%2##<YL^}hd<%v?ht1U^pB*Q9&o?m|t}r}16Hg{6WbHXoCITIm
zbV#<3A_~n~?I6i6{QL_-uZCc0IY4;kc#gZE*FG!Ranu2jMJ9PtZ<n1u((`(N^!`9F
z%>IPjUAXc^kebuC&Qi6#0D@Jus<=(}TDw?RHZ^&(<7*{#AZ+uy_@_fIdsOHLtS@0n
z?ad8j7a8caOCE}E7U7M0m((k~62NMs4}`5RzDy7n1)6AC#BX4GmU~`VCs+5`3N$fA
zZNx3YH{!QzOY|><v^IvO{D+p^{pRS3H;fCotZ;+&P3f%hVKRHEvuM0_B0F?9o&IZ~
zbZ^Jn^u#A`icJiYO+nGAaj5e(=UPpt+^kqeD05J}OPu(#4-2!(51a@ghU;4Q*}Kk4
zajqwOEP^z$5-1eWN6ydLEd&<*37~8(Fx^0CTLi}?gpub!`*{n7iT73V(TFc^dw5C$
z(92>=W;g5cbG*{)k)^WwE29hery=Y635zJp@I@JOv)MkI;|lFa?P_VO$_c)NsiJ;|
zXcc2kg2o;Bl*}1I6IJ+~jUj;m!!-?rLzaOO;pT?)yTr3{n<dR$a<SL*r8m|kqT~=8
z){?_pT_Jli%*^P`Z>ZZ<`aV;X-E!4eYvgImP?^~!6WgcY$UBJLxf=Gz%SSp~-{en9
zhSjmdM#g*^{p!~vc$aU@Q}1A+0~{U1k#QyB#U5_H)koJ0Rc1|;I>YZoDt<x1pAMmE
zt_gdi{Ty_-RdB;_Q?OlY20*T}b`AawV`*H_=X>#E<^u>x63VeNUF^m?vafmbVE_Z&
z7M0Vg+KL80GcLOlLH|ZT^@8Ua`B35&t^bzf_@8GFvoQ}yf4Q}zCw4I|&T*OiRUa6N
z)CJIkcNsT??E>A4i+bP<Ky`0dWW)0rQ$Gg>x~YK?!72gb5~4QnWR^eB*@4z36$U0Q
zHFXL<fe%Le_1|ifFtw;%PTGUpwVZq&vw<^DzgQMqK3xcQ3`#iB>X{L7S|dW*htOX-
z)$2;Xoe?+MFRX&H94OnUTu%-_pN@<#$k)S0<tnFZim+XEO$r$JE%rfW3=8A7ZV$)Q
zjimxyPOF?wG{h$)lVQU^53_lYC)hpF_c+nfNzQH^5#*g7ZC3a7EH!1d&IqNw#q2X&
zv1k68f@kySR3}kw=q>M1&~p?XzK}f-j$~dlskcr(^|QUgpJ&Z2v}>L{`E>HZ&4pq5
zN2rdway%R+DB`G1mdZZ(8qdHd9=7%?3P9V;5bS{MuOx*@<r)x$OMUz<(X-7rxR~Vk
zuPo{Bul?=#m)!nKRPwI~MI(1;E1dwB!^Wj5C{utrI7$6InU{@G|65{;7?<fXU|}1*
zr81nCn0#GYM{TRc>D!V&n-}IKqX^Yg28+p|M(eSXG(%MC>UfvHMuafYP^~kqSKz!c
z>K<iUy-3#rn5^J9gj9?M6x!$_yrn`S*XOC^2ERcwm6~bqnYQ9gr3QDs*cu7Ww*6wz
zD^RlF5Ssd$Br@T2<QgKzHD86gh<;kLzWWB@OR&VVV^1LD(YEc&ui=)LcO77^b#=<_
zR#qA0AAC(%-Z&L3IM>FyQWyKOULq`~{7UvThG(?ys-{tr`add=6Nx~WsaTF~C2c(>
zVVAVayOVIi>!JR2jIxJGfjxlTo9VMB%gBfnR}ovFkeeshdOVFO^Y>{#Wvf@7ddr;O
zL08lA100Q&u1r>Xysrz>Z$72u<<vvOw4qy6#5zBJ7k3@G%ZeI<*yJ0nz-;4WT>9uv
zRCVI*yOZ2dXi!Ah3cowi&P8BWCjnSkU?faHobz~$XpZ0g+@Odf(RS&{re>l{%v++N
z3Nb%*?L?@+i5O`sE-<A*$STZM<h78Nj71@3gj&`0Y&fDH*Kt)%OnR9%XAI+i^XM?*
z0(3Z}o0#*96El$~A3szWttkp!EJI#37*S;eQ7gEB?sdM8vWWK9=<#X_(KwhSvFF*C
z|8z1PbTC~i^)4^=Jm+j28YWmE_SFm7QDrtRAss$~DxZv6Y-LjO=$<j$8(=tp!opV{
z*E(SP(pJOi-qRB7#dklR_`q|Z>l{$=OlTu1fV|`Dy_yY8cd>Vl+UjvcdQ7}cqW#N1
zHp3f3v!}e0wO*f3#=J+>aITupWI70X6%OgQURC~)AsukAo)+Kgk>uC#3eVyLz!vUj
zgrxcjxapu{w^D_D{5VI^%C$@WQ-M+!g{g1xy1OP5ys*>8?y8enX3_qUn3@6=M_kY}
zHvuwC!r1seJ<HdzYvroBt?UsKwJ$zcb6z~dIhY+)4rvs|%;Ig8Aoboc=5ME?TU_0g
z5DZIB_VEVL^>zqXk}L$Q|62G7jAYGv#5zdwc1mW1;war%-e~>2rhetM7K7SFDm*Fr
z4mpL6-0HEAR`x!gbmQ&B1WOBSiNUs|hmwaJGh(~S1fo-uMYosyN<OfP|A4*r<K0F!
z({h&BMN@k0UHi~bi$&!on9WJ3A8*|dTk^AT1G8?Voagmh+d2-s<uQ<C^k<}WcGTo#
z)dKTZLh!mKa)hs5cJ$O-?VcUuQG#jS47yC#tZ>Rs{Mq<Q)sLwuCebke?a?xbd7S$D
zM6<qVp|h^!@a%(iV>jbYOTS^vK1PEG@noLfMv@U|@)q6d!=}de0sP(aLc~-tYaRAB
zhF0b|#d%b7%rGv16No>o9>FYt0nQdFQ)4o(hx?#|92I3F5KJn@2|zFNy503CfZSKA
zRPk;}lrx0I%_BS~dAb7e6ToPsZ>!UNq^(2%337}b7#8dIl9;RYeK2%6m~%S>O<Gz3
zWZYGuA6PV~)*n=@cz^tQN1|m^{4~7nexGN)T7ss~YhC$PvF76*H7rZa&UPzWK9OcU
z5id@^r?86Xsb$#MXrK(&!kKK<XoZZxW?LW-nnodeE7JkAlia`~MRzAcDx;h3>*aQi
z)y8OUSwR_gIrw1HLOtPOexEzl>c~ApAZ#G$@}2{ztmRA1e1T-U>!sjMJ&lANgYlF`
z))wO{3BkQStJR~bsNOr#P-0e}rBW#*y*Ep==^#}9G_9qHR!q-|R=h`=;#_$DaH+ww
zb(%X8^j75S=$79Th0(?bJ0m{<A3;C`?I~!S!RPyAC*bU3&vu4-;j2k_GJ@duA8}K)
zmS<Tzm;UJseer8jR1h4sHf_VkEy=NiO{?kc9yc*vw@+T3^D0s<p~Z(OXW3crO!>*6
zlJoH@e<g>0OB(&}_P8p+ZCr?wyhKuT&18frT+Nst6s=ZS_i^TV9I(OuUE5pFys4*h
z>*8wxun{(ySiwAbME_d+<CkKv(;uf}LWL&FHSi5fOZzdSqvr=Dj0Op{Ux?r@lTtd+
z3+@KoA7lF^^b^hxzOF|L&_laQ#`G<Mj!N*2@m-}>pH@VmvX|+^n3JzABa{lB`cCP3
zXq(?B2Yk1DNwYy_`xP8ff3&TCSUQ}F@Sg@w`W42oqu%VD>Q&EQZw#u)i0N4c-$~$i
zcu9!RuJb0M=Ck+Th6U10)rg6@kxewWb8PS$N9mR?rObFfRIL}Dwevp{hmlz3OzFwg
zF*twnE6@uWu5W(on{VBUq`qZ&@gsmF&0Q)%!v1>qna0$xcoL%0>42Jb_KAljS&w-J
z7pNFJOyn;11(p0IWJXN6+Ym1NZXO}gb411IQg&*fL<Nra`BhH1vYGeiq#iMD!v4z=
z*xzbbn<#)&POi?%Uqy-k@qEtTGJ}7wEdWtY+0E%!MPE~ut?;%r@_3c2ufMUd_k2C|
zvh?xs>(X#q<pf^psK!m-=6TIqVpS{kn7t;~(ONrY^yd*yc-Vy@W8xNBrCVXOf^0FS
zsNeU=;c3e0N|i+aGvul9_kym&KJ__%7(@d2vCU9@J?8zGY!dPZoZ43*53!U^oEYyL
zv79n}Fi5q~Z=n7ORfQZnXR}%UViRj|F>bvW*<fzE7C!d`GP%)><qNW7WS{?zXW-c<
z7>2M}jKh%Yt+Z4vR|hKv_k{6jwbQeCK9?B-3+KAE@5L26-3?kBpX#KdDT5MbmI_NW
zAzpaRbi~KS<)T|1)^x5!b<*`ujVO?<6%;#=6+JZo(d^Dzgjt}j51%XKX4J-mV>Nqt
z`D;~C^Ept=Nc5AH&~Pzt26?86+V3wz9SpW4OunT;rizIchl-5t!3_0XJTx>PZR`uq
zdKGf?QMr3fZtNKLtR<<1(u~5*0+kn6I!Owt@JY9)8?Lo08L;@o#MDtvWB8YRx)$vO
z%oYzHZi1RB{d;F@Khk<A)P+ZCx*7~_+ZxVUR4L34*yni5KE>~1qnx+j+>k3Oq!?P}
zTlUaT*7mbsGwFM4R_{Gme>SHQ#uh<aMk)H>GoO&#<}p~c)GG`4eZM^Yni*rns=Now
zTs^NA_GN8JrbRhYQ8w5!!QZB{tW<{f!!zR#YcMZkzCxc(TUDB5SGZ|2V<)>%Wg34m
zFdB?HEX|a^<s~o2fCg$Tg6J&J6q;_lz5aB^%G<_udY%(B+cV$th*(0Dd=`cb5XCOk
zcU!()eD(4?66*=h<B3={y<3>K6)&q);5%qfD|{`chnJ2sedEDF3HZ!a&+;RsrTLR7
zw`U&3X1BaPlyZbUA8=)Rq36GLNevR~#iNUhJ&uYmKlO8%4wrwj=5-^lGG)-3x*B_L
zsGUBV`Hse_6DjmgN%{1Oio+qZ426MmlZXIbfbW=8zM0V~`l*AYk?pmv=4OS5rjU8(
z_d`=g*Br4{5M&QUhj)qe$S1vt)n~cW!8!J((y9oq{RVNpsZJLAl+LM3YJwQ`IS5RS
zWJvwP?GK7i!L08&M{CI6%2U)L15E;S&3NpB;~iL@@T5(5NGBbn2<$Sm5WP{a0w-Bd
z?dG2(V@sYuHK%QWU^)Ta{`J->ae1cU%n)5t&Yj0n9JNZ=`|nt60FGyQ>E}g(L;MTX
z5mHV$t@3UuYV$+2biEREZ^Js8rZXDfD$8kk4DHTq)nAcPwQ;Nnc<0<Y%UD+=87PEQ
zigy)eu{>SM8ayC{$WuFbv4yK(4>)2BvHmHuDKzuOW2KD)HrHB;!;dDJ0VO^uBuiU0
zbNsYlPf4fN#%>nh<XEmv*lhQ5kK7c?@KwX&+sZRG)|J`Cle)6sgYGXiY8_*-Sw_c$
zrNyqY1sTl(D}0#*cknYD{pL4^3eF{m4w~W3MVh2&Qt{~`-nZIB567fR>_DS~G8P&=
zNl%Xw>pdFjvBN7Ycd*Q(aO!4TmcEQTo@uph%dgFV=hVY7nmbeFa-U$jM{Y7Fc1I=;
z(D3!o;NEHWV@<1Kufo<2bo-qCO3bybk~=fe)~1CTUaX3y>$5)r@8b-E!(^ont@mad
z-*iXX&0Y!H5gAQhTi{*YN*ztJ4Ay=7!BRN(L_I1DtZ8EdDKOmjR(wT_j}%+-=Nons
zle17mR(S+n8?674mkWcMq@-q%eHL2%vJB-OH=>`5K|ZqJ^wVJNOr<D8IohcESb4nI
zdrCBQNWHiGUY-@+J{HeJ$v?DHVX>X2cA6>U*?^=8@wldc627Jqkn{XFa9Z%j2~@?t
zV#-)v$7MyKqq<OgX=*!9uU-FS54a=O@v6YJ$bQwvI<)>F$yQ2aZNyBtyF9~jN}i&h
zgH1N^0ev<kP*v61Vv8rT&H2ij!p<jjpQU9@v|H|@4ZH_FS9pzD{HZV7E62q29djd=
zx+Pv8Sd&oC&`&TdN6UOBc@KvWNNKEtp`-N*Sh>sZ5P^4fU5pT`>-hMqA^RBe4hFx4
z45Cxf&SNGDGg2>`xz3<{##f_7K2Seit8M2w0ybFTE9miDLsoMdMasOAJSc<ce4k6N
z@FZSQ9)z$xts<s;zyxu`r9wu+bL(fC8%3&6;_T29=ZhAVgObbX1CD;W`^--YZzF+|
z%G>7LC;+AslBf0H`hSI!{CibjwKmX+u>gnQs6HyZS@-U-KELRq56Vr<(4@=x%aajp
zRq^44===eGI}4$hZaIg$yUP&$>NJSC8F1T(-FhM|-<W#axFZK@nDnlHZApx~=#izR
z888*mZ=dM6H&i9ajv4xLAkIT~iki}r+<Xyidaw*z+D=Q_Cwo!Uf3PaP-S_OtEQ0&$
zRy_*jNy;Xm0mw&)lZs27qG*O;03Ih|Pib8JKxd3P&zWPixtLe#N&#Llb%6uNXb%(=
zHt{7K6#X!4i#mnsb^JCN!kW0Ma}VJEsm)%IWAw0yqR(Kkso`lfDCxdHb}Tz&_@Gd2
z5?2gvXxGRmJ~|s(7<s{A51c-Du3Ej>Z*O*A57JZD$nCx1>Bl9+L59qcfNDEY)f8_-
zUj#j36zyvrjaH`WEBXl-E1ZE+uv<KF$&$(w;nlsK5l^47`h_+3la`XV$`>kJ3I>NS
zdYbqB_b>*6N*&*3jlc8<f5D;Lsc=GE`}cz5|HyaDtghajKWi1cBUz%;k;)#bIJ}gk
zLXwryTo_VUChS|doaAedHTLki>#0*|_0D$?nlE4K!@m)HS@Xt_mEs&C@g5Pw^wc0b
znc&Swnu_^y{V`AXm8=FBQ<X{zx5-rR&ej`_qB*ugQ;Dq?tan*L?8hSNAC0-?ra5_B
zbtKSGCx?xq1fTD01)U9SV?2c;-b}6=2c?;JH=#XmpOlMlCl~sFRhk(;4@}2%myvz_
z{&MFL&(j%q6&+LPJWzRe7xlIHC%~cJ3S_kw0(-@q8GW=lHsI1f)VLY)W-Dl)8q|$(
za#v0yp-X3%ad#H^Lft9_mAjG9u0&g}kRETud(!|iOUfSBh;b;vzUvjg-x+X)pwgT`
zeKNp>m(2hbQ1<e15Jmacitk2)uFoh(_#DBAEJpq!v2tI=oU`I50M?);iqad9Mre4u
zeMEOHWXJTe4frVVa~=^P)NSZupO&*uMb5(<<C<Q0sw^J4s63J@%q@G#jy#j!=BDyp
zlkou?o4a+bzHmCot#)<3VG3~3v_3io-DsFo_hXQ{Kf9%gWqGrz>pnoCm_0$y{_>6a
zOP%4#byH{W!h8K`0`Qj$+yil?Gt?2hHw!xBWRidEY-NASwmj}JPAs7?1Rfndkj;~#
zcylh6wbR**WH@XftGw2{z}gUB>9BfG5}-D?<*Qm`)Xiqco{&#*dX{E?TG4Ds82?)I
zAp_oF7f~OFl=XgTYKF4@2h4^azy0@?q4KHmp-M()vhntuiF%iJ^hqu?7k&<LShG+X
zoGQtvlQGx&W}s|*V5#<+v|cI25FO3XrJ5QxAHnAi(Mc~d+Sp$4CBa)Q8~ycz3rdS~
zfZE|`d2u*c8>F=E4c7c**w*3)>|e%wi=3246o~CJ0K?q6r(d3VvChDj_iN&YsL_p<
z8#AKe=PF}zt_Ent6!N+@8Zu^8broO6Z#}MxEn^}$3t~VNis%`*Io;0Wc7pB5eo_{3
zw_T#;k`4he5GWn%4JS|1YFs%TJqoIn6m24NfZR>Yy;&U2y{2C^vkvx^$ral#dyGxH
zD~b`3@Yok??_L;ORW>dwL!tWw;OxQ_i$<)m7EHd4BSrY3pljE7(pnHPMGntwt9lEc
z$%N-;<+F@jeeKHkv2u8UKFERy4omIRc{jj~a_7<C6zzMo_QaGfR?D)(tia3|tdh4C
zBq%7QL~qAX6Uw+!JS*o(u2wWBNI&W2F|Y}KXW>^Jf-#Ca4cS>6h+Xs1&2qQ&l#UPO
z^X0X!Fne0~6VQCsTWn9WGxEcb?_<<HMnF;#t-{4m&wlk|YOJOyR&VURJF!#A(vzAt
z2R0`19CN!`czBRa2%a1|hTW|nkBrkI<VWnwLpdH@$F0#^P|c6CnuAWHD$6JuLnZx{
zL_npHcSv-bbYfK=KSukt|AcT)BD-TOPZodJQvE5!ZRvE?raD#qjR@f`{&OZ=0W;*3
zeD+4Ne9r7-1IOw8a*2f18$?rVOlK(Vc#-eu%yQJ8iF!w8I}2gj6BNBR-NG%SM`<F|
zoJ?bIU&3;SaHAGwNeMD(8;sOOj#sHz<m`0fEXIC_Oanw1r%zZJ;9c{r6n1RnQue6g
zA?1W&$C<V$v$D|1P~E4E-bLG%^@2@?cPGil^+Ghd@4nH`yWmUVj08cCK#r?xK0ST2
z&13#ST_#JA;0G}S3IU26T>?}=(Ei>qfA>i*R!wzHGW~}PANO*Vx7IYECiQs?_KOz`
zdPd7B6S4Z04>eW9+}68MbI8<}#%-6z-dleDu`0R<{(`%c7vn+SzOES6fi9Ens|F{H
zicP)VJ56CkRNDqclE!3Zm{TG@zqr757+GslFELb@HMI3Gba_fesC)5g_E2X^wn|;@
zAp_KoHlaQG&no?&TX)brYl<AIA8Wr4@fLt|X9PsOKz7lcj*??W82L3L-AP%m9xLPQ
z1V~k|^yo?`Y1GYug8l%t>7Ac|HaS+sL;}j60GkYYMw3MH?@NhiioSQG#j~zvIVL}^
zf2F#3&>E>@DwN|D2qXhtMIx);MT|vQ(@$=iLewHe5S4QJIy!kC14v&|EdHZ0YT8W@
z5{QX=6ed@+LUvWTw*391ni;+%yLFQY+ZY1F;Q-N8s||m7Xg%H=S}V>R3HLE;H49@A
zKtxCZj1r=%$~!q){9cKDFO-@1Jd_E4IZ#*ZPE7Ug(@k;@xRuL#)VET~9feL(6GJhe
zs%f_tTL1lp!<cA;>9Ec#IYgTq=TX!<Td3&hQVLaKPWnR}PBZfr{XgK&e;tkElIq0O
z+D=*S;t4k}Pch1CPHX1OwORaCk)l2*N-<%h4N5U^G_bN<Og9NQf_P7<8aV2^%W4?y
z&GUATWRUlxUhSvTQRe;21K65<OC#<$+i=M=<BsV;@7gUVxE~IpwCv12Gn7EZVUFQt
z(}i?>AI(DcBz2*(<qZz=IXLOz;v<|x_1kUu{^za-rAniTh7zTsSen@;0v%$e8}PP7
zrqXr2I|CNlYtfM8h*%K4_(8S-!hw7K-4UG5Dcr(P2+n06l?)PxFUYvQ24y6>S!5g)
zj~m%9i$TnwH?F-H9x9-;?YvVYgcMU^rxtbdZQ2-SuV=^5HmDUBk1rMSy#`aWLv|0A
zQo02=<dhi0AEygQ_4{!df4Tf*+7>6F-7$VP6;P))WO5;yGwreT`NQJlvjS2H;{_jG
zzDcjO%2aC<x`9VoqtpG7<z$Dyp~M-APpfXUL-B~EVYPm<-_BFlc(dKgC+tE72cqhg
zbIEhc#(5<GH9rZ6^q<S?1vq>pE<N=}i0p6f<u9}+4r;yf_ku+7uY3Qx(*AGPaw`b)
zbhV!mctQz7BOA=b{J*ovhG4+zss=gG@VX+5uSro!Opd0MUoirk3KxYPeGot-ckJB-
z$w>HsomfccY;N2Qu<RY$FJhE@yVug9dP8|{U=#sae3xymOO`UA*1yN7+u+@}?{Z#N
z>Zl{evAfykT1|fEq@iKGiq}L#f|pGKjLK}R=OkTd^Az_UC@ZmO_h*Zv1F^HN3SWG0
z_z57X$hWV$R8I&z(JJdnn|raosV~p^0g<A;XHd-;EjoGSkZe<bBz9l<OC(Pn$57>o
z=%m-Ef$RjzVoBCJxS*s+X(%&f`u^H=*EC^eStnR4qRp`|+jpRqC4gak4DopAT;WR!
z4%{Ym(5J*@`WThQr{UagIDi%-=O-3^GD%c^IwB*gZ~p9-Id@m5r2Ij6&Xiy6uFnKM
z4Vz`aJUF|Fmo@(Tjmsv(LyJP$MAIYdzR~;`^tLD7+=<Q<m^*(h)Y;xD7U%S6C~rmJ
z?4CA6&#8o+<m=-Ga{IY+s{(cL9vSy&qtM7WG5qv{qE91|Q@V3&sKxtYq(!Cl4&*iz
zKLH>7$}oJdF6N|3y0KJ|1@;b(G01TU1-M76*<uU}_mgj7X3?=?vD}^Pt+A5xD@s2$
z`%zO5EcR=h4Vcq$TvDc+Q`wMPUosPe#clbCdf(q^vKVT-bu7(%qeD((^^tKGLoC@j
z>AHyqqJ&SV%j>-93EPI>DspUa>%Jf{L*V>b^(Dgs0{BcJ#r>shwXXR+x%kn>NHR6i
z!9A~#q2jI6DEp*xyRx2qli29h*Y`_HbD(I)XydrrR1RUGIoX4sfOU{>d|&Cb_mLvW
z4Yn{3Gc5euPk@^X<59$YmFIFq4bZtVUQ6fTmQ@d+|FHhV<w9_MqDPQvD7dnRPSj7d
zvM)KZVac6yRa1Wsq2=yEQgar9S)+qkjtXcdkz|LzyIf%zS?bu0?O!$I=zJ4Cos+!=
zF3+2otBrk*4BH>C-#W({37c_5S)Vc5r=Luob)NIIMU#3#ATOSkwbF%gtu*G=q$rEb
z0aYC{T;dLK9X67c7;Wg46JxgY)?2G84xXque)CKMPBVU2GAuRR1ejddxjgWe(UsxG
zj9O;7k_8FG<m2r&*ND8CRpZW(b1>V}s;bc12Y7b0RX!wMg@a+vcU}~9YY#)SMtmJb
zt4$eo=;P^SYVOv@w6WZd*RJ(RPE49}-hSf0JPR&mKO-v;xiVJlJs_jxiXP2z&TS2C
zVGo!^nrfbtG>Y8P8?e~e`BbA_i$XG{JQLg4BN(2gH=v7JLY}czpU$N_w4<o~yR!w|
zx28LX%2B3f-SJ4l*smO7bh-oSCB+7vnSSq2%nh7ws)aOEteg58#mGspm;f;>Ta&6g
zCk{X{{w?Nfw71j-D!#w+85>lnC<YyE!cC*thu?!=4!PsXJTl<Ea67&jBgCjhkg9Cu
zk$l2IX@UGTyY2LI1uXTPvMx~Q33JnXJ|C}hs0qrs*A--^q7#j^j&MROH}T|>BYeXo
zc}P4;)pz)Aj;|gLHeocUtDW9Voq@UiYG?W9BzKd5o${^pr(n(50|^f|nIayk-PnK+
z!N=enI^I1NJvC&=hoNH_u~?tF?@FJe#=<Yqs|f0Xw+VGna2o}$>=C+5bEPxs8oH(u
zU|}Hbl+2Nsr>wSFAOAWcD)ONb3)7JH5N4bvT9q9mI^?(PT>4_uP_8$2!yN`|X0F$%
zX1|SQM8QTM`*THaVbepG3|MVnd@ZVNi<+=2+6&dUU>?{x9s4xl)o)ajSF_t*74D>F
z_=Qdw3Ej`Ix9hWpi6d)wwJ4uNi;K_a62WQFTn)&;aW#;b49s$QOg{GY5AH=&<5<3C
z9GKBU2SpBB+fNhQB938H`vLt-uA|?85ZA8iB*sEplT5b4A$DRFePkhDI{dLmV!Nre
zwH$BqyVy<XE3JP5QlmLjkn43~hhy*evBidK_hqf_lF)UJ15W1}eX*UdAx~X<sDSx2
z9c)Ne?ty|*`G&cZs1>c~Pu`YZa${}l4X_g`2QY(WRNx02i1h~&?zQ`$W41#MHAyFo
z*F9h_M|;irQjN>o6DB>60`F5V$=?kmCnydV5lC@=eHm#%Syb{sQozmU6fBhIm5IKT
zZRnc_H{#{1DB~P=66@Q93wmuFda8dq<5fY7?gQl|ztpC@rx!7SaN~f;*VBd?Xj@@z
z#ETaxn~vEEFv%mTbIH>1b0REkTPfv=SuaFn&#3=hJj=+IVL;o0QlPH*5V}`{>CHlC
z*-=R~N?V0l4I;8eCebD|Z@I_C94U*dT60HRT@R!4VZfRZ9q$h7g^2F4xnit&9-K0b
z<*vHrpJZyrHtrDJ2Bike6~7eKeHO<XhfA9z7hE%R`jr~!#$nO^1p@dT`0RG(FF@MO
zU(QjCyWU%Vjx`A#PaPZ@jgKa#mzpUtJ?@NTh94AFsLD|IWF%ABT=kw#RKbv{o@F$m
zg5vrv$h|EYh8pfN7=w}wva71?8iN)56ii{D*!{fGn!5GKhHa2)O8yh0h8I8%hex~G
z7&5DhtJ$1R+5CF~yqwxPh+w#bM2~@CjPJWTlIUm^YB)rHhNFJC4OpXpI+nd&7Rmnw
znx?iVc#|6>8%9DLwTvzjOz`gKxfbuKVGq=6>F+Ki7Bi*kJd<BHm~Q6}SGT)z&}aK~
zhGT0a%K5glS&r?4zF{KjbU0K^q{g++R*{YRsJQx`vs%sdWc%K5u4GSG8B@>VzF67~
z+6IHL^Xhh)4{@WwRx#l(dIMsH3N<=53++O2vU7E==$r_RpMZ8f*=Eh~M#pCy8a4fR
z?Xu#Fov>1-w03l?ettjk>9e~n%QaV{vPswM1x3QXH+k|ufV*i5CoYALPd*h~$!Yuq
zIZ?@91*Jh{gdsEGN-?T+&X`<F6xmF{={sD-DGAn!)E?%Z%*Xq4g$cJ5E^UbfpvgD9
z+W0&VCr*i8s<r=EgZkTQwSN|^{){I1H}uHwkR~iV#o@0sbM&L$S=%F39rAia8b`FE
z?K-usrgm7-f}N2L&KI$^Cvi^M;CIH#kc}5e>$YNi5g0O{1&PBouu^+9?p~Vm-)47`
zliV)>jarWiB+c&Lozhk~Qh^<0u;^KQD-f_V9u9UC_j-tsQ*v$O!`?!-45qj3krnz8
zx})EQI$LW=<3&4ki7`KPI^gE)`t;Z(v(o9CmT_+{_TF0Q_)7<m7ouB@PI5TZmQ%JS
z*~FwBmOS28(@hgLFH7vq(J)sd#`yl|UX8GmoL`yI<?TjLV{M|RjN-@Fb*Urmb<(-}
z6#}bGjrAhjy%Z@~!|u<$CXJ-4^5cp)dp<L<5IXcpugEiFQc8XT>TjQpk>y6+YeQ;B
zCPr-efb;0MI(v6>I2IS2=xJIcXSJ`6^!u%o>Wwgn>3=%!M5q?p^G6#K%8IaB;8IxV
zDbmmnTc1C^$=w>MH;MWbiwWH>xL401S138vx5E3mel#0V8KYxqRw?7{u@g}!XW!mF
z7Smv9{6XFtGY$iZC&B0X9Aus#kgE}Cm$kb>VxQA^<ihyrnxP_K#hrzts>a7gNv9{T
zx-M_dco$95$Y_$<#w9_b#y{qw^jN)cG5Fc2|7$Dq;FnQ8(uKggOj2u&uRx9xP-Eb|
zk(Pz(ZoG4`k$0KtHL1%tc#xHo<{KiigR$mdPT&5SZ)5l+YZYE^bd4NtPd^3*MR}9$
z9~St!&T(7n4U3+rMnza!?jIG1`dr&j7Nron7jG+vgI#PENYJgg4CZ-r^w8A8^zS5;
z7^rlmE!ZEXFu6`54yHl(PClL{<!>H-c;&%CcoHFlOfOi7Fac%-YL!n5wcH3s*Bkf@
zm<*)9`=qkE4xKYm_vUA=z1ZH$@$>+dt$r*n#%Fif#b4hxIF%~~9cDZu(46bU>M<J`
z+C?l*-F*vD!p$C~>-(if<SMbo*fQ$UGK;pP{awfM3a!J$APs5MDX<=<ky_*Wtv8wl
zuAk(W%-6C6-m~&I)JP^mSbDu_i?Npoe_<~12p%|}PTgRS%`AcHWbh1?b7EOTI7h6`
z#G=?)<J}HfP*tPzs#QKc9r~kUdcxoQ^&2|jR^kQO$20?*hQ&X+b-cp8p_`G0dv$P!
zXUE|MU82eJ!{+_tH$OH`9F)elnOn(O`iaw$_Sj~cG59_;>M5<TeI4uR7>$i*PqXjY
zxl9_PsNWMBNNjE5YenaoT$#o^yN3Mou@TcrN8Lacs={=;qdoB<Ndv9ekA6@2)D$rS
zLj<@!eE#)A6+L|U5%h|*?3iIyN2rj#QDMam7e9GK@{b0g66-iUa~h#oV;6@B*;}(d
zCW4xAhvyq~K6R;H=!S3yCFAbO6AfU!!bQ&f!1Wa_lAK~Q^sUVWBBv0)c0WcB^TBfw
zyPOAwQP+KIE^!6<o=ZyGrBk1u0IF$T6&$Ld^>znsI%}-caZ`IB#($W8kMq+B%`ZIr
zPe8z`X}e_IhL>KEorm=9&Xj!z8c=-pn53a>YAnY~qeV`hx~9hUeaHCg6{E3LQ_mj*
z<i#HAo-VzTGotX33Y<r=&&d0B@(@>1b_LviyH9XBmdBo1>|GWwU%Z)7V^X)cIGL-d
zU^qGG4!G2}CJB$1S^*rC%9M><&2YUd`bvrQbi#-A1-JyEdZ51;*zOfiv1JM{T`Dg_
zR&x33VvUXicjXv{X0yITQmXCj?#7(6q2r%?pI&Hng>5`la2o`ZWKHSD9o{b~zFmH8
zID^H#lgL_z95#YT-3INY4;?8wrPVqFoAj}<<gKb*kXA@k`WcdPS;gCWy)b#pHSg|W
z%2FLAT9a2DD7h5tR#ag}y5hI~DF{8>D(-SpMx2Mk#)8h7kVT-9m^qr-1fbLv*%uW<
z<x#cc==$MJ1gXU<-ETq*Qr!~Y4=#ASWx=F&8;9cg1+I=k%HQri=y>pL+cbF2`Ibsv
zN8mlXg_sL1e1H$hUK*8|>3SDm*&A6`u6yZEy+f4m`>ltR*K448r&aoz81~aDy^_So
zuY*7GPsPLxx?BxV!^)}6k!fS$8}L~NTMblXk4$5>bz}2!#*f)K{;M5$qM&|kC5O+u
z6Frp&uKfzv$HEnNK&_90eERI?TKcDjc1_kQ=$;Rp3FY<TR*&(4Q#OnjVD^P?HQ1EP
zQ2U4POWyIcb9GaLmrusZj^&>FP(b{g#nTv>*Vs+E;Ts}D^^vAn)0EtjK7~0*K{)Y*
zgH0CDj)o}v-S}l}x4y4syb@%u{wP0>#lF|f^M@2`1n=WmPRlK|(EOUz;PL5OJKdAg
zz#J3tk$#%%6rB|e`xwD?hNUP<1iW`No;NM-RD5@mB|GFK^v6Iu(5(IX8_gZ3viu^8
zTP<@!+ee2XXA)ma3&c}#t5h@`Bqv#GYZC8umOS7*8PnxRs%Zt?5H4$)RrnBXrhI#3
zlcr<e-RgGRPrw%Pc&?6F!nkZC#vR@Zg}r<J<TPR={AzHn$zvT$I)$(4gy$7{96L{w
z9D|M&X%6F?b_dmbW6shPw|E2GH=yEM9+imR<AAmb%>63VMzUtTPGDZFf})AQLwcrB
ztI+?L?uv0In8GbANLCWUt)6zu#ltN;0HVH1==wj;>ZR%?cK#yKqxA$;ZZlf+Z6xzC
zLAV}!baqtXr*K2|CtVB$T^#5*<w6~V{X>FH15f8|FlTgGDLe0U;a8g(H9v=>`p;qx
zW}4f^n+wT)t1)cyi=m!{K84=pFb23b<<*lQ2R5A(y#{ipgW^5}F7x&Lo4()dldKP8
zbcG*fz=>W_k$p4rv8%9U>{#M4ZY5~X00QqGv=5=o%K<KJP-J9D0p*M+jKXwrOF29_
zb1a$3xK#SSw(5-=Ib&JjHi|+W0TCrCsflw4d^P%zk>dq3W00;ev9zV1{glI!yO=V^
zCpRbDtW#M*T|6#=yhs~`KTvniB|Kg@an<?YDPSM{gnON{C!$Hs2M@~W!99rV+148c
z4lXy4a!hw5-WxdqGy8w~TsW)0Je_GEbR=&NtG4d2y#i@<;+AkSmY@chV^*e&-~hF@
z`~7oGY98CLU{WMiayUSJ0M1y*-;uIJaxLmK@^5Zi1`dRq_8%8_6Dsslr5faZ$J0jX
z@&ENQqf;ySSBzCDs<m-{Oql(}Dtr>$I^qXpQ4T%+l3uZu0mMB_W_&S#ES7`SA^Gnb
zFmOLBGorl?MOb68_~?e82)dT5V^?ZjZBSN17kVR_KXV}E<<(W0<py;E;=IWYUB3~M
zqVqlNYrgBR*&?I-x9o`$jO|&p(_*FHm?p_)-Rh8zUQII2Nw(#Ev%Oz>H|E(Hf92bE
z(W)xlGE23ZAxRkyudg(Ee1^!8xiJ9W*u!G$$-ad$Mvi9Y9x6w9$S{*Jwq7zPxnwNG
zN)m+(L`XE&z9Z@l$ysnde5a2k!%|j0`*<fvB&`+kY*^slu&>;*>bs=p&@>&56a}yD
z9QDl7Fa&EQ&!h6ICgYR(tHONtZ#R?~6C+I!w4ZC=UcP)o^5!#a1fqepKt$`BG&gy?
z+HZFkiI%aqwl^%Dbb4!aY*BNWm!8{BgA3of9LReoE;)`4bA~oIH(#77mQD)NN2u&9
zBWLv+g~mD(ovPle8&8Q(OJBI^g9<k0+`s1yun9gJ$wig2>o-ICUWXhAzMH01_i;~p
zd6lb+Jyn7N7w6;uFuTzRFF^DB1QcQ7o#q-Uo+Zh#DyvHx=ziwzH1;gHF3Yso-;X|!
z`)o|1sa1bDjtYrcKr2t;><k$+l@H5#Xc8)G*1r3g$gFfI#XJ8R8->WG9zE1J<&#Z2
zdigLkT6A68qx5;!Xv0d!x093p)4L|*Y7RbH`DV;Vo~$jHh9YyRr!u6enh$uwM~)7|
zTc7$r5qUqBYiIGXBY)So7bXbnK(@Tp@{AVf6knC-p7?New?4ule+QFauj4Jm%aGD7
z2ZTzyr*t4*#+5O>>4ites$?9M#5`*-;ti;nM4rkLBiW9V(KRmn#*4lE#YzEIEV{P#
z3>hGFhQ&{S<x5<my#93_lnqc=t)%c3x3ryS@7Z&;j!VaEC+LJ<5o4PvSLG1jXHB69
zStoBvjNlntYV68LetyStit^{8{?3Pm_F%)IhVVDV;%Wf_9vW2S7RCZdEKlphgjIhQ
zR!F^8aLtRa^3#Z~$8S2(o$pEX?Yc)nMCCi0KX{$L`rk?v>+u%Id`J?*4O$!Da#$#%
zix>AZxh{AwEV0rLtlwx9uI$aC`rey8K7HW1oWqPP_mAV3X83HX<Z)vJ-z1l9=@oLc
zcgzCE+D-4CLAGh`*AO$husgMej2|OPlg$qqeCCe=tW}A2J})VsS}c}IYeQJ-Ew*`?
z6oNN~by1J<Pv|JyYp+)Y#M;`wPx2AXZ*>dI!omd!?NsxQVGeb%LUjw#K3v;cz6v|I
zdK6VXm|eV!-Gf5-B-&z;M8;3kzFFHwx9DEc3XHr(ldz}A&ZQUJMN81bx83or^LWg?
z(1;AE9Bl7;J=(d}YK@K6oqV(s@qyo$KP!~f-7J>Chw8@{W<s~9^$!{2y_;MH%t&<!
z=uf~}sbM%Hd*szdwj}V12doipm%U2rKw2y=UJ$m=9)ajwS-mL3C716ygDx<Ya#<{h
zPJtTFdD1X;2ulh*5z&%>PqZ9w$?$<~<+}Gr=IoO1w8!$X+d4VjB}cMSx=XQfou2M<
zTpQMrnw7mrz-Hqs_&#TPJs4x7@mAI(P7a-qH~gJOmS}3*;;O~lLEb8S-p~CJC@-Il
z!`hee62dX;z@<<AHT}Fi<yxNNd9>`X8}U5w&x0C0R8DwStX@l%K^bxjUna%{KZc11
znKH5<x{d`HH(_?Axo=POd0yB0fU{Pm`KyxYyzl)##@;ik$^QNF#m9nzh|;C1ASeh(
zmo6e*dJh4VPD1FtivrRG1f=(pKtM_e9hDM#2_ZlrAfdO=d-3F%IdjhW&&+Sld2_FI
zzsXA0O1}5bwfFv9j%?jFANqw3dRF;{p~m_n;j)xCl}JN47L5~GT=Cq4F6$G)H4%fN
zaYB&!y#>nT6j_u9@L(dz6c<%I;HL6>y7P_DS(1|L)w7cyRxS0_@ylqOc*0U0@`+l!
z_6@{Sopw9_r_0ddRe$N)eh!>Xu!Q(!JLbwT)86G0>R^M!I462Uq-wY^$rB?U<^FDz
zRqWPZJM%T+=l2GDvlL1a-Xu*9@qEY~R?D4K#iRZ>xw8@2B2eU_BIw&+vcobD(M5hl
zcXIJ0f6C^VW!q*^+Fjwui84aXLfx+X(PtWzs3UWax!(A6pABcB+r$n$<V4{BY!)Rb
zP@F!Qz3n9vTE?E~jkW(Gqkn?|!jOvA;snx^^~g*8w%Bd)v}qF^Eo>ayx|+2@w7wat
zy?Kviy;Wq}f^xwnT#7M@GwJHwq@fKN8`v$~^I&3<uiRK|K^AmL<iDE+xi`+~yoC=i
zCij`|r{$bAr-EkpGYbj?BUTLI_F#56v(|L3+)491A{+~UAw`ECbSfgx)FE>^?A2~P
zrPyv`xp=WN5&lF|nTX%s7TW7|{SN#~R@1&<<|EMyb~S$=Jv4Q1r=00Dask*4UC@YG
zWA}S%8z;B@YIq)~Ued&dF2c#ePzy{W@W2>Khhy)=Hj|0fJP#wrqwlGM`)&eL2}X9T
z!K}uEK6Cd4AK5&cpS+wLSC(fU&eVZe&2u#g6^1a-CKT(H^>|uC5~?qx&kb#Va3w9w
z*>m)@n*9!`vj&|TZXK19!t?Ebi#c<E>VlkR$F(a|?<I(syugR&Hdzc9<wzf?O-Akv
z>XmaHi=;AB=btSVoWQ1m)zMxtiy_~jz(3fFdvP2+zQ<L=b78rxf{)s+hA<bpE-v)w
z3b#{e0)PAG!;6je+<z|AG6K5M+uIc`g7c7?7kZmtgKt<LR=uGYs2t`k7EmVm6W0L|
zk~hkjJY{)7Ym#p=>$8=}gh`N=Z~r!gXt>uQbcVM{N<ex2Jqz?-MR039VSvsrEM0;B
z=s_k#*(BOCg|-3CZ*?OPRuQND@5M+pNzGspl%x!+5p#U-pl4j!kJrx(Vn5Ukr!j6!
zhJ2V)4FanZbxj!29DqFGsuHnCne+|gZ9A$Bll|(vl;9$P?t)UJXrB^QqSlQ6RZ1Mx
zjm16|JNEKJt_30nCXvn|_Q*|YA!yrwUyufwR1~kejcXMFTBeCDvva*4IK61r5%<Hw
z_ccCYVG7v|+xqK``?VQi4D;G>o4c*FR+XI01u2d;6phQUCxYFRYeoyn_JG2NN9Ff>
zsC5BPm^U5FdiNGev2eeER^I4~(S;M(z&q)E0|h@;w<|Y29^%9km`bkhFyQ;YvpKR9
zuG3bSmph8p(D0Ey_MN!OfGo5?Lukd>B=Yikf4GG|MyRSk7L)V-xa3B2iqzuea~}h>
zVSxVy>TKjXlGz(4L9Gs|29fz_wFEgfhpxLF7J4(jdLg?eVV#{H3C)Y*fYy=1Tct3%
ze-+&St|rH^Okqp6{bJ-Asm1jlhj#yVzn&kEpjc!Oxy<^s|3D7=pZP&T@wgx*3A$_9
z<#<Y2IaA;tULtp!(#gjl_G&0_=3%ID8Vl2Y^cgK-x&1OXe~>~AvBXsMjkpg5-K_o`
zyO@k_rFc-&xMGwsSp~Nc&n-PIdLJJ{8jFlZs<dKCUO5IYctbBV%fyShjCk(={~&Bn
z_I&HA?oCfkXZO=s+q;92`M>cAJXpcF<!UZ;Rk{m<<%hM8+r}5ySwd_rz<T*NJ0Kid
zVKsK)MM#_LZn+Q6x3<j_tIZ|e)>bQ4lilm655NXFqk`^=h7P*jrLk%0lxdqqXmm^D
z0N`~mZYSiy8&pk88ab)QgTmdUx=#Eo-UxTqlMczNlu;`BA;)d2+aAo6HcN}79&eoW
zN!X4B=)u&qeP3jliU&aa_{}VKL)SX|?cRbluq!{&uo$(?<x~ad;Z!Zyw^S&ant8Xq
zeB1c2l2Q%PysKR2!o94P3%Z`gdp4eMAEa5nzO(j&`oJtC!_IQSMg1To(t;IiDAlq9
z!HIV;I{8oJ@{Q;6O&)v@Q%;lDui44Tqz*sG8wdr;&mL8n2LOfDxSBt{aLpQTrvvBO
z)=u7AaV9jV2`n~!?d_I&Bzxzu$|8%-DuKCTiO2o2GY-m!i+Gyuy6APmXQcK7eK2`&
zqNBjUZZ}6bq`wtq^{6_@j^pL^CAq7^+%k*nB8;wY4tet6S&fEM-O0%{FC+~j2VJ4}
zIbN4Q9VUc*qj!_W7ER(A7O>UJ;C7~2C2QL6n~{21498-tKQ3c_o{73ncfy~s1!?-N
zHSLXAkJY*EY7+a88SncVYesaB@4G$OBH6q#>)>RJn7yhIlg%*H`YfXpBbK($)#xXo
z`j>20Yz4Ox&yC@d@=m2}W|fX7DV=rWcy?~joxRSX6-|QWmzS&Fa!PgOTwSgv9!)KK
zXg1^Sw4O!jH`M$?WJmPX`4`fyyY!nOLc+PNoCn%@Ro3V_S&^Sd&pua-AyulrcM?i<
z@wuiEzkkOkD$9pokh_#RZkJhvYm)w<h#Y@BhF4LoCa)*U>_`<!yW326>cp3B=mSjD
z2;ma6JtL0M(;3RAZ*%6<iFHS{W1a>XN(Oa3dt*!@67DL;;Nf=hiUZI1?yicCjw_b9
z`M3}13)44(2nkpI66YlMA>Hvv&$reT*Eg-`ZdP+oo6y}9(+>?)5T34!P%DxAE=cxh
zH4JgH1@MB;<%Tji-Fs14cwlaZX<I+o-pb~b{w!-Hh6-EDUlLwaVw0g17xD6+oK|V$
zhSj4cmtS+_D)vtI*J<;(j(Hm8IuIT>?gaatVeMQCY_tX(n1FvRIh_+45WPP~>)W(%
zkfRY(*tV=68oK!8jN<6!i4VUi!OD%bJcR`?5tRWZb)Y$gt29x`+rTq(pfAkknkj@_
zbH?RJEU|d19|jCQXfC*Da3*6vv^zFRB`DYyh317+G+6v|PH{LcM%$gDJU9*a5{)pI
zP-j8hF+QH4Jho=f1+<rToAtd>vn9@T9aiMtT{ZJ&(k=3Vpx@0cm;=@zKEh@ll2IO!
zdl4)0V~6mkqV?Q->3L>f_-#!1F*6pl5z{2#gumq{nP+=9&)=y8b~w@mc2s&Cz;Y;e
zfKq{ubVlba*GYJo-b3!DOKTKBup^t>(TB3R3n=qBjQpHe1lvNwSp*oYZF-J$95KyD
z%-nuKkOutOO}6ma6+Ft>*$%}nXrY%hHT01QR7=+oEU$h^l7_!r6K$ek3sEe53#}6U
z+&p!*TTVESHtp9!sS{SdgT(!LIoUx9q#7G~%LV4xzml#E3&_|j78x(SXJyV|d=j;5
zW|XXtsUnJ=3p`X5dhj-~UW(Qt(*<0-CUN_Uu_nQQ6p4FakOiw|Gf3{KWZytBUJ198
zUe@C`&S$Ewp-s8Ln?p_-QWbC#uybT)_KLf~&M2O$>e=tlXKr13(EdifqDpID?fHvi
z#+UtYwacpD$dJT3c#E@tQoux2eNj_TMk%hPQ8#%YS=w56juhsQn)jMxH3#;2hEST_
zG8vi@<Iyuf76|SJzp6MDWqGx!u*+S|oj1^h?&h9+Dx30Uy$utf8^B4DNv6>CG;QcK
z=+&s98&&1V+XYOLlray%B}OXlbKH$;Q@r3s4I<(ty==c#Ly$(dv@_Qi%YNtbPTkwP
z%+#+7i;dACbuyZP5CBek)ojTx<53<pkWiE2d0>`(Kfk|RP_fLy7ki>YpVyv+Jd5!?
zqMEZ*udVWR^L&P$T|NxPsth`{zf71GLr>yyClYEosE+B3J=@F6Wuo>uZ;<RpR#?}z
zF1K`Sg^#4ONv{8&EO^`I0dzYkajA{3T0%nM2}_|N$(tu9@X0^B{aIt_*)SUzkKn&i
z0cn&VZs1z>JJ}y{tL@$C`x9(h!D-{YoxkWdU6EO-{yuMrH>W7qW}hdey0OjBV&#Ib
zA^Gj)R2-+|u)O-uUAg3=ROIdORH1HJM<~shSyo9d)llBni_$yHi~FIn39YYAf?twi
z#otBLvL12mlPWvlf?lfVpEGw@+)(<5@a}5gmQ05Ng(flg;mMqi*$kV84JhhYK#`+|
z;?cQf-C)+2#CUgEZO~c&d`NU%dlc>+b4pRNV=yQBkLps?j0mtZ`{^N<PgN#@N<h_J
zGpT>0?(UUUl`sExpag+3$Wk?az3+;lXQnxs?POT;pfXzQ#n#Ag#64TD#xv`ka+ES*
zh)<(D{CncJG|3zO%dLjorgXwtpU`bqC#GDC8I%?(PM()qTQa?MSr^gTvq=vwcte`a
zk2~<z#v{&czuX7?iX_gU-!Dcd@*nQCb9QFS+2LfPqvIu0O*)``8A(koLkbck;^lvL
zjx`>tHKnuxASPiXbfeC-ro#Q3?C;BtbyP&CFhfBO_;*fK{%9;0*D(CQxy9vbxqt#^
zt~f&*P~g9p0%g^4BjwVbalpGXYyoFF(z9YZU3|RJ3}S2Et9k9s-ka8qf6iR|(JfyQ
zA1r#CzCT228VK+PL`itUvT_POJ|LD2d<<ar2&}_)=zZ|y*)d`-nzn{_pXyJnnKV_(
zG%0a@nGjhztP<Fj`B>zdmk>O@hFcD7A}P>k6n}OV1dX99fC!lxc2I@<Gu^e>VodD{
zMo4@<VZI{5LYW;QV^K;Re|2;|-Y`k`$X%t_;5CwUmH_|VtPJ|>(qQjUp1Or_of-e=
zEq>~eks~gqE;iFJ2#AZA8GnedBFqP1pt)6OSnQ7>kjXONv11aD*7=IQP?8kATr*s-
zUJXup=Gvd(@WLdnO%o42-JUH+<qArkf@~r9I3a2y&MLZ-$paKUyJFBX-Qw57by<f8
z65pSjw(|0l^kB$$?}b*CVc-=T)Q76E49R@Qdp)>ZSBZW(Cr=-#?~q46`v0+T|A)P+
zfAQ-b2~ipH{rNYY|3iU*e9Rk7!fujSwf}$ecCa+P+atB?WP<?f`i79G94>Qeh4jzg
z6VoWj0cEK4$*6vk^TrV?#~Y+*lFH?_r0gnLC4l%EBl%R`uZdy`J6c8A%emrVFZQ*M
zX6$<6mPMUmN&>qUarvz)56;jyNo&~}LqTZqDC1uuDEzLyJK@Hv4Yu}jdwy;5Pe}8X
z*fJoN`xxk$Fi>5|XMIPO!btr2)1NPz-&x1{@W4S@@3sWAl!pmrki8)eCQ3El_Nn80
zl-cD(ggA|emt(BZOAFzKD#5l!a7+gcTbCQt1-Ne42zr@TIs<gZbK1X`eO1ZQ;pSr0
zoSdDWSbItQ4L^tdMEJZoNyWGIJC?rvX`jw=S!G%5(UgLonriOOFIJkmPkvJFezXhV
zY4*l*vV~Hs|Ea3t!ib4QJhd`jdfnEB_*?{GngThTaCAT_)}!oiEhlCUHZ@&sBeAUs
zx($)q;UDUMg}piwd{sdbG}R{}3h}~}mga1G$dK;FIXW;}{!El}AW@4NG#e+{y8Ho9
zDpE5^$|c&5S0!;-$L`0^9$Kr{il|oH(i7T?pMv3nQy(WB7x6XVPR=LGJ=Qg;+NZX<
z@pF3vp6rVg?`%g_z6CIBi*?ogC3_v-^0UuJ;@rHQVo~f)C3iYMooMV<^&>5fAdp#a
z&ckRx_n2jW988jR!X@tnBuE9rYsp+XFZt9;$Z4Y+i;4~RvR^J0H8v&rhN#E<Zu*eA
z0k&)+PeiHiUx@XC0i$i%MUK>|?Vil-1b#RDA?0L>yt+72Txf-MopwV+Mpt0G9I7YE
zX>-41mu5MUy<mc!)WoOtdr3{u4+~Bib6|f^Q;lF$4`XuMBW^q{w_x09>aOltb+)})
z+d70duYLAQ=wA%ECz9~fc*v-1dAO)h^9=(Fd!x&ItDM=lSmA11b3A44Yu?Hjx~nH`
zVP<<{YCKOyJQdbwQxcxcfTe0~BKIwXVmz-%2m*%$z`^jH)oKS)`KUB_;VRi<*78Ut
zufkdhBc9@=EN53CE;IwEKn^1{J})XBza8A#GVr2u+$Mgm2o7m<`R8LoC2#u|uf<04
z0NBq^KUYEMK{@+_(WRDlHJdlbq+vQ^;{5~mgB6tmp*WYiq0EjQv7%rDUGkQ%d$XU`
zEIxwo<^C~$is0KJwL?8!C2o3*Tsj{4Ln1#O)RtM$OBDMkC*WFqx^sAwfo(~|^`J>T
zdih9P>s@h5rRS_n6#>W>YJqBmnBfQO+5x_c8p%tYEjP%|uvD;Y_0`~=-&souvji9*
zN?>0Uy%}>-<Sb+cfQ&Xgnb1ZE-{SNd{JGSsTBV;ts=!Jmr&a{i_By}V%y>&8N1951
z%2*X0<VJVnbw8#wU39PvO5sKEFaM~}rD=|vq+TJ$PrRRtNv3R8U}f$i7C*UiDCj;f
zZmf9TD0>4fBT}|;H;M~FRdYFQ;ya9)ct1B4OpC-nHd#XTvj*03fJ)0wQu+T8xj{bE
zvWah3pU+rd&~2OgOU9A5$I66Ra5*tf_|oNGY1l~o@T18?^?)*gO_MEWNde4o$K;75
zel6L;nn3kNQ@1E-0i5@!Ho04S=i>~vF10GaJGKAu;{MB1lKpIBtj6&f4!4UqT1wpH
zFJxD${2A;vb4F5ry1nd!yt>-@BJ(bx6?KmVqYxTi2ruQmi;lin*01u+zjkdh(mJ>F
z%&5fGoJ4N~rcZo&c63;obZ%m<x4YQ(!syjBBrC*?(M56FT)V^a4&JxtCrZ5D2>$hp
zcvubzVf54+%*Gby@@bCJ8m&KQV+PjndU!7{aox1NLu}SJp;J*0!3;_IQ8V}B*(t#8
zP%h@m&EP^ah(00%EB<2Cbo!z}e_;D+0^y#)Dpl_+NU-WzETg&mDSiRncv;w1`qlwY
zyPol_Nfy1fqsXPfd@E`B872;OqR+-B^_NUBvr~&lBMs8-#Sw-;uQeIvp)+GrDbk-N
z2AmlctnJ_`&!w&Jw$CLmUy?>npLCImv<f>)t!-SGq9mfg(OS66`8r~I%eo0kB2ZDT
zVf$W|Hz&ysI!xizgyR;=*_Z1#4?=|56>jCXlJ{=|7Y`Ds>L_kR512K=-;MyVO1An5
z1p0Bn<f)mvk5yHSXmDeNWs5Tsvk%c%k~8~yhYH=uZ{5+HZ_FEXa?}+Nh;y>pd@qhw
zI*$Bc>xJuN|4!GMUMj6JxaaEuc2?@Jze{}p)BNL^F+b-lG#X%BWA3XD>3F34%ZZ2*
zqw)2jk?8ZQIQ3`VFY(Rv)swG_V!i(<znp6Rdr=}V$M@%UqD7a)XQTCr=M0u(2>nmL
zHfyBoxU5U*11s7c5`<iC<2sAeEB^_ha}C#C1t*QSJ(@XT&=yo0ohMBlN+#~vh4jQ4
zk1{gf)uiMQyc-1W3k=%1n4J_-t=SlFo_H68kN0tOwm8w)MK7=D#cLd1{v~_ybhVG%
z<q8p$@`pq@G4W0MW!pW)as_z2c4EB36YlA_S@pfvN>vqguGj|NXb)>@R%CrdTdxJf
zlOrHnT1WPo03kew&Z>tRgr?&$KAwOa|76TrJo}fd@U_}$Mw5dL&9g^~%*4{kz0#T5
zccr%m#p^{#bEq(7i9yX$iGI!U`<XELm%5}W)D7h1^KV<@EUJto*We5D07C&O&1dK*
zvSIL;3slARtwzi^Sq||wiqp_hWJ4<v*4zsK{I<~(D&vCGopP}s8hMvIY3<AeKUvo`
zuFRdS79df3uDx0Ruwi2CZg>u5oUIYa6b&uqv&-ApR6o3CUiT%D^&8G35QDqTyh-oM
z$7pa`2#@09)cZ?j6jUbqZXPk0;zs)M+U-j$w@P$U?R~QNKzMB&aw{xB67&6Z1xE@3
zc`-o|+?V?tfGXzGZ1AHOsC~ollj1ByWT2r#Fd%~?jl2=|g`x12kpvOczMBbaabn!|
z(5tHcr-JfFAH7Eo?||RlS#&8&V8hSy2UWzgk#{j*f$=Wwd(^?Qpf9eV0X}5b6&q}z
zKf23r!AsJFmjI*cwC7|Rd{<c^GDv0utYU+CC4-h>&e9%wcWV}8#>}pLP?WGORsQ8?
zky|YM0`UMza{LfNPr7%T>G&d!{slDV1dPmLZU-Mu#r;lmKVyz@|4Zf_hk3wxXh%X1
z%2uLaq`rgdf1-Z>TWh#BYg#%M_)F#cv;Q5<{{L~21#n*fXBc}1@qxR)&J9M|+uv&0
zZw8+4dH^BX279;Ep+9vkP=q*LCnqi|Jfqu_HoNxrjB|q=Z{me>CGNqxt8wPZyEt{R
zN$cV3&`NJ33H4dyei~0WmEQbf)1E%pU$Tc4t?GugNL;Cda|g5%6VE+>2kr3ZnVRrI
zyZnf7uNxBgX!z&5O;+d;gz}OIELwv5ISJR<>dKt5n=zkg_Bivp%Lj7?HO_RRlgVsz
zCD~sxvz*}Ub(Vg3ROn&^i?aM@L*?}RxDQ{#yuY9h#k6(XUUn~BcUT2IL16A(M*%(~
zli(T2v7sva<-yvM2Z+y9#Y4w1BZd`&UC{)Ahn%cIpou8N?ai%aPsyjQ>4ygeg~rtg
zv64~|0-Mr{O$&us9!er;;&;`ZUjCVk;`J=$MnhRDYbtKb2Oz27oV+F|joP8uhP#<!
z{+gR<m%tOHU(VYH6;M<*K2a~Q>0!pwMG{a$B0anOW5&Y5sVVW!J`aB)ve;|_ah<Ch
z`;S{}Eqej9o8i^RA@UtUa}^}s9M$FaBJEnjz#v9nz+?XfhYV_I7<n=VKUR53A<@_I
z=-wP)O_YsZv?uOz?=+1ousY|{#<p(N8zPeN_bhWlhqv9{=4nRq%6n(GwP;eCByp`;
zT0>yUR69M#J5&UlKaOzOtaDuaa1`bDR_1dRS39=y0#c=EYqR83sJvgvn7{CF&|WG!
zV_&pbbKF0^sj+d_xdb~R@M1yIF{DLy(6fmnqO*&CGkE}iyeQvsSXQTD!Z2~p-S3X;
z8V@}!IvVtD<7oP%TI)f*O+<bfHU7#nVqHltM!<Y4I+V?D(-~!v=D(0FNZ_RQjHxzl
zk80frZ_6MkQ-q2QwBrd)=2oYimI5wWZyLO1J|tLnBU|YIWb&CY>?$KU%>W_&EyAvV
z0y@?>-iak+kqe%A(J{|c7vhY<orjK@VORYsUoMoB-pC;K3U%wBHD6l~T8H=m8g{t|
ze)knhl|x41X`$rh6e}Ol_SnPQ1Oox-XWN^6m^gNmDfV|8`~7+=k)D}w_buTPm1H?U
z;uyO1SOM+%mn<?vp|z1>!hYp?J`WD&YuVw=aj(t`BdB@~zFXw(1=wq6zR+nHM?8mb
z8gB^`yAKy1!Ph>ZkBjUrX%1pFLhj2zyeiC_QulKdLg_Bb4=SamHH|*w&CO5mR~b9Q
z59k5JRX|768uDGpQfwgPZ2S&qu7Ph0zDqvh=GqAso!Cn^yWCn?%!c`F){O>y>Uby8
zK1bhbbqq*Lqr2V<gxIm#1k<xe&Uf1>rOL3iC`$_>{JlTTn?bm{oL$V{*KJn#xfHu;
z!Q*kBnKM=MAvcO%miIsO<OqJZawyguMEC;XFGkk#fg#KUwMdQ>kEV&Z3_A^tSLVjA
z;+cNU@nF0!K)FXw-Ze@-gooli?dx?tLm~2jJwEK1BEDm!xtXKnvbJ+yc#^TcHenmz
zQ)IuEF^-6z{=MTDCos+$&LJ{1OG;3fo<&AgiZP*Ye>i_iYNeKkJ<&j7n>1-VMyEP6
zBB~KK-B!rqv-YZ1KbGcNoZ7Rp_?@=QrD@G_kjY0wpJ$tGbAVw?i;b5OCHTCdcCSs&
zN-hUM9@V0X`z@-`=#>0rLHpEUZeDFZ_ujCd*K+^9h&7AKb-A!Q-Q`KNceE78^;KaI
zGV_h3@en6vDp#y=BW|kn(?NsNlb>HN#*nYcOVPC#)cK0@q%7q3Yi_ao7xZaCW!ipJ
z1L#^2EmE0tvlLqCkDz0YBDtA`&0Ltp`f)+VURy7_Q+kisb8K8Ex_%mLl5Q3L=Hk;Q
z5`|g5<U8m7w|OfIVaY+~wWB>Kmn*d;V0pQ%G=CFq8z6nBxO=BdI!x2q#3YwbrfS2u
zVq~i$kG|N2jP+U<OXRcuAbQ&f(}#sRt+%u^wW%76&QbJ^?-v}0W+$Q6NI-caog%B4
z_p9+{J(WB7Si@=GU(lzk`T>uh>_{HNk{p+$SSM6i`C<v*v4stAClz5yNzu6BhWtUD
zhMlYK9)0to)Hwh%MnVTvNTZLL*X!tAASQq!2%THx%g0?_Y4y@+(zh@8gy7v5G^+Um
zOU~|klU${aFZiS&cL(pT)5;CbHb39gQTsi{4^h-?ScXY7OxMUc@A7<}P{YPgii>97
z()3U32+*~E;%u+#k^ACoLTRDuwn7N2{!0=<{(iXy&FH8R=ab9fRzSDhZik*aRyGMA
z3(zb|-^!7SUh`!6OC~H4o15iXsiBcJ6v;!Z`6CD#({oymcNp#)pQ7zcIm_-8zH<%b
zjJZT)md>iUqQHTdChUL7)W`5<594G=poL}gar2t?dCl8N>UQhyKelz-iPI!Gpz|ex
zG}5q0&p=~^X}uJmN)OOe+CoF8nDY~%Hp6h^L|88icMzy0PEU2*aB_Sv|88R3n=L&x
z!=kSN<1{KybsLf)CMoq;2<F>Jw2!_KKc*C0h3_@-(c=JEKQmV|^$~`Y`#=mHBE6vh
zNHQ7F*cB(0*8@ID1m}~2ctYE+pSnx(ar`TG@cV$hmzJ5W)*zZ|9ApXB3f|RSC!tgg
zq<IxD!(4>e8t%wc!^Z7Gc4MGLdG19i`eTj<w9J$j_G7Ed4AbV_1Jl^6=X+!ZGEEq3
zFMby{(O#^?ICv$ywtG?dNasdx+8PR5fkMufS0OZ5u1zGsLBa!0paFMhmVj~9N#`z2
z%9~CjPYC?)Y9PZ1!+7<y0S*ZZP;5p2<LHtZ8u5kb6q4@AM}9q8{-#PL#L(gF90`#V
z7o&4$YlPCqw*(ko6YkwxB$=z|Jj;|;m_f!zpr1?TxZ~CPGMS$_g&SHk;>#at1Xx-{
zdl;+B7P{T^#F-Om<c&!sh6Pz25^z!rZ~7lF_x~@2o0O9yIoDcygJg35H!uIc;wzqe
zk6imd2jpnI3>%Z30(bR^6&t91lQ3U;U1J*!>%m}NP}sDZ@8|E1%6wgdPp^h+liU?=
zvU2zzXh%&H>Wrt-KU5{jdFAQxJfky`q64%7{FT^k)jXurnPq5Z%m#h<SbI(U>wXiK
zw7H)DMES4VeVcY5i2dXuG)f;a*>O5|4Jo;>4Vh+XCln<ff(;d_Ne9Mq%hfZS#u}G(
zlb8h@CDypWw;+3@??Im+>u)5G@W{lyW8m(b&)fx{nzxWZ8~^;Q-?w<O!%A=`;FAid
zx4FuHs!cLWrLn$Bf3}hD5TaocJ?_3S#|Cxy+FM8IjYKas$Y=;HV9qBbDtnK%^!VOi
zN8O%1IXtM`HZa-T`mJ)zzkm}0nuo`jgYHly%xcN>0z;_YAKDv5vB`g|%iTQJht9y#
z@n3ZO7Z66f-0;QUU2Tri&sPy6cun>+EBbr?Bz*H)ZrLy(k#-}K-gpaI?o>9Mmn#k@
zuO#Y;zo)D2?+Nxkpa*fZR2WO1$4u|SMxP5ytBaS0fL1T$i$%KxNL7_pgals-<AroV
zoRB7FqsdO*tC=EAXS~p9R->yzb1!#0cc<(S^3K^8_nyvuv$Pc9w;j`>H~Xf8QSkVK
zA&Iwl0Q-iyzxHcz-Ojh!vT=et2ZxUq{VZ^}0wPV`>qOa}$Z~h$6uwwLV^AYEVp}iC
zv}TVfEjspP6lk;dc_=GS5u<@A-Q_Mi`r&4k)$2B5jc$Qzq^}3uFG(7iZfL*yIrccZ
zTV9y3Rn203$UcsBICI$9YML2(*z#5Hnf3V~U(hmpDH2`V!ZUw*2nX@o{i!p08JAkE
z-Z&)n-WX0IGd3}+RDkl$e-jb)Tz8Sj-~El;N*0MUWrg`s^V_$l*Sv*%)4CeNFz<1W
zt%1J-kp4MFt-DKo>;9RDl9UIX;(tEK^J39;)7$LkY#wI0n`dIs#}a|ftrXF4Qh#7h
zM1^UuZgptVTK9Okj6G#Tw{T|9O$?y?^uSuD@ugY!oSo$<s4(s5pH`0FZESTrTB#YS
z1<}V9B2+D(uW?MwHiBirT02g+{*vi9EMnZg=Px)}TKex*(VQ%p?BY)o20Rh=9jZ+6
zvcV}+eTgWBV|(-6>!@VkOueKTiwGlm`6T!I0md9If|CV%@G+W~t1U@EgNpX0N}vMV
z<LrhRR9ldyC$o5H&5~}X(DTzuOHR3kCk4fCQ=-EqcszvpxgEKp>Q9d$$LVeBL14tr
zcEkcA8GJ#MqMgmQw`QfG3}CuDnDvmQ9osxdePT0(YdN-jF??M7HI(z>gnp&3Br;zs
zWi~upP%Baq`Qpv+bKs5OPrDTx(|#^8({R1R7mzgK;InHg=P@}UtVUkP2~K*8Q>)nR
zidpJY$NSS1K_(e1UNK>{<?F|MW$W6<o*QnB`jgRn>zH~}m2zrRhK}#|F83W@KyM<6
zi!xr>W^cWvT|HulsVvSWay6%&l=4vI{4S~<+qSEfkQx%*u=`79d_(cxTqqnW7Q;ra
zed|of8dU1ZK%tPNcs0ew*Y^#WQ;N8i*NTBSHks!}eqf9+M?XlG4t;%g9hllqFcsVA
z4jbdY?c_q96=109EGEb<-7k60o}2sapErc2LFx|I>4M_c=G(jBNeT_fhNBXw<mr>v
zg3`*(;sOUPD2o(>x6|jVtpJrU8QvEQjK+MiDSHcSTGE(cww>a(oEsB&+gy&#5H9AR
zYw3@pUbFbEeB)S9a#a2Xp;b3nu>j(4>@rjxg-7&F`wYwpv8f)z;1BufWFT?kRoIwk
zB&?c%$^AL6;GDX1cXr6i$LyK2iOA<3(zUCO9jUy}rjT*)`Dk`Jp_yRB=wx@eD@Ddi
z#j$E1Fydg)Wvzmd80zT3@W;uFpjA*eA1YXEk4wxl`eq73a9fQ>&hO^%`lflQIm0$S
z>9crirRa*JBv-TAFAoBM-h_<$W{~REphIjC(Jb;_1HA|@LdoUc5W>|g0=`h))Az19
zlc_9|&v@Xb^{{_gr-ZeEXrB+m(gfL+SF%`TM?*Yco-LID(VANh19xwXeVE2Kmch?I
zG1M(tjy5|t(5Xu6Wj!~1<h&ZxoslMf4*Un47C!_L-WqjR^cwWP@$wUcw%G87Q5*1A
zZR|1pjFo0)Y6m~G@t%6;uri-c(@s`JeMv@|8AbVN7~Rw=`7xICgc;Q~ax)kSLm7x0
z>-HuzO|yp7n$PIn`6|3e<(C7!7O@)Y%I0x0VLdP1Ar@Vjn+}F{?>6=h+FOTsy&Xwi
z;q;PAdUr&q5#znU3w>JC7ntH4)>m7;bv_etTowT-`n6`HiakveGtY#WxTL<8W=}D4
zNks^AS{b*#9YnU$h`reaeW6i?p~d7#sV*V-lYi8nzS*Pp_Yen>!)RH!Qf_=ZCLI=e
zZ$%I`_+I1Eq-tUj?1>T|6il<O-s+9QlJ4=Z$DR+MP8wkdJD1F2BN@`UM>T#6mo`Ap
zzW%l+_M&yIQFQq=e(f6y{6O=+l{X>%!?Q<lJs%_Xl5=A&^g|RlGuUf|W420om*%2E
zUg`kikZQEDB(~i4C}{e0`Y?}LuFWNr3HDLV(!Uw{1D^tOAk;AP#36%b<q*87hF!`<
z-q7o}^4>;IYw;?*Zw+>VzmapvM5x{(9+pknao=t6f~Hg(&a(ty8jtdnzAsX3=<(e;
zI)aV-0OW47R#7PkTv9%$R_cW&<)QkcQ&^q5bkoHbnE&kS$M*KV0##%kLVcW__)TaR
zG38yxG`FCV(^rFhNopMZ$89cfNhWExpT}i>heb0g>hDc0Q`LRG^ETBl{UrmXaVI}f
z*$|`C1sXN3htuhrhHjCTk&Tj8hb4Rc=R`j}zmoR;O){zI=WXvvn|}T$kM6%YcmI9a
z+<#R${{!On@47cMshO(`q#%p7lWnMgIB#qMfU$C<Wm#a&M>y?bS9Tc^;Vo;G)voC|
z%8W@1z^+zh>6fP0=k77vVAg<}&v?i1w0>n4DNtF|G|Nal%RpHTx+mHcsrw+$$JYQo
z`xJoK54r}q->EyqP`(ALrfape_vv`V#@ExX4|Y@e33gMx$F12t>%ClCVNa=`ZS2?0
z3Ggcms^O^fi)L1c7ttQYbTC=T833=u-pqM=>(cmMfAyrQS7NH}#~s`rN7@>oR(i7d
z>6k?3^R){@^6<xMVwCBN^Q{3C<xuS-Gn2%438>EnFERIF?U+uAt>1msg$8mYwl>XL
z^;r1A(02@ovS)EhNapkTxn<PCQVehXd{%MZ(rd=sz8N2CNItfL<u-Wx;+Xbsd%<yK
zp;3~wiFHlPm-p}7X2wccnxiW!D&jplGcxN0kuC$S<+F`n_%k5Y&T@dZV;5}@>1mK-
zJ5aMV;1<hsu7ZON!0n5sr^=H<0v<h8F)xWyANY>a$T$<oy{z{A+DR562fy5gdonK<
z-%4<<KU#b!?OaqFA%Az(p;Dpo`KRey4Q$p%wu!~^Ey*(eL0}1K=I-pC3x)FH76G<j
z+ASyR+WD1&8h!$#W@#O_j*VTXTh7NIP(7kN1O%d_QwbCbM871z8^pU5*F3Ais?liL
z&GhlNAl%&`Gc^+O%Kg!J<;u9%EhNIuVWCs0sqCLt+F!vOks5njv3BcMF5hw2M*PL4
zn9E7ExyxofGg)3t^R2n3zpjIj3l{zJgEXk$59%0{NS)egPb8$Dqu5YNKx+|GkyjDp
zFPmk-QSInLjT2LmPMUc(#V&PiTN|-+kT#)NzOnxqY4Mh6XxmDf0?CzX!5&`P_DzkC
zYJGb=;YXF+u0%qC2>{$H#CTCttN)3oQMP>4V1XXG+?13o<-*|m)8ZM;%Si2*OQ6sW
z$5?{;zNS6u4Z|P_R%q+p%}DGb%}LHXaS2ffviokwdSd4=KCt>u`@pq_4X6Qft^j>6
zhS8_`&BqZ+RI^119%X&O)}5w2oU=$7>B4lE$7=E(Ao%`5(;3AGVgxI$V<wRMTe32G
zXh_;H^qa86PG{kZ5Q=U?t?zo~%BbS0pJDEKZyTp6y<Fa%8Kou{;!nbq8$u2OZl^6J
zvtGOEx#Bdgo4ecV=2<(J;^U1!lKJyimP@W~?!m>Iu|Rni#T7nn=J|i9zi}wBm8WX;
zZDo@RZd-KDy)c}w7XCx@^ibo6PKWDij}K*b_!4)Y4o*a^IEs<p{4Y6Tio=R<NQ(+!
zr9uoI(8YMe(L0Xpo0Uc*>QRXL9AaeC$?Q^UXAW3(Q48EZbq7ZqX8LoEmZ~Wlap#I{
zZE{a*)knTT|0Ro7ZS&J^wEhW~mDd>yd^L1r<kl;F7jZb>>vF837aW$6ij>evdSlux
zb%*TX*(07qIG;TA6i(Thhvk^vzCq^NCbjnO#^p*+K5iRSImka8h2w|BsIO!;$0nd$
zKqKf2YOb_dTO7#?1q#lldOv6WOQwHnl${FgP0&O-7P-{{Aw8UH<~a#}&eP}3vMuZo
zpS{BGJ!~tebyHo7aStqT@TW@Yc>S5>`@$7HQ_TLh`BqD6YRK+}a}2k51Sc0oz9XIY
znY$9JBgDDgD<C?DVfc35`EjNCWH^Txd&*rdSinFwk<A)+?^ssD;K+N=&zE!AV_$Of
z`4P^~dihm(VrNI_3_F4_R4OftGr-V)9c%->((yf2L1%+qn&E0J3@mQR0CDf;W+MrS
z6CWN#tT;X|)*m~b!5cJ(g-zp9R{6BBqO3j#Ma6Du{0jICY<L(JC|nVwaKVf5p}=ow
zeD7QCD!(=nJr;)yAoW>t?DYZl+k4IyN?_VlN1_=8m3m%Pu?poKDduczyNCZE0a<8o
zE+fNjBhL4D$%^edwjr6qy+c9;pfBJS$F|Y(&H4@r#pq&QL1$hxU;7(!3`Rn{2GYZ4
z@U&6$N2_Yta)$t~c5Pj*FItUxYfw!&Fw1~qzS$rxzu8dKdd7M~liu3>*QdMeDfb#n
zqqcgc4@w+ss3ofCoBFsvm4FMRS@n$U1xuREX1MrzWWmK_do{u<aF41O=`TwtvC*k{
z#sQvX_*%KMtAP(KK$lB>rOYXqzI5JH>gB5&gxbTm+~b!h4uIu7AjJ4O&)wS>;5Odf
zSGI-HS5x_Gse4hJdeURl8B0=vo<>PRPgaUWd(^kS+T$G(V_iOt;nJ5O(k#UeDFam>
zZ07YFExyCjsEkIo)&}*P^jg7h)zl`(*uw9P-%US+>k-vh(+|sviW6ZEnZ~2y1a2C8
zi*bsB?l9>*k+3$jfgq%Rz(f$DH#O45v9dEM!CRLb%<Jm;fMqXCL`{j+xc6GEDUnbd
zUZKFr9Jj7)-XTWop=v|_&`JN;KwE+8=pIP;k3o7FOUAA29svQ0gGwKxuctv^_YLB5
zBV13jPo?-Kt6i(B_a8S4w@S^p3*y;C-bSgj@0x{uWiKMj2smJVPeliUqE|BmTa~t0
zp$+AHXeSM3Y<rgJ4zmQ|dPca(Qa^-}$w<5<{!FiEPTn&4uwto!#TB(zs>HoC<hC^n
zdu(kX*}867PpxaRA!KpLgNw|AHF1r{J5Q~kE8ytZL&A^L#j1yY$vzt7I$rjy<Q$k^
z8+khRs^`@iCUHh!bGh*^Sv>@J3(Bxs<u9PkE2X-^UbjlTHRr#Qx&vGBEky<GTZ;iF
z7$Q|z6BlGKv;JG%;B1Yp@-R1z%lwJuwRf3p)LG4>d`hX|ij1-B(XV6V?wYoyqNhwU
z^00J(J3d~L5t~@TXOg(izMO>|#N^02EOVx=Y(6CkoXCF{8ov!CrP04qkQOx4k)1u4
z(EBem(*H^(9scio5p}MYm7IoU|1bT=d!%N<e|(XD*C?lkGNB-E1{*!J?WlA$f4g;w
zR%!&rUcOoAk89kp;NP`0@YJXHBW+x^o@V+r-j(mSjGr$qqTy%#_G`Y;LiW;J0KjBL
z;Nh0%$<L)sr2lYwTx)I3S0Y}mc$aNImz&a~F;d@v$mQVV!=8e5o9Rwk%1pcQ@}cCG
z1@$ha&J}Lrs?Y3L{q<?k##D<TtTJK#JgPR`)Rk#s$mo~`Z#e^GMGmO)98_k26_W}D
zRSVt;zBTF&V1~@oyeqqhiM6rLGO&96PUzF#r`AhQOO1#kA4IyySvLHMpgX0s0e1V#
zHv-ptA8#C=`nq`2k|mXZ0_%>n{s9Y}rtI?Ca{XZY<{2h!jLz7NyCX~BrlPT3^(ggd
zh&x)n{h5xgJOv6{Ea@7T`t_nzhQlA7WPLOK*RME*zhtz)`dZH)ib%J8mL-nO1i<V}
zwGdXX`lFj?x!f9EP|}Qz#BoN&i8v!hkY~aL6JLJUJSBj6C5FwzM1x)$azAj<^UT<^
z3`7aBs~-(a-MuJfa|80@XNa+oeEm-5{Tb1C^!5Ae-7V#v!Oy_ZSRFZL#cJ;}e{5v5
zjA|Mcmr#ADWefRUg$tFh-7&D~9_)joHqbTZN^DTl!y!J}b!j|ljbWd6>9He*Z3-W%
z-Re*MsfDv_U1f{$-lOn{l`Oa1_q^DY`E0?CozCO!74=+i*26Pw>rpFzR|Tkwtr(Wu
ziX`VyN8i7)0%>l>H!C$FMj;W|CEc7HogR&uyX91beh5{DX5TTdFcgyj_fyVan5>&*
zXJOcz8j|S<8$PJ$m#p#<6wSpe>!;^J&x|eWbOEfvUA_2hjxTN8qR3m0nOlG;c`nKx
zg>78pmx@EG;J;);O<#L9#kVY!Wnn{U1szQaHze7XKBT(9{Tbp4rdN+jOeeW&E-|v2
zwu8@E8p_~z@r)f{D=3|imB;a-EaNVGn#CO*V((}lxU!rS{O2-=*CxG^JO0K^ox?c)
zT)1QJ?Xh_c-v!&b<NVmby#TvU(xx-cx#Uwi9Zr3l?HfEYA5SLC@*Ed5krLWGY23mS
z!U;<zI%3Gkh3I1wTSr<B`|8*~caL6en6y3GwJ0TsG3-8yo*KU0;%2H<5VkM<m|R@C
zi(Q~~LZBFZ@cEoO3x1X8iK2df)a?snw*0(lH2<`L#`0aM`|rWTT)2~?9PQrL+p*py
zzJzpYy<i0Ribj*OW3E|%S@-zEXJ@y=A-=v8jlMnm(%WK1y1~)`QxBe`r{btjS03UD
zccJ{A-3Z(B=WksI$!Mqw!KAVv)z7J(nzRe@<?ue-oWr1foX7|%GCGQuj24^GuCB0+
z{v1qO|H`zVFoGTn6y$rEWLETA`q_)YlL-y*pL26P!t*LhVA(nJ@o$Rm>gvXfjV<Qv
ziDZhuWc{t}z0s@r@hb+QWL*$OVnUO=6d_fJge&!>FP?(4b;W<Y&EUusa=KRcKIo?D
znAn}F?}L2RC}kJI8O6{i?Ff$ac7GM6Ei-_B_4F6awF}uubOWzd@ytQld_FyxVaqyW
zPvNn$B+>Vh&j^d5Zj-qB25|P$nG@m&cQ(b9k+#@QyZhMZ6%<rNyM_)>@{pf+beV-Q
zxFw>na77TTPY7Mf|HJTSm#w!M;tT(aqu83I4Xs)FtuC$?3^@XALFF8x4q8i-y;^l6
zgYE7)jxmoxNn+E-s->%9`Hm*->a{oGn%T`&T-boB$AW#TVn%k5%Hndv--`!uX>r?m
zZTf8Lfd;XBkas<mr-Fk=t=^Pj05(g$8F<%7lnMT9#gbEUxyd@eOU&EsXD_6My{;P7
zjLS&&`$}Y!T{j`Om^_3JSbmrVRH<t*%Q<cq9V{2v;TnFyC3Ta0_<+j#UthsqC=b{e
zHQN^A{Ef`<QN}ZJTS{^2?z}k#<!a7czy*4cl61AkM<gAZ_*_<$cWOFUgpAOn@sy8Y
z_Atu%vu2|C(~1xQ;`k{2<N}mW<lDrPmvx_-Vq<-vo@{SH%plF=lvf{geb<%8oDa@_
z%X<%TCaN53>Wt2^$?XL=^zLO5>8F@1BxG<Yvgd3C1j6vEo{CFyFGOcdY^bU{x^&FW
zfnc){P5;4MQK)~Zw%*k8qq**wVg_K{oB7@J{A;jQbzMVIIHUj8^^=A!r=vj+t%(sd
za{*=XO<}3=TaeLIBH!!4jWDp>yY7L{?4SZ3NO#m-zONo?Tww%G(x<FQ2eLaRiExDn
zNT{Pbfe}i4IGEZ~z`64Fw|L&H7(jW_n@)tVGF|b{GCcGXlNtSu*ynpM%h<yId7Q?F
ztzAiiYd^IEisRnvG=DwAD$S($6V*|T$zD2iD=z-|+#tWA!Pi?lHv-~k40!lvi_5t5
z*PlIZ>-6ki&eh$%Ud4{NFG97UTML28{L<-Fa1$Yt@c$l7Y07elk(qtqQSnv9xsGWE
zcwn>A7qwUQY9jQ-qH)oNgL=c&G2o9^z=vhwoRS{_H=9^*F21NSq;bv5kp%~zO`hiD
z0=hD2mo7T?igb<L@EO`7vFJKm5a(VcZ>WX%iYTnZg?zcp`(hSTo|1RxV(IGWIm7$@
z98j%VO8+xDERXZGV%KVqA?Nf{$1A^$EPlJwxSaMlp-L8K7M?YwPA~#HK6F!h%i=JO
z?U&t3v0&D09nf36ZPI(;pM-ikixw=O1>|t+3|)*RmXxO__PJdCsFz@`gZf%Bk@;iE
zr|+TNgUTer86Rax##E17Hn+0`Ju}-Xs?@saF$XzqJEr4bUSi#!%@v|z6ZJV=RXZvb
z&ogeH$2BgsvAG+zPZfh}=DPHXC*zHR7e`6_gj#2mD=8i=(QuQrqx;|C%1x5ch{I$$
z{KqJbbib&dUE04qE&Ei*gSmgRQe<BLbJy<Qdr~#;e-$}Y|LT|mftzH)&~^PU*5yvq
zU|Qp^Kc;8vW4(l&dwyf=?ynu|tNMFZ<#=ffpL`#79qbDloo%%zh}Su7-fR+Dw5_(3
zTQ}bJa(&Gcn6j=HADIR)506%q25R}D@X-2GtbK@rfk27)o{9@2*X4Q@LDSSzkXn-~
z=~;S&!F98}!j#&9D*@lnAf~ZLkAz7yjbguY2>ReJnP5@eY}I5f$iE0V#_&Wdyj-v|
z?Kd7T?()_1pt2}Q(yuXl)M#QkUadCiry#-6&XlXg#2PXw{p#SbT->{WMcPy5rruWW
zwm0;k*@IKI8(;iUNqzw9KGINLYbbC-oYSguoaO$#!(?yNkSZOYaODxz<vOM%;uZF#
zS-%PGfc&=6)0SB)*$x{9nd*(`@1VY9bdTUN!%h+3o=xD^1Zld)yzjyS9&{dlMArLw
zia9}jORPT|=7Z!(<@`}4m?~5Wo}T8W{IF-=SWvf>@A&~UyMu?VHfd)lJ3nQa_5n@^
zF38g|F>d~lvkhsvX|b#Fg%iENdtWruHHP%PU-C4U<w`u1{tb-JEFa8j;vD<zpSUW=
zA{F<$-!BDWnf~oB*@H?vVpa6uTsFA^e0g<<!h`l_diouh-UN6S)uV}(JA3_~W&d#p
zb2_pJ`=&qs+MAl{jF#uw8?+sUfr2@+pVY^z;Kv*Yb?`u*8!)a$J5tzmCF29wS=MS;
z3330|((m@ouc{YdjoP$<1jQXsW>D1O&3&R<C$A5^7!u+968Pd-=@%4P^h%m(l6f1@
z#xL7Di@CjI%?B4&4y38}*73anTiq+v^o$TamwDfV=aQxJWbk*YRpm;)v_DsQ3>F`{
zEt=_-WN%=3lyIHIiI?Y4eqD!N5LQKzyKtAME14$tEqdz?I~H{3Gl{l$aaxuAstPLj
zld15yL4I!}^i9BFg*MfRpx*dLj94!-69*Acn=l-`$r(1LRDaES5<Xs?8sZcZZYz~)
zEcJZM-PS}#>MZ3LM=SzE3!f%>gi<?UlS1s*GJE3ea|ByN!s*r3oQZ)A@f*^P(vZhR
z$7?#ERQ-2rG^j6W<<ev<)1Q)0jT6isKoXqenb&~DP~VzmmA#LBY9l00Y3O*i#IGNN
zpfvFzw0SQnJvkGIsSjoglcZTEI0_b7l$F<uUVx#s*tbX3(kg7Qkl|-nI~U;IY@^wV
zBS-WS;4x0SL$Zof(I+sqXU@;Sw!P!9vhUTwAk82IQY`ed)I*Bw;V`!G0aSzH+xy)J
z5_E}cag-cKw0*rx7744aRKqMyZj~yZXZxS}P$bF7nYZuov0HAKX?VzDy9_D+2^wF$
zY3-Pb20q|)w%=Ymp;$l2=ATGiJ2RG@_L03wPbaNi>mnkYklGX{H5yR)Jk`5E=AY}&
z58(l?CQc%k^z?`(h>y|&a)|sJ0d2?hR86@A)VvWv|3Y!Ej8uI{x5l*{5FJ%S?Mf_O
zM?HysIjB{>v?r7ijgOeI`&RvL>>*XpJZ6**lqz9K9M=dMoQy`YY63b3%Y_ZIoPU<<
z06BCG;qU;7*~9XeI9NTk7>uY?X@P81el=(rrih%ag)bO7rO+4OW2les)|-2?Rm-qS
z1y*8_U{U7d&`d`|Duh$o!Q59VO_;?p_>Uzu?=3^10nWoO@khqovOHeQe~cQVfAO{5
zlh9;fvBFx~^k{miQc3agnNiH^Vprax)IUZgZZ<4x7Lsz`U`3S=J+zdB1{)~q7-hpn
z&61V5>aUr-DhuwB<tqqq6V%3*Y2TWiELw1vOphP^KJ7i009rk$;X+o7<R?3_8ul{?
z-Svj&PcB1eI!@B<j}oY~d~eD*oms>Eh9coXux`o{6fqafe8j<ELZ$eoIi~Il$!QsC
zBmWwsX*3dHYZ{NoS6d9R^Xk}#f63)n@|Zq251*oa+}AK@IL>{4_U#O#e2T$n$%D0{
zycHrLodE1689$y-x23NOe1+{hm$R#p=A|bd8EePE#0F~enV^Cj9l>6+Kc@>T{rmMn
z+HanVmk8y6xS!ZbE=3lr2@VX9NXWyf)|BKHh1(QgK3X`$(l~Rtr@I}~s3W|8p)&Hm
z4;#Wtzfaw{8DNCJ{cPZu*%K0-NsKggn8X%+Ib=ad7TwBE3?~y-rZ)ms0n#T8)uv}d
z`Lwj*w)l6vm_6oRlBO`Bt1PNW$5ooEHP!MsyQxy3UzVA}l$Z0JL9aE(0m@>}U3nQ!
zTifa$B0fkM=mpLoXOBvWWkoL5ub;YAowv1Il9uQNId0CjU<CIL^4y*cjmMM;EPdE8
zOD5E~gmyc;ARNX7;mf~ZAt<1S#V6Ps7I`5;171^EScN_-QRP~I%d)s?1vtjFJl4b9
zIsZ2P|Dx+F<C=UJcI`k=L68zqKx%+UcZjq!Y%oe1HbQDNs0hf2Nq09Z5Jrb|cQd*h
z>E=26Kks>d=lsrj-w*fhdmpw>i|fAf>wClN<%e6lfQP%=x?$|5?U16eg>MpkW98QL
zqaPLox$o-^kc$_6s|U7kflh{*e-NshtWtLwLl@=Oa^(ki>~yiryRCw9wCYkkd5r1y
zb-*O_^z~-y_)3@4ah`p<v1Hj-RP!+W6N9Gq@126$OzFWFKJ60cglc0gj-A4jOU*J3
zMtXPuQjHD+2ac(Qc;NzCT4#@o0Zp3H3R9UA0=*I%*tQThe9ufRFLvO;DL!R*AW`HY
z$(}N@s@ey~n7N&!(?WfTRhv=G#jE@F8eKfif^$3|@FD+Yx7`_;V!M6xpJ)Hi`G1Q&
ztZ_C6z3&p({;m7`N6Gm==Z6G%5<YUP?ElQk{YT{y{7P4y;Xv@y<k7P+$}(eIjFY<!
zkA!<26%`FvYxml{)+V@CNEmnV-{6~hSf;cb6oT~(jo~P}F-){H1jXp}tf;mG@cMU)
z1MBOGBnv%7d82DG1n6)H`b=6pMST5)Ti8LvqV`CFh8v645`%_duvF|}7MGx3XQeG5
zdX91Mlz_QIFxC#Fv`^b63|AHHI7fv4z-sw!CE<qJf^&l-OPp8@$90%2VP8fH-7Q?)
zyQ-bDzJeU(Fh0%Kxu`d$)qLZ0MYE+}#<f)WswPDJ1NDaO+fsASG4BDyhZ|d}5gaWp
zvTL7AKk(JurH@6^D6Gld(|g~N4HFHR6Hrl3k$qyU7~t|Zdp0#+X!73gU;6Dmb2~Jx
zTUB)^t^H4P@53Lz<4^e9FoT-UwZCekH8ZJ}fI#zvrZVHRICd6G#bt^f*@tkFk?$z*
zekVhSmx)NxK<2D*aS|=6A{#q~DMEuANuJJ)w~YFFuj+B?x?@jEi<HFps74lq7Yhvw
zt96wgETpz7Nc#f%Q`M}zVm5@46sCbV=&;L|$)jiBbZa5_Ao~NpE5J~8_>wWR5prF0
ztQEae6!tODU_+12_~Xl2`U5FX(Gv9&FWsMDf-%)jyfXT%<{8>3=mgXiMf7a%*L_}l
zXMlL~N#(D1e+&EeFl^Nkt6Exdttz2Ul3O_#VGpswg<L-DWGSg7e7@UZi<u@bJf}Z~
zb>Fm#U4mi3%fYJ>?k{AK_eYe&CI|H=HZBZruS!CnCg#-cYqdw(J66^3JUj;ptQvLI
zr18#epX76e^|BlFEW<###|1R)j|IEQ_^az_PIkT$nlcqt-Hhpf2djkNud7BbWvrCy
zP0`1hnRAVFI|O5PRo!0RNDjF^CZC+jo%uufpTyo!v~wD2i)9By52ntBXKV_M0lxeJ
zf1G=h<iI2G4gL6Ct=IxihOQm&rhb0*%L03uoq>H{d4?!FE6mYfv(es@&)y9>MRoWM
z^8bits@lA)!q6nNzqRmkEmjt+TBoT9OV+sVEi@~+-kA96_Z9fcE45qj^a(SE`-Ar0
zq<-8`IcCGM{}Ny@=H;EW`~G=$hH$IrP;%TFmd458b>+xAH^n8HH|1r<Ayk*6nU=(H
zRSVe!qedK}+fUbDWk{uMxKblExnUjv?pM{1rLqv-%#YR=#Mo0zQf#}{B*yn}yGwz@
zqm9&}v-3s|c@>WZpbEpXu`=nL{*;g{*IshQlF3!Gs@XO9)cb>F`^3Jq2bie75zpk9
z6n|K{wbh4t7JxFXV4!;eZ5{j3K$ar9YiHdF3@iz>Bh|#qwW%87aUNS+nr~XyC9;L*
zm!?qiZOG*{i4A-xHwfO<gNz5=EzFeP71Q~WpZKhI3@u;PZ|5~cHNmtT_6j}}H2p3U
zwQlN`n9OR6?Apd@hMztLU(h`Nh(48O04E6t=V}F+PoYsOP#a7TnZ*`Ecl(3fFp|V}
z0sUNBAtno*{nFT0C{=WiUWgOOoc;`TWDXp6aWg#w7CI-^b^k5FSGu9SsR4H$OZvdj
z2Gksj+dcUuLla$iAy}c$MW!_q_FkAy#l3vx?TGetwoN1bX8&*nW;niTLnk}t=!V8^
zj6#MXoXDsZO2&U_AYf{(tKqr&o94sJ`B5QynY-xm#)n7;wu!KBdq)MkZ^*)uu&?lb
zJ8C9g^KNKuo~ZIxQu?;WEv+36UQunUb6S?TS?TTwoD)x4>!`Vtt*DR`H=xl~MHua)
zjclxdl=Vg$?>RD}e!OVhGB#LLCEvK=jas05EJ%9=ooMU^JzWb74;X@G{x-jqObfoO
zkAWuC4a=TN^r|w4UdPi4-96hBROeHOlx2Tn&?h?Glv2hoPJ}m=>Ge`YGH};6{8`@p
zWgRGtRVzKu#wf4P{xq~+b~+6P_a~r|J1%=ifC0WsX~Af7qlXohPen|On@U2EkGKUt
zXx8t387Y-5mwGRvIh44P+xaWo?{9?pX@UK&g|ueSqnj6CGJ=?`^Xj;Ws~2MQ@wY)K
zmIFW?V&b0*Q5^eR*2&InN+@l-O^gB-_P^DyG}Rx*X}}3qQ&eb(;3&RO2q@Q+T&;a4
zwk6qjOVn0K5HcP0U@Ix>_%0J6gl)CyQoJ#ho$)9+$$Iv3E?4}jt}vA_k;By93lEvN
zW}guA-H7s;v3V30h3rf!p5t9b_Dp0T^^jjiBYo4~VF19z-+T3LjTzQLh3w4?;1xH;
z!HDhf*;FKEggmy@Mf8VhLofq4gxr~O;A#>DKsqB25b<+pH_E{=t|1}Tui0Z&%Qond
z)wQ2qE*m;|!KVZP4~_i!J$2$7DY@|FQc<LkG!BL}*HFyEi0g0k4`upzY1e%pXXmlh
z9A?RglB(MSlf2xiA2t3+kIQ63uUyS$Wluf4Gy#{EVP`fpcw;Xvv~YjE6?dm-dHQqv
zD<{T7mI_ZYGc8R{3V&kLArbkG@i-yJC27*%;guAW@rV8%-lO5+*TXUzY@8^N!{BL*
z*T5ha5?egt)NTxYLlQ2lJqkPJpSV%Oy^He&sOW`AJe%J1<xQ<ltl)bIXG4V&(o`5}
z0uWFJ5wbb#w{}L#x_oaD3O95Ahfy~7e_+7>;9@19fpkKb=YPF2Q9W-b<6ah=YA54S
zAzdKH)$FK#vTLit425C`vhb6s*4ht>1%?-H+Tu}Zi*@d8-~a=>D3{OiVrWHOy~nvQ
z!N#G1(C%wdnrdgbpAu#4+3HNSMA-ClUt|q^p-|xsg)y3?*P2F1APe_jmQn^EjmH!l
zVmz15$HQ%>eA~msWlll^s%QM1)%^S+N4A@LOx=at2J&LXe-Xj-n@R$JGGaPEJ-oPg
zQqz7AW_YF!#cG2%i%;4kEUB;r=OueQ(uZvmNdVg29w)stbLYnnT#TLCHAl*K_@(D#
za-Htbn=^<9l-nkauHm^qy}wf+6YeriR6ZmakcqOsk$k(7;C6TDw`<Cb$-~bdTD}GC
zjSIw13BYL!4TT88wAp9#3PVC5V4$M%Np`Q;-)ZwXf&TOFiz<m4^BtF<xecRNW8EaR
zU#=Gp_K%KH7<+TU*_oRbBYC5U`=08p$s4YhuBF@*tJEA+7NC+@$TLgg?`%P&&JE!C
zE2h5@{y)<=;&n&O$yl}-KQB=%u2OXjAZmSCwB!28uSYze2AdZdZNrD<tJ7W!TZnX`
zhqE99$lU=)#-DV(On<KZIWGGfrHEa3UiR?^4T9*13Mht!zG~Sl>&N%Vuh$`0=pDt@
z5vBSRY;8kv`KzO~&Ybv{V;`e$l2{&YsYbd{SzS;5tSDKm?>~#Yl-TfNtVlRFR*49w
z0$hdrv@ZGwC&how^VT=B<!T6ouN?{u6OSjp9a-`OL{wC@Gz%iVz#=T)`6c$#*MOkL
ziXmxcoev7L9wWf9#8)&q^xB+OOP4iH20*aguVQaoFu#<Spt!G%M?81_yHRrbc5253
zdo2piAKYTBEV=Npi@{LOupe%;E=KLCo>nw+q|?@=SgC}fQWJ2;Z}DOt^AU$^d?%q9
z8>hg^5JEHRr@`##q|1G!U&P;|`9qp8Hu>3KdAz2)%<<gTiBv*{kUljnDRYdJ@?<E?
z?VDJ7s-0$locudHtJw0dTTcOwDmYIOxy};`7@g#-2t^ZesRUhVtw;4Ui}WjB92Eu2
zPpjm-?v?bJ+whIDIy$F%Tb~jF+7du2F}>R(V&8(#4cmd&-b|L<jH-xcVIX?##QvP@
zHPZ)Bf7RiZbTRJZn_URqLq*~yw!sMRG~9$-?hXR-HF&BdCgLXmNsdDG9y*%-W)}Wv
z98_&D^^t_p-q=5+p(&hCwm&X%H$=zR6Vt3d6gMk@FV#2P3YQ>)l{?uMdxkZzRw!?l
z#Zm$zdqQ!v_#~{O<ExT$=hvniQ|0J4iVH;9W)W~vohqYs5@uDqoa=dx`S2vQ(Fa55
zZ@}>tgv9zpL{4OT)t6eGup0CBeXn8r39n&`=-KF9O^&OP<7qUW;vRG7FarI_O3u#e
zS7%H9XQuBEy%npr#fNCi{x6hJ>qin0K&tU0^OoqMim@z?75?6dy?Q&h!YJS`GM@Ka
zhRn0?f@T69P)Y3T8Z7EKt@yOkRtrHUnM922yj1Htt=goKb3!;z-m6dFMIP?ySe2u$
z{Oz46XEpq%n1Lm5o7TC)bzA4+)WR}^SB)bKZAh0MipD&+_bsV{^SF{))?xl=<3?bR
zj<lbhPE?Wa8sf1czdjdeCRfmgCl}NW&b%rbGH&b)kEz_Ok1hHr8q{Q&vjiUwtgR20
zQeOL9FBTJbD>B6$ZJ^Y~OM95rGPCU(H*9Y>TF6~N%uu5@Fq1dH*SP~s(f~0f4T@1s
z?QYkQhs`|7jgncpYAZ4*b_Bb&DCW?Q3O&HRu=cqc<+PLeQ=nU+1IHV%{Lmi2y~>7h
z$W=b+^F$?L$=Et)4xeU~V6QCg^O}aHJ4)KJnUe-3S4m7{3u`Kka!dc9_EB%Emn8um
zxVi8P3yruJuS2{J3H{?Hi;fXOyrf*wC1FR#t7TdxVTCmeK&cQ??I4XgaND?Bfg4N6
zH{uSzQ=HGvf(f0`mNwa8hvJS=lt&0ZvRmj7thg%dCa|$!QlBJ_`sCY|xEVk6a_G@k
zT4ljeRq*6n1Z+EOuu?DrB}$qG75n7R%`5L`u~=7#%wAh??cKZ6W4!rA)Df#y61~#u
zDfQLBe!A+O<<_WO-?Vj|Cl2u;(rRsY8TfW)7(6{svhEdxt_`A92e*w<=tnbnRs)}J
z+jRg1l1c7*I?f6%9cSi{yeAZCHNQS3^p3vIUQweROFI<kk$T@P@3+4N$HzX35y6?x
zJe#FMitb8r*xtDw9zykg@uHWL;X`sy5931kW-|!kV7}s_C;^@~so_@BO6912g5c3v
zldDC6A!KD7!JVl^JNU~niyawWfN~=QiTl(ecBq?Jp%ZRW_$rJ&?Q^bUQnt9pdkb?-
zhjS;z9x(iIB<U9pAqXf>N%|v#PzI>+sj;YW5vf@A5?Bf8sX<fML;7kit1^fIo2%Gv
zA@GKyA7t(LK3?-pc5<!J+>%~}=6Q0=DJE!BDxOdPU9x79ED(;dF`bsDEa)F9UDcYd
z>^wmy|Gcq#x1~RR3tK0bK>d!&0nZUL5Nmjh){kH5_KVexKlK?Bz=aJrt!71$0+$gh
zcgHd<xzrOADas?#n>9Ka=%m_Eu}qSq<YANjz>^}id7#fb8Fj7eR?g6~YxWvHJBv}W
z_c20{FolZUQAATxSGP1HS1^oE7te|_oPr&#W>2`A!+}+D>YQmQBcb#=l{;ERKQCNV
zBT8RTuT)5}FXmWT61ecFcZmri;ul(3yYtsmBhjZ%3F<`&fpY-p)qQ{y`W@i}BoZS{
zaA!md_<bV~6<Oa$FG2~OkmSVw^auXG2ci6Bq@Q#6f83z|FE-=fFos}c<^4(|Pg`g!
z<WrULDD`{!LvT?T)3k!N8mB?8)2X+PNpsRD(@UpW{%HHumWVGi$fi7_;pzTSu@bD=
zS%|Y-ZHJoDHsFusTd1tez)JJV@@0`QRykr!yn0W&Q}PXT*drPosusK!ZbIvKN{<_z
zn!=ws%0=)`VQDO)N0L^2Zs!V&?zCZ@-<R2`?`OANWLBmaD6n=s>uDaU!8x=iB%dXA
zc_iyVw-Q<Wf<7S(L43KBq1{0Zm1b`$gr&V`(zU8D<ql-1r{RJX#RbPmM2YR<q#Ezj
z`8r;Jrh^q!G@-uwZ3KaS9(bmX)=^oj7+Pg}&pB6PatfutSQng#hn;s?ooM2>fJ>Fr
zOdX2?*U&WQkukQ?Z18zfY-$m4AoGneuAxZr&0}hjU&oB4wN$8h5M^IgIC)WneOvxW
zqE)p*zd$HV^bNhWg`k+e0BvC4<G1)%;{$UfS48}mdV(z<%=dde)0_IA5SpbRq{^BJ
z0PD2lJuvnsy7mnuZm<CQov(~uVXXcxf8h>PaQ5f53rqH~kGMI-p5pra@KR}o(e?{^
zoTVdI8eh4_#1U4atgdc6k4%X(S^aNB?5OR9dNkEM!f14%ea?$%6uo<;t~GO#R`ERT
ze9KsDU+B*lNF49CTU-;zSg39G*Pk;ZcPEZfsV+*Cn!nWUy0$_^VIX-|sQ%2h&%;HN
zJN4<F%#jM^I-4%=&Y>%5dse}UlB>xHA!0JZ03^YTX#CKpwjlF~zZ+o9pVFqVhLhl5
z^{@8Bp=rWv_|hqrrN}r1PGhn@(k@;@xlgsrE#(i!dnACxV3jR0iq6V%hdiW_NlrnV
zv~CTri*c9<uecKH3z$j)EL{Tl%6%?chgPGM)duKv)4{N$cV}sjv4GRRb}}D7`_d#%
zVCJUR&L#Qpv``99rp(Q;9cPn`<`TUX=)OVHn*A=2r${%k9kO0___hUl!1;LkBqJw0
zoO{OTdj~8Q)1WRRI1sS1V>p&fQ_+#mLuVl{Wh02@-~2}O8ig+zRxg=6^J~@18!jdc
z-PMTs3cfvXI<&@YwhwGC85P~;0H6${j;Sy(_hGm!`e;7inI2#~H10^!k%z}8#NVYT
zA?84)=%{pTL{r;CiY8yf>x%^Eb!=Me{~`zRNaBF#joCc5Ch!lHgM+6KE3Mz=I8OgS
z0QYLI-cNH{(bXKAW_N<~N9@D$m~ml|oYiCNiH9!A26yU?xKuqj$9d`MXSQgy_Ra9A
zMMX4~d(S;;KfOK3<(@%6mfP|^CFe!yc61QsB*m6gBW4LBvG5!YNi&x#iHBQ;*KG|P
zfErVGS56NG!!h+qZk%nzNFrtVG7F}nDQ=W$&Xa3oax=5Tq^m4YtR_NBb6|DP$MRm7
zUQlJiD^5Qv%43YzWlN-I(4zNt%2%d9rB^o;c>kvKvcT_(+$=xJohn$cc%K(EVUg#8
zyz;lGqm-0W5T74iL;cj+$f9P_OjzMuV-^+(3jeyL%G@axs~_E_wh`(%1-se3&Z|?C
znygT&f9gnZtBdl^C<3H!dXvcB?GmOe#+Uuvs^KEJ8V9tFp_DYDInSW)L6-&LHIkDQ
zQ??D1gsd|j-^m=j`(c`4&44EUPfjqI!DA|gLk8yip9vyC@L$waDf3)IU`=tONv@%*
zJooqrHunlR#+}@05jCcJUJ~6%*k)Ycxp@uUvOft_w0QDvPbKOx{j?6x+1s*z-4diu
zn9dNBrC5G9R<ZE6!@$-o%o4W4n755}huL_zw?-SzG}gny;%<Li6WJy-&9!t}Whn7t
zyj+Z(a@SYswAtxTq?NIkMo2SzZZ`e=hIjsMtQVE$=eis*WqiNugY#`gNA9jxRVq%_
zZ^lt$qQ-UL8>P3c)D1y#Hq}Ja<VmX}fqu)vomZ+}^eMR>-S>~C?vtF5M~Tw9vsE#s
zSxlo8yiid>@j&~IgPYHgpVi>sa`#0?4Yu}J7`#fDcrnnJQSMh&w<{@}K+|n%tNuT0
zKaA~n(mc-zixXI;e>d)j|J64HU`Uh-rkL9T@<G&E*gu0_0q^<Ug?38$T{4}le&b$4
zK;culLoz3t_pwF9gi0W(fa)MrO?F6j_{DjE!(6WSgZPDTTG<IwGHBS6Yk*X)o8ZNd
zr${mu>Ft?~Hm2xJPuRD=Zbu?2W!tUQ4vRIZfIphJgHFkDxIo`~Bft23^wCd!A-1h(
z@c08NE^Z11F<~t9K5b397=n<b>+9sjL^}uU7X6YU>Ml(|2?kttej35lrav=GxoVd}
zTkChViHXyP7%!iQ3q$(m8m)V(UH;IP8fV)ca?l(~-TUxhRYKKEAS|5&p?px3Ixskq
zv?xnGGgUal0ePO9ZxK3#n)5H@ngG7})yycky;eXGWu#ehzs3Hp>G*@JAoYahzg}aZ
zy{m;GZA;fnW2FwSZ#JGBqbV+;c-_QGk_uVJ(5lh&^N=55w%^DTkBg$pr_$sBI@XR}
z86171vEs3?Mb$be3+yx_-Y@V;xg$7}&jBf<WmTZq7PqKU^(yL_*Vlb@$E1X#P^Qy$
zh~R2^r4Tqg+5Qgh=LV|L*8eBm+^0L~@vnApgtCieVHiGwjbn_OiKu82lBWE%oQ+y{
zOLV{95Ey{7R?e@xWii8sdeY<NZ~>Bq<qr!4det;cUg#)3g>9(qLAugJXQ~lcIsnTO
z5g)u`kZK6CvT+DBmurA|1ADt7Mzt(90+|RuLSe0G4Dqk53OgBecL8cP-3(hgx#6)?
zrveYcC;B4&wEQ`O9lTtQkc7`c0kv=WH5(th==GjgkmAV^@@YQoy{gTr|MRSQ9XtPL
zTdMd@ho-z6PP543qHOA8q<UrI%RyN-)0C0RXSs4K2<`$>6V{=ac|mWT8im}CUFwjy
zOX6T#lbT*hHWe?OZ?N)P`9*1?x8zoKKg)><6u#tbK!Ck4v!jI%cW~sIGUn@q-B_8d
z1`UF6W5GjsxbI3}ZA>^`oWcmeDI*hKXd83!|G}hxU{f62kpPzdDb@aOkAD~+LKUBC
z*(O`Yy?^u}CF%PzeD(-<Lh+Mm9joXs_}M~0)SM}cDX|Rs{%oPro3_FB21wG1yDqei
zv@s%J3@5LUaP_fL_@j2GFqaAQmDxUd{LSl#CaL6yx8;1L_JuSeW<{t31rcNQaut<E
z-ivh8$wI9Tlh%-98_zWEw=)x4ZxVfk77An-L=2a3o<J;`Kh#P52ls9QCb=EFpR}~y
zHCLmv4(N5OS>O-7MWu_$&C4gp4ZxD}gs05A6?tbXk5Opn<GhsnV&B@;$I8qTf7pNP
zk1u-$SrIV!Q_{^?N&CW)zASm*3&>JRSjA(=-%XmJdp~ajj;hJ)?K+}eraqe$(58;o
zISQpRJB4lciC}<xRih5UDtXN(z7pm@5zAS{oBfSe1t+`y0!H>ud@22LW?qxwYo^Mj
z3EjP?8%=~!YQ@TF*~1OPN_bJqtHjNfBMF>dd(}<@v1`o05kpzS1L2Fy@s}56)r6VY
z{KV8$z5SY@x+g8A{!L=@A~bx5yu^M2tA2M&ypb7~E}JvRLt-NZ_JgZ_{9m0V8VBDp
ziTtJu5BTUKR)I;+?~g)+g$d2I&!)~{CDwIviz<7+oBq(N=h~_bg2W*eg<2geK9<-r
zNE2TWUArdGkN&LKBmR>&G<e4SA^qAk=swaAx6%SP2pAW57QC~|?`z8Xl@H*9@^aex
zkUF_uD|!O-P9-!?^T*bf3e%f1%wR+7-#kerm~8&?)9&Q@r3U)0I20Cui|5JsOc;Iq
z|8AFyhR)>6uuyxC-SqYHf06P^2dsP8x1u7OjHlEEiRS@s41t!tj+D`5D<(JRm{6VJ
zW+jRhRKj%C=lTZ1nsYoEdDN}<S9+NBH1P$P877!K8OYx)wC_Mj4JOn5K8OA57SGK-
zLB$25zMyQ$3T{|tdNrZXyAL{7D^?5)S7SsJw+SmJPG5b!lBS0h4|z<!+?@J%mAoFd
zoH_kTD4wa4Ad{(Km5d=J?+aEztc1DFTmg*37YtaF=^BqP+Oos4_U2z_0=<L+`A_3M
z+VnBeDWWW3!vS8YalvM~&bO|h`3Am~OZm#VSxp5n`i5NBm;1;QjHin1Q1Qef7Z|ol
z@nw{v)_f>Wj`NG$M>($G9Dm8#{3%S+cY!Isn^P|fryaX*{e{Ru4xEjPAzawcXksxk
zGS&rX?myq&tyT6WryY_@I@?AuGPKi_=6m0{VwSpYWHNO+R;oz}t4L!ZR!QwRgl^6a
zbeJeQF~o=HYK|P>O@i6kZbe6&y&9J>{d*#8#LM8E#(SFzD-)oqcw>1vd+bLwg6;7{
zFWo0M=ZH{{%0B}?@o?64y74Vi;%9rq03hgPGZFx^#YHv^vI%wY@)e<I&xn_I_)|&2
z4WnmK#YYP)E3CU-dFuD~=SvgXelw0l%dsZ8p4_<oxG}!+NuTEap6RDjG>Zc2VJ}+0
zygb-3RZ@I}98VhJah}i>(HglRI|P@1(+`8ooS>@2H63Xw4r71by2_V;uBoalGRitR
zakXy$md<|d_z%Gg(mF#O@)36LcHG^WcyxlysD1f>8!CQ5q%FgBh+_1yFgIVh+y{BZ
z2Pj!)XEmMYikrf%J_gxm-I~MMElJ@k^%YE}oH;%Uh(fIH--oU2pX0Zan(sA8=#a_!
z0ti#JF7+uCs%E}qTxQYC1#g4KE_9LhEVHj-!yF`>p(#Ud0bJFJ&PGA73*ffsFUx!M
zWlP-VkF{+PH?4{}*6OkS`$VaBF&XsWC{gPxTl0D#vZS2A=a@N_qhJZ8&5SXB?p7_2
zU8nS4>0JRhX3a#2263_Ow}%Cgz9BJQ$#I@mCunbvM#I}lUBgROcDIe_YE>K^D}M?1
z3)mW^9COCP-IXI-dRn-}=#M*~G~uyf&MIHJX)fM37BP_CFoMruW1x3c(Nb(}1oPBV
zI*pG{&Axr_dIYSjEKtQW%DEPHF*X>MvmIgVMQFJyIU;z%Ira5hz%!a=4DCq={^WiG
zA=bP(FGAd(-!Vu}PB%5-pY+ih!u5Wh+@y>Lk(RV4l5*g7317FzM@F<7Ls-=A&?>w6
zVKT{Ko1D9LQ<sstr2Y@>75G5X_{rD5i?igTj*h4q1>!F0hjU5C&w~GT>t$J;UQ=8}
zx8LrUAHPJ7#K-c`e=C~3PJ<TW)|4td^1~d&<Ygq!uxx+Rn?v`4{N5&Qsg)qK20ztX
zb4a=GXl-0h=WW(4^!WNSgmAf{S*Vtxd+L)J7>Og(!AYwN4Zp1-5q(;<!Ob@Ugn7~|
zej3g4j~ow7A|zV9sxzag7JeZTC7|nUjau-MgNcx89Ca5IZazqrUK5sN70r-Po{jA^
zkmM;-@qJ14OrMsZ*n^q}ey%#o84?X?)w;g_zB00N#rm^t_IPGuoMa)7XtJ@@oyObJ
z#gIZ1uR61-;`L{$3gZmC3!A>Qs(2f{Y{?p+$9I(q<ww~m5jVy|_ank*k<;%T6A_Z5
zucE9=@=ARwS^58vmO8jh6?QeMag5kWnof)Yf~F>mq-Dh5y{X~g)Oy1c)Z}tT+_qMA
z)bCwp{1|!$XE`IdNdz$DDbg04in8U?aOej=Cvl8RIQlJAlE)XIb&tPO)GD<)p|^@)
zw{7+(l(R*|w4c8q6vRlK)fLvbj11^vWNX4kOj?to9lPP-^}4-x86yLwx3on^-Ft88
zM#FOjjN+4W1#Vf+^2tkT67JqgFggPE{gvFaQnPz{r$oRkS$T>9$K-s0NY{q94l|$K
zs4m7N#;B!Rt-9rPN_I~!MybuGsyfjkI&?LOo<~InvfPhb&7Bf|bL_MUUqiyWU<8IV
z78bs2T1TRp#HSPQVqn;tB$XTy3tkBI9Yhe40~A_O5;*2M7+YuRWP!=)XLil9(N4<!
z0)2(v^rr7bVtyuGn)C!aPm0DIZ@`xHbB<l%&R#K_gN^L~Y6kzK>~695!&`EH0zvL{
zg(*Az8^G8*7zl3-=(^}uaO&h@65RI2NK`odme4~7h}?ONR?+U-D;*@MW>h8lc_QnM
zF1aR{(g`+PNQeW86%$l2u_JadLKa#Z`SP0_xo*_n2mwZQRQ#Kj%Ee2!|941A2S=eo
zhzQN`Xczp`arfUHxA6r8R*PVUiKaj5=h4=Ob1P*$+Rd3tc$kEyiL*-rs`IR!73uv&
zMuWj^a$>7f{%F%pkoV|l1gX3#c<f!Gb9M$ss0BO_G*aNFyhPXQNOk-^26ZIvzNA)L
zenP2QX|cu-=yN%5R4yusF6<$yoow5K>`F>Nc{!ge{CZNwY+ZsC-hZ0GpZ>7fA%Bg!
z<G9z$bu`NZm%8fY6@(g|G5TcU^bnEj+|4&*)OFr<l6A0-uK67$<nNW7S>4g~oUEY`
zEfV2vW4jq|GFc7r_l>9CGgMuo<{JFMyBB?gCWLp4_z&;(cZdS-ckPrC``4BEBZji<
zrbx?kE*$34K3%~R)Y10qf8!=voO`1Uud{Y=Y|pwT8ZXA7c~h~tu$?J$jb@2M!`(G4
z#r31}2V;@`L>rRSc-nAV%P-Frk9*{(N{2r8KLxpXdoy_LO1>!SEc%E%jGd3(GJa_(
z)M>u$EUGefDA{hg-?Xq(D`jbysJeZdK{@|#gX&aqCGq5@nAVj;N~dSepLCA6&RqcZ
z%*Cp~JoIrHLitQIG_ADc!bIwKBl1`bZ}SPb@ZmrdB(a!Gz3LeIm|~n0LP@(RlJw|s
zweruVI%G#D87Qtevh33yr#^o=9lr7^t$vgx)5&}x`33Vt+!EEJiC4M}bUtdoOCnUD
z0t(4BcZceo_ua!V@36VH4ti<R`v#&AcoklCw8aO^wDIhY$4_XIW9nBb!AJD51o!Z(
zGPycv>NF~tvB>A?K6$<BRH7`)jkLg!$EM{-CX2L^anR42Y!R&(YYU%vmz>UQq=7l_
zi&OPnUHu<UfpM?%t6IHSe;)bc43#QLUtgjl6vnDT%qoM_IW`#trS|Ms%e90p)qyVB
zMQI-_yaF^0rYO(LOVj-Tc?a{AM&FsJjv2E9)JX{gZ^EnPRjtD1U9dQjWV^&C7vnN{
zlc;2Ad15x)&_0W^{<XUDCjSiSGBqMTJE<*ZzB<m;0M_;La`t<jRPzA;bB?Uf4Z7f^
z9s8)4Pb%JeMG_>4$c1Fnl^jG3YeuWtvfEqx@N2===x*k)2waUrjVK&zNk!5>dky&T
z*2>PTYS|$qUZN1>T<(<QTx{GHN7qxuk2MqUj(ZdD-aVkZ{$@#{D5^<(`YGUp;KUc1
z$&0Z_cWVj7NgK7|9b*N7H-Y6BNl`BkANHRUmDD(@k5$J;c$dU=2_Bvnb@k$0bOTJC
zg^PyOe0m<-_J!(gbM^9hrV)$)nFSs}ctSKdFGYdon0ZXFWby4VmkTv0u2RClZG=E`
zj?DR=K2Kr0Vpj(8hfz-P<FG&7-3{W^1G59pFwCz&@n?9wCyuMVVkVz*9(K;ZI2+nB
z=BORVRk@;h1riE656(1=j)NPKzE+Kl_2n=iqr8USiG(U_Ft@J?A@<^DIBS<r_tFzz
z$d-zCg1NU%CcXq7T_cHWHO4jTtE0ADexs`a^5kEfj<MzS03WINo97{><p{I3#%{t=
z-y>{Xt>@o4o&ASFs|VIrf#wqhly@bsKo!biTGa}@z*aS_vb5lb3nn2aG=V!Pkf@fD
zDHEHu%P=doz%B#T71Kic-L0tyxcSSDl`woGXEFP<enUjP)_@<aN1MGe7q7kO>Z4Bq
zZcT5togZPdq!t)P>~>v^$OKdpS@ae(mUu5yynth;FyXpgU+OJ?(i@76S*F4Hk=pRA
z1&Y`hpYF&s+|n-(O6_5{x7DE<ipD)xx!IgMdpD}SiAyAc@19*r(c5Tnq05)1iHXiZ
z%?+eGxNB!Il^6rhw_Qx1O-t}wChtR|)Ma;nCn%#{UD8)A(0$O%`vA1%*30Lf23h89
zk`?RDL;<3QTs!&k@7QX{wcF$$Q&i;kFhye|aVx{I%ia`K_O5QQcr|oT|2WP8ziA4(
zU4@aoRHEo0X+K2rK$t|@dyNi$3&+>G4Q_7GnkG|}W~-lcl3Pbk)%Go<Z79UuXi5eP
zOnY#_7YtVr`ye5~yn|eNHHUnp4Gh!S_ZZb5a>J(n_{?qjo@b9>!7-+^8Md@tC7upB
z-#ba!*AaCODZ^8zlX5s%w1bb9avXdvqf^e+UI2$2#7*s;oJ4+=-uZA1ZF1YYH><j^
z<DyKcDD3(C#d~>t*qCCRzq=zFFFDg4=A_|K`iC=KL38+|^X~Ootb)M0DcYsIjU}#q
zr7eCT{kiQ0-nLY4re(kL`{mPT1LGqm#CDEv8gQUAeGCj`A@<5%@_SvXoBhi}&8P=M
zKAi)r(%R{Lb%=$A#u>p~Q*?L!b;8$N(s()v=$Mj>;laE9-PH2Xg)g%&e&ZrVI5O*y
ztgUIZ2$i9s*(u>cJhuB3IlgUCf%uRnB=@XuIKmhCV_%>JG(K?0sUi;=m!ySs!PvKj
zM|PgM%LT}l^Km+5-p$JQn)$mQuT%f@&-v&5qkO%o(5&_OpX%yfvww3FhB>X{EGNsX
zS?c7s5ghLwK-RQNm+Q#}xDwF0nTC3vDh)U`jOW!UM(bc7=l{~MLw<d-gJfr7+}-q%
z%KOO9ybC=A^@XzQ9tA3it>4Ax=BrhyPkG)*1F~qd4MAM0#(94y>@OM1d7&ITV`Xj>
z^v1i#a6)`nFO}I@5kk$J&-b9fBJXZY?3S)W0`DpYzi7cR=K(a_hLKKdf2=uTpH@0p
z{0F!JcWLddj0A1S;qF=vsL3gm`djN>mQA_mN*dT+tsXq&<w@RqOk3vEY*tQWAWtO)
z<C{E&SSR&7A%5zhO~fib$&K&JB%DOvp&)tHP<ta7ehV{cn)F@%$yol(!6H)S5Tmdn
z{ai^W4;ePc?WHT?nL)g<#j>FaRoAcJQ2hvh?-29sq#hUT#L%UV1&`P4?dZo7`P7>4
zqb3F;86V&NL%mI+ZR2v<pga)AcN)Z}lVgt#(8rm<`Pe3F_83lzVj7c&%9yoR!uHd_
zXvR^oBLTK*cs}p$4ro~m8TDQlzN<Rt>OEq-Y+sVh!nC2Gh+Ky(Z|!K*`d)esz0LqR
zLIlIR*Ew1y!GSsz5Xg)?ALOu6IJ05un!^&|u<eu%rjZ!=8e$HV*jJ5&47HbMTg*sP
zj7c=}uaB?vD&}x3-MeNV#P&yg|H^o9%J!s9)G9YZoclK5Vg$|59jT^^r+<H_!Q$6D
zC`VBU+a@RYmyODvGVbQ)?c{6dq6oy;15-J2t!jQ;6T##3haF#7rorkLu$~3K|D1r%
zXey)$k5vD`asTW|<a=6uefD30Z!N+lgB-cW|D6W^zpw8pz~MkHmcgN1uORMZ1h^pb
zjXbF{q^>8;P@k1TJl*wa70UnlCm)=&C+jO<C{PxrAhxI~4Q?`VpWWCpe0z;l%J266
zq8xZx<TW<;lRF(Wa=#W(Detsn*p6@`wa!M}G3o(e0GaLVkjVpVJ6gKG2%JY)o5l+4
zmlrtL@0gO2<X9waOLEW9qGev)!mCNfuF%X3#%62Xer-=Nl^VHPXVgPjF)AnT)A7=s
zdsg~hTG2lH^%2VGgHZ)p`=}iB%rzF5*zkiSU`yMHZKIHQ>+Q-=o2U>L(l(w?Nh!y8
zu0>jUHxm<deA8^teEo_zX$ie7G$c=pcz@n%$?hteO~>t^hfid}g}E``k&K-#6cY?;
zqc*4LSuerjTYmWbth$2cmWr*0GDYYSQXLUi0<@*(T8Vc2N)CCKzc8M=(5$n&K5p_6
z^DFnzX@s>gAGHx+bECTgY|Z0<i2`}awT1A^|6{^x`lr?B#uBAo(OJ*Rz=W^o0au(v
zgsBm|fFG#`JKg&&Cp=?_Xh8~9<A$xD{EE7LJ`SCCldI4YBOdRRm#VkMkNIy$%GXyF
z(%ze{aWdxHmtE`_*r9*)wzp-5FG32uUC-hXuq}8PLp8-`LVmW@qP;oG7}M7ym~uO5
zd+37NwBYb9JKgbkNc6V6{#$MSbblD3VY>Cth~Cn#83`yUj{U7%c+T|0g4G(wyBoG{
zoZgYj5s9cxm1NoJTvd5i@4pR-z7OTRgYSc_`ozW!SW^oC0Oz1Yy6Vmo6&)#E@Fv$$
zBP3aQ9>h{1vOIJwey}C2w@a7AHCUl*swbKrsJYzZESirQfR17LOTJV+6zHE>&!zm*
z-#ZG$P>P>V96qjP%;Rd@XTWfTdEL_}|FE~zfcE;7WV`j%pHOwO1ypTIG7tkfNB|5<
z0~`60D+Y(Vo4BQB(RJLti%hS}y^oNC8$AU)bx=7_XSmL{R$@&=q=|azjK>q=0<q*R
z<8RsDh+ls(bZvE-QZY?^)VEuSnN$72-9`t_Szl35GN;DJxB1GMRJ_e?rvP@U99^cP
zO(LGxM&L+btZF%-^fw8G3mV8UErS*?x(&IZ+|yeeA$h$y9;dwQqN<CZLK*ySt85g|
z9M(?r%8cxwT;=2=q##dn6SxHfS2Lz0NE~u3DW+9I4<b`wC=zoM5A*qF&1eS!%*tnV
zaC7b8kVWU^NUn&1Y*zZ)W8s=PQmK>=?W;TlpP;spN(&{?I3=&#qZ<l}bzTpvC!W-Q
zrrWaPW#x?~n68sFRqIsqg$+J>qT>Ts2lvQMCVngDyS$$-Il2IqpFL?0@nr^{26Va^
z`dr;T!kA|2X%5N6-8?7sar10g5_Sbz2@$S8D}J#XgE58OGhX!E96z93>KW+m?PI{p
zz(FvpA#WHD_Vs>nb9CM(^NQbxb~)?Mtr-k;Q}4FKPti5(TA06dT4@j%*|?ERjPcDI
zWuT)YWVGjUL&IDpjmz)ko=95k=Z<C5XSiaY*o=c{0x-<Urne=3oMM&Fo?J@Id3c?j
z#kZC|sWbZhIwYR)mLG1Ty%jcg;8_Tz=}<J=e%PZ59G85<3~MAk=~5GND?o<+z?zHd
z(WG?htxDOb%Gh7=jtxFY?mH)X-AS>(a@tI%VjR`75$7z!tQA6B`I_KF7(2K|8dQ>X
zYz{<M@)561c#mPCEiIwAd!EPk8ARtPAAM`j5fy(%JZKCHAzzXx(8IPo?cObeEYt`W
z%^B>Z!JFU(BLti|0LpLvSzxVV{OOE~p0||ME~H4bXr;jIC0BIV(^wdtCTz`rVYaHH
zuiGcOAE#=0Q{!Tmg;p>x3gm}fxcy1eUHfJnxDZ6tA%o={_5*g&0|v#pm<0QrjxiqU
z+TgO(GRc6y$&b0d_Ja@iLernx`)C$C#+WL>!?+csQr#p8-xG3TPc)Vi%s*!9MBNxs
zcOg}qZ@#oS8}6OwQ|CEPYyUVvzFvv$0c2Xrrc8q=GKmillXIUM`#F`_4S0Eq(z_B8
zsD=NrRQNkkmCg0ElT8QFP?elJy&*_xW)Nx<WwQ9IO_GpB!K5yFzbpY>8oqC$q|!F@
zve{Rse;a7%jW!<kVX|eOd@r;xe1JI{bCmZO<tYiRnzj`Uudfn$&f{UeTSsDg7OkpT
z1jr&uZ#6RcaTGf=|As;%E#oOia2VPcT44I)zSU(5$mrqTE2`MtiDli~YMZ5RF!_Ai
zI;d>B*OZMP6`<yHS1^!H)#gL7?GUF$ft+EzxA%CEkNiIJ4B7U?(P18M@K^At*9$sp
z9q*ts>DUx^U}!9k_tU76h8%gj;mO?TzgH62zvh+p@Hw_kXw4wk)b8g$RsvgK#a=pK
z=wrQc;YFNWhe@3-=bs4C2-R&o0<aE%t8H(ExpK|%*n9!pb;83(tD{dn*C^7R=>I%g
zhp2E=bn7mgtHT6{!#FzCmNFC2f25jcv+8cCNmBvWA;S367}Sp26W0BAGdY^U9(cWo
z#oyLZWp1z9cG9lv05q_i4wWi3pHWuR8VWP38C^XGW6=RP?q`KH`hzovviZRKu$r0?
z)n2K82hDS=i>RxUtK$-7G>LwO*KGpNA%fB)#Ni2O1tubI!OidT<hLp9Mo12@m%0ER
zUB-V>X~5`JLwX!gjbr$FdQ>EbKImHvE9x>mt}I`TFH(P88Js0YO}4b!(FYaulr%~Z
zK1M?W@l(H7`)As_H6G}6={k*1>ryj;%w*qDWb?+(`%YzbQMz}Jl}1|mz;$(ex4lN*
zA(SilEsJVQi#!vvVeH+Bpa(^Y_k7I1oFY-V22=%BQy^g}WqB*tfnCE<>#*Z~hFp_K
z^|;wlu-qww+Q{vXOJ{+y^gC6H=#H9_v}0r8KEczBdu}(UQoDo7CBIJ}<mk=~eWg{Z
z$gdMtX!9cSZ0~Lq&ZzwLsOfrcEVaVLQl1^6Yt+zWOUBCJUMFRHlJDtX(<hQ^l6++w
z)jvs45^>yb{+$w2n#(jV$b!>Ve<i@Aj?sUr0c%`22}GDZj6iM~mBn7%+GNZ6m3l!)
zqDyBH5ZC>GF8+NG%8tAKv-0r2&EWq{>GAD|DS|G^Y^s`}I-%egFF@|^PJS+Vic_E^
z8XDuHW&SR{BstoChSh@Vn(?igf7>&%?3{Wm5d`f$7ONP=wY5(p#&ACzb}fZ+gKWok
z932IkzrJ;Ne0q!}K1*GcDsBGaZW&vSaM>}QSt8NC6jQg9{SFcbb@;z@8#O=xdONji
zFLAIT1km$%rMqOMp11n)Q=gEqEhUpX+OsWd#p*+Q#<F`a%$T|MFt0pW{ADACRTprb
zU6aDoNr4Ju#8&kib%0N_)WKUmf=HJ1;-}Q5(S}B2?$ZoiHCU320BJaIv;8i`<0PER
zvvzkVN$t;I+MvN>jK54sPJO%0xKz06Vu*j4+avR}`+Y7uARj)fAwy;9<x0bA=wq=`
zw;VN0+ekB%RG4_@0=PSh3wtq2uuE!1iT6@>y!Hv8j}_=$OLgr+9%O0E1_Z6q?bB*e
zf?bO(4enAfY&R@2nJ?0&3gi&UI_gT|ViyZrTV5zt+`9l#&Kp55pS8t-EPp5tIoFS_
zm@)0V9y}5zRwzbN8(y=F{eq;uge!JBCmS4K5P$YL;~)z1YJ7$b-}e|td?ctys+hwl
z@crPizg&lM0<F=hh$lW(yMSfEB8=~H&RVdOk$|v&!~rdr=&oUErWmM|Ca@=}=bfck
z?D8Mhsv~TDJ(q{w=9Eu2<IPU=!=!VjMQrmH;?bE^*XWzr{3KxI=NB>B1EbiG>5WWH
zVHwuKKj>}fWq(|C9YcF)ehZTN3H&fFF^)G+@E)vHYlaE7Hk5He1H&QmC0}|hT|!%d
z#Wiszdo-ulW*;LY{&lPAf!gfoe*T}r97JnePuGXLVd*By7fZ%gm)xnu-_rW@mm-!~
z?`x#IXcbmSJV;aHHSRq;0Zh!Hn-}>aQ(RgJ4x%{1pnEcOffn(^qei>bAhu@smi`;x
z)>ys?vWEf#UC5GPEpMXn*v5m>;HjcQHC9$n>Og5jAnk`8{z?(uYiRyR_QuUPxWneg
zhZN^qz%mBl^fs4?hO2CI#L?aAM5;d4`?Xk#3zAeRpd$9X(DEk3*qlKIYkx$2qO>!o
zKgMrtQatczS6(@CdrT7+=7eafu2Gy)Y*PR$&yZvtjjZzz9(+r~A!#mwuymTD!Xvb0
zw4(z@pNNE#J&3NzmZy%LBEGnlgIxk5SC5MC5p!wPGvtz{(0A(nR75QVD_FJYYKFzx
z!Q~P5l9ImlWkS90a5m+PA9dvwLo4V2+S@mi1+_`;<62}Tk<3&}VWV!Pwr`1+=P!UY
z(x}T~7xf5+YL9ya1BvR_j%bze<Tcv5%aO5qP?Gtlm)p?K_}~o<Oj@Yq-%Nw+q#<7S
zH~2z7yh6~TxTSQvKNySl@Z)n>75U_n7M>KxrraD8S@4phx9U`CWE@mJY#tr0ojO`J
z_%?-Cfk|S-W5s)E3Q0$=Z$*oUqw=YuZp3uxg`bgr+fx_S3;*WQK*zy6T`mv`S76B5
zaDU#-X(BY)&vYHVV-RBY5`XK}SO-qm>d9OM#J+_y*yfcc)hT1HN(~cyoGavK+lP73
zv%-m+LmM-ROHe-4snCjQ%TN&etQV(1j|<diwwEpG%dJ!=j0ba+6?5>sc0&Vv#ra=_
zw~rhb+Hb4acWfA2DjnK+9l1^?tI7}(ICXyyDJIj0stjmN-Z>~JZyhC^6h*cHM6C|u
zc_qx7N=<*={bu~DBzGs^WAp2AgooLB&QB#QXn=)Rn%8T5V18}joVY95c)x3Q!|cIp
zV)L;KgtCHR<ORs7aASjsOaa^;HQh8Le1bVgtBX?5$}_H)?ELwp+btnsKOT4+`QiP+
z8U;l7UaFk3oU&1CsTuOS6i4P_Xt%xK_|ET2>AESLPn0O=m)|f=ZeRT0)l~ya&gQ&X
z++6V4ka*YTMrnIJ<CWbzf!@4y!PS`g;!VoKEko-72jhdYGKiW1cU#r7J7=uA`!A>`
znXlKbaHg%^Yv287eIp*(fBtZOdGN1Wgu<Ck&)@Kfpwl4x)g=CkaCebSZ%~mfjr9aq
zQ?@T8?b|{^nI2_I=S-lz@7PkVxyandPj9SYQav0AdrEY}Z8Jaf97sC0bZ6jh)E6L;
zDy(a$M0zsA?9Il8{{7kxegB`Xy~em-+mD`gK?U7-W<)Kt?_~2hLcUGtj6cB@bEH%m
zm$gDk0rpWD1dD|6<EP4}cfRBO?Dq95w0tHrtb#*0$GZ$-RxB`*Fif*deXv$V{HH1h
z=F6O8lU!%f)|4sTVf-}rB>B?BB|=kC-jz`E=hLVHpfF71pqsWzy3LMI<AO)Xli%v+
z(cwCDy5iqIYJFc1juxACkTe`->%>PKX@D*4S__lIMH4Ts+2}C584eJyuUE*VJ7ce-
z&s_`XQNJ%-bkn}#IxH@8ymSTIy$FL`t?mEZ^Sevs-sLc4(`~ixJ=8r<a9=G@IvDBj
zrjEZ?LFkgrZv&`Vk-0r6wbmZ5lkPzV3aH63m_Tay{20K6th;>*vjv=4gf|f|al{!$
z1}U|7d5v)4<e_giyPSQuzmw=jO3(kDz2=yP3z-u>)3BljGOue$&V3~&F+;CW5he=X
zWH%YctEyH-IIns7Q0ANZ^*@f<z?^WwH36EPNM1FP`+HT!f#-f}TUwUt(&|I8U&<Z>
z!VkJ%2@3j^ZSQQT-MkvCw0?DM`%;eVBH%dRo?(`M<tLW59_%;Er}ZhA-O3R{N%TH~
zeX#gQfHnRI$!EPy=pqJ(5$583S^a|IW#q^;p<x8n@tJ@yWPt?a!O>Ae^iO^*7Nu9F
zDQCs%_l-c+umutvQozdp5IX;y_Xy^2LT}$!I=TOGmp)`YH&XS@If$RS=9p?u;6t|k
zr#ZIX4|?}I(H}L&xZ`LA%~N7qJ&$X+5^H16ZJ$061DPUw*G#GC@+xN_=Hz4x5AIqE
z)C!jTIIU{&`8ZnMHeY`yUwphxirft=rW5mk1b)G{2rPCGKX^6I$~083dKTcyT$@@_
zk>X@D2n<=(iRVXILxk)FqigMFXvg0dDJ550JQRywMH^8xlyO-DSDY@!1T!*W#Sf{M
zHk1tv3s@4r6)ekg4@W7!g7w12c&U7Z>>fAjOas1qp3I6qSG1<S0+fy{>y_d9bj8yE
zv`KjFgqWauTZ0%m)<c3mcl(b;scm9NC$qouYM(_#ylo$dYR~925VchNmykri5lAcz
zLEd{~{!o}hq#rHwC6CrSB*xXyDvz)lfzWZ#pI#`{aWexcI411)_eERE0Mh)i<#-{D
zr%vg^0&DC~SDy+dGWpdZhukd5_*jb=2Fz*?)q}ka-^@Sg7ii@x5o6Hw>yu(I4Qe8x
z1*MwDiMIt3cEjtT^sj>?wD6$ed=OBKPCEsW>*sERzD_TaW3p(&9)1SR{_EDAqi$NX
z<ecXBwH;|}2fCxb68nTADAabgePZRT(|Ky#8e<EKpJl74sI4xx93I&1Ni^ZP_<VwL
zr10Rnk||K}^d=KAEn}Iz6A`eX1*PX2L*s-4TidoMgmvz&m^4*C^7$BlQlx*nR=rr6
zs;p()CziUO4Ni%*m8$%Upz-UwU@6AFt)>4sHyh@14(g}pKLZq;0hNrNs6VulfbO>c
z;tzBA?h8oRcmJ9_Zl@nB6n!NkjFkjM6rBD&!<fJnT4;#llQ@fe9q@XRi@sZ148c-l
zXtymYE=>8`p|8{iRCYvv|8<MvO1$>(O(xSLst>)f^YgyYr~O*dN{Qm-HYZ2M5GHWn
z8d8*~OhB?>#5_FSmGz5hO+*UN;B}DrONKoiKjM*~y7^0kaJ&V3-!9JBOog{&Lhc$m
zuvVYn=PX=te;^l@6Og8EI$O~DHQP^!L3yCIn<O%fVUMOaZf1F+N93BFx1HkM|6=Ps
zqnZl1a7`>&P*jj!6_6^ucLC|WBtU4=AxP-Gi3m!h2na~%y@Y@W0Ya5tLk}$!>Am+N
z-yFZWbMMUD`{N|6tRyS^IOjck?{_~h6r0#7cw83JPP+Xi<44z)8B<3!X=Az@fa-Fp
zo7S6`;h&h!d5glL?mSl{8HDT}uW-{W(HTC>>U<f+^!JYd7iPq0mm%W24_)=p`gU&X
zpYabmJpkYNZ5K?@RP}>t8hmUyeykxbQ&(HNxrS~`<S-B;Z0i%W`*b8usy{=RT*$YR
zzeABBh_km~4wq^cNn?FPDNNB;FKj0)V<y|jkf^q3%^cM1YnK3B5-X;WMj~LX)=LQJ
z-?xg(<@uzLXv_`hHA&KzAq8Jw7`WWkts?E`PsWkb4m)p~P(<}lB=L-0SqITCPn4CV
zbkzELsiu4Uh7>rHEG)#R75CLw{un7>tLuXHqZ@O2a+8CS`kV><s7iFm*`r%=|NGsv
z107S5<L)iBL&`t|iaE1B23qbMHg(1?O#Zkb6M><Zr<je#SU|VXghZ1qrH?+vnTXwM
z*c=IZa`{Z%D!<CaFmVwD!S8Gm%8;Plz#=>ATv0nQVIvbhv%gVarH5<KHmx20VUVAi
zAkyKfbMJR7vBF#CwBM<8>>9Kc<7!?G4DXp&EO68)VWhcmt!g96Uw<ed9`6*yx?=Te
z(T3|>ioxsm5z)y$&`aXMZ+Z?L6jv-Qbf8-6+of2dl0)%qYrOFxZGe^yQ(IlF#a9M$
zZBk!56>93S6A<~jD!jZ%{uSEQ7xm}DCZ$hgg(Br>KH`Rc#<>04gB$GqEIQx@=os}>
z<+BRf;EPSnY7Oj<(YRo|av7I^5@&6>LVEY2ZuNo;dF=+s#u`j{7PoX>f~*lMroshP
zrS@!T{<|=9<9pCLh};0ytq5tXIx~5+8(>zpNjod}U=o`_x`ejHvhF5&*<2WFi<gk{
zjNeAw&cjyAIB~9O&T^h?XIKX|5cQ-N-6pIJ@QVli_F8X4d5PADWH9{<cWDCuQ5INo
zf4;HF^%bFkYvhsha}rJ~ed8Kfy<3Q8qDcMjrOC9~YAQy`b~ce%+BCTb*(x?OqWGx)
z4OJC3xO7~iKf{8pWo_HhKpN!%0xDj>H>+j)cn-(-gZ+7pZv9uUqLeT|>i78cIW;=<
z16#C;PKQk2N*O>~{M}9pEA4Ww=KVjm&v3F?B$KUP`0nVcbbffbB4NO*k13)K&3f?a
z8%jT6cTH9plDb;rZY{PZbW6*!u6Tw6(#*_xDy#(llK?btv)&KH$&bGWet`5}$??2-
zVV!iP*~b46G|v5^PH<ybZ7K1vtOtj8^|oJv#K{aO?7MUA{<R8d-3%E7)7<JuvJ&O5
z#+9loz;7f%$ieZl$z!<AXp_|)5FS`4`+)AHb?f;vYD59!hgUM~UbP(-iNsC7A4aMv
zPS1N<W!o}X5OFR~!f>}*%lNT`&-~t<0)S>WFnO4#9#pz%EHaWN+wL@f>MG5uAmNrf
z;iCxabq;c;xR57M6~5sk?ZX>W8`#aBzxNO_bUgh_JuO*4vM^KdQ6a->o+<1?nQV{u
zQP=EW$N1nTfN57mDCs2vCxW&CQ<_IWBjN*a<ow%A7rKf+o&;_Ivbp|ia^AradvmTs
z*mLqM|9?Z8JO6UQ;RKuz`2Tr%{hr{x0?EMI!V9~c1b9)xOX?aP&ipN^3C7;i7$|xq
zX&?Y}BFOJDCK2UjUQqj#-r`kL>W4+?`E;K%4QjIh!<cKcdKKrfzq64w2W3UO?zrUF
z+RqPiZ{^OJsdSy8!ew%rj1HE!D`NL5Z6-#W$?~+%<=%f{V|{rPfVj}^I(yp+e`1tI
zQP%v+Z*WN|<g@ULBPv|=%PqZ4a0z*+^^R%br@gPxRF>AXCB){nBI~3<p~XV6`Jw2h
zYw3;%hRphEO1!MbJoE1ps#`O(E7WCL;uNwZth-z}xIBw06CSD$n8lyMD%k4CT^<TT
zS}oTgr!f{er^0CnAuUC%?87M(?xu%ok2~IJXpfBl!xQ6}2YY(Z=j?Ey2fm6um*cWu
zJm?$Tp3D-U40Q@LyP}Tw;O%1tM`d|zvfQ>zi>Tr*C&qb;#-AZ4Tgm93{UgtTnsanh
z>v2N(Ix@!TmW{(M?a(;StmZIe74PJt8&)srG%su49C^RxGS6G|aJwQU0=pK@Hagwl
zw4JkJ<q@@ZuV(vwe1Nx6+ikdtbjuEW2}S;@c-?Zd_xnZMC8d}Bz2<_4R$Q;$S^t1B
z>b`bE%X~5mjMnC&SK`+}BJmvySpN-;;;W+7gedr>g-%WN*n+YC6Dk`~E3VZ8DxZfQ
zxw}7&YR5ivO|CA~+r3W%Qc957J|xQc4i98O^yKZbjf$Sa%3qnu(->=MFHt0e?QT+_
z?gFtSI0_SOd%t9{q08W$x1fjTG*a#qa#G7fp>Ran+>11Wk|+RMK3tq}QE1^@2>MlT
zO83SM<5$Izev(e+H?=xksST+;afjIlBX#K}U`L&j1NhRMV!ZOqtUq`-UacFaAK#Ea
zW`uV?8ZXjNdmrnD9O7B&UM(`?6TsXrl$%4l$Wx`Q3JuEUgX(LDS38yGHXszIS8<H@
zSGV+MO*;FY|JfIOTaEeJc+BPSlfuiEHqzA1EIjJ<&oY_)QhFZdCx5eeB&_PpxMP!(
zuEvzz)Xbu3#TjYU(r_#KyllHT_Q5?d)-pVwZHMt!3`}G)4Oee`yNCg;8tnDE_=WZ1
zm5!Pj)+S*(?<iW%<*-a!XdC{3TWP+&F#GAynC4A695F}a<Ss!3SNk*8OHN5ZlW6@&
z4R2V15`VAEPueNzrl(Zs&B2V+?K-k6n7DTpN#|5U?oCMRrmIX@TWozHF&^o|1ld+!
zq49i8?hbwFl?gFxNq*E}oI;k#;SzkJbr7jT+2}CCj*po~8ALGQYl6uYxrua<uYE0(
zvU3RqqZs`gDt3=c&UC70NZILk-)&lWh^_vd3iBc#hVoz6mh!S(TGevGy!AO>ztBtZ
zJ}u5?M!%i3#d-)^7n{X*c9tn1`ePK<ozw>8>hI;OewHh!qy(|uHOxQPx&XP=HB4i*
zbtlH__MVyvc|1)BGpn*$Vj18(0KFKM&KcJnU-uuwFeD~qoB53w6{4UMW38sc8s`#m
zuBl&A6}(ncsz%TBe@S!pcPCbl`6oD6dg$+vE0`uYCp_5!(f1^Es&})ZeY;pZOH&K*
zK)ju3f^V)YuYWA~Q{I=5&^ns!$ll(H>VYbJ{<P1NL<UV6rF$~aPhcKG<+{&MX97(@
zEB-y>qaH&Z+V_qItOXxF2N^c?mYY&+X`5COJcXc&RqJV+KbLWYCQBuzNP<$9OE0EB
zaQ3`7E3&S2@*^3$?i)#eWso2g?v#oc3z?>H3VGIpuYMyjA)Ar%PC#rh1Ufl%lDUmZ
zx6u}OD09fTp)v7#Tkc9J<39xJyFOI)9H#|>%wBSOu%a4+xR+u@_!ZNUi;wZz_3-<>
z5o;@0g#_)_x2;m@KI-iXH$f~H^K55z1@DX^h#yRa8R+Ssj@m#7P^?f29kf)sppGaO
za5EeivFO$dTPX?a9=#b>b3q^nYaFOIMo<!nVj;^qoyUV-!t(yZ3Gl%bO&dX(_wPu>
z_6+j$A<r$7$(2G<cMtS*m9(`NZuVRTN=~*6S@cKGq1k$9Qjr&3J@D|KjU^M*9dFKy
zZkPTjvS|xixG)&;>V#60o!oE|AX%k3IVehZNIjRfbbDgN@*?BR6heGluBzswIfHU>
z#XsoXj`Ueisd}m1RB6iTgEoV55XCh`U5%<)8QLL>C%Kju^#FqHh>u=y=D09?)<ENv
zDQDQ0Hj!~-u^MmEB;j9?u{Z#5!V--6XIKD=S{=#Qe*rspz5t~vS^7Vd!bq*;F`#(B
zt?oDAwPig(+(4ugmjBjz0L3#xo&mI$V#45T;IqK?&;K%*7G8WzQ6y0RKUcY&=r%%A
z93tNrZ(S0z+-;QVzPqE_pC6Z6QOJg6!2P)#X6t*x*)6`6&q<f5F)=t##Vb36y90~M
zzZH5`o-M2hY^ntGid{&al&3U!`PV>p(;fMGQxgTOiy3ee>R>m+$%Q5{V}*iiX_W#=
z+jkF{qwJb^4)PbC6FiYG(+sM)(owz}OzZe!cBS&x*6;F<Qf8i=i>5U%-QaWDv*Jw6
zzLx9B8Rj)Gd<}(LW@jUpg;(_`6Tr1HQgZ4J<n?4ZAT0St>j{6lV4lA8FlyStvV}O*
zc+{V^myNt&x9)%`-Y0I$44kz=H!}PDkbV08`w8ZS&$9b*Hl8g~SB7$n_plA>8OFa1
zz;+c%^KFYf<y09&Q3-SP_$Xx>K3Rb>A4B5(PLfr#=Ahd9)lL;-c}7bl2~~`<M_V`K
ztPMp;*t=nVgw-drBNe>gloGm~a(>#VliknWGV)*MtZw%lghW0R{MZ;dt)x3#HzE5%
z){%L%(aVsf3PnZd13MY0dR30#A8?Bb7$+4`ywYj0QfT2$J1uK&BiT-HB-oYxQD&xZ
zyX;omTJr9%Q}+JKNKm=m9zvoOx;{rbHgKR6e@b~ccuzOsEX<)r|D-&t>DF~%klKSP
zGl+Ag^WjqM`{0I&ALp4GCP<$|cZEW%xko|kWze+Br}hojKfQifTTa~~jPbmDLs4If
z;4yq)Q~W_tn<F*WAt>`q!*{jG<DBcD0*0VxHcZr8f-z-iFI`Ryp9(4m5pBCJG;Qd&
zl&7ISCyE<qIx0M;8P<!0IB3iLE+IAAGtHhg6PTjM*vcj<NdHjj(O_pt>@HSKcPous
zyIg7T7SZc=r+YkcCwD$t^>LQ1VZKwK^MR>9L$UyK;`g%KQ8qW|wrjpdQHUOeWX~47
z35<|z%yAvt4Xnn0kPqT$ey8NxRqmg`I2W`Voan%6kX{I%o)EKTYDDZl;s3p%r}{MI
zJEISbP==hSOi>$_8zOaOQ0@n{S>AnB<y3pC3*f&Wd>3b1DLH_YoxGr<sl|YZ$o}9I
z-QycqjiV#BaX;|%oFDBoq3N&IsDu(P6K#ivKWZln``^ZNSOY*f8Nm*VnkQI9mBJA2
z>+9NLu=LvPoUq;1GP&rf4oTD2<pbWnkAj92vdK)@(XshPN`GgasD5AwR?}+t-a;L0
zxj;@Sk3o1ag{^rfk}G1{5hIKkLiK$Wnf5^hL5q3{@JV(pOIsiS6T#a#h2$@%-Z&|V
zyGC1yysi}*#gIbQd0)HxkA7w1QRJESnQH6maQLLSg8Wl_5pCFL)&zKJ(2uR8Yj+I1
z9!ll2Fns?gIoUO4<UcM|GW(?%O&jkKmIfy5EVX2z=wV1~WAkJ9&Qb56aY&5!z=E8j
z3(FbOw@A%=`$Vi;$U2(WO(dh;)l~MPDXj}*EFs`*!FsfZNHSqE%hHy~>4<Zbebl()
zOp5aN_2a7uN3m6j9Zyv<&gCfhpDN7Vz$b|#j+ZGG3mx9N$sX`L+F=G%_M9`Pk}DU=
z@Ck-|Re~YUcMauJCp{Ja>|u{I*E%2ixoKgKb<tsy_1j_y$a!p)x+vErm3*B+?v;)2
zN4aVtl>e%5&(~Q#`dhzIFEQzwQIY)hKK|&nzrV)b{)fQVQ@SKAild}Io!FbM{W0FC
zn2HI}50R*Gqs(Zsiv8hLV*h0jP4H%3n01r9M(>HktqF_GiRnr<Xk$?|<?UZ<P=^Oe
z8Y-9_1N1{tSeJjiU7L*?$cqEiw9z48c1i7_0`%u+M@k1XA-Sv$gu!{F{6)r<|3W-;
zbY?q}SGRUF+4k<Q1h8+Y>;B|cF?{PM)Egx5Exj&!5rIW`@yARPF~J<`9dV38&vTzZ
zpI<d4ifjUvxIfX$o}$@kS+BP<)$F%^@%F*XdC_XyP)aDlN2Lo7&|E=CJv@B<c4-<&
z6p9hnjn7fA6*oZZC8~qYN;G>{^Xxq15=$NzZUD(gVLQf0K{GY7yJJ5&RJW#)Dupm;
zKinJ0ce$GxFdbK=*4%*zA1_SQhH+M<lzvq7juNy(RAek06vAVPx8xKZgSEa7Fp5zP
zn&OhQzmp0OD}0;xiAAxnZY!`#x}B7)r1={h2F}WrjEEh}sy*&F3TK;k>zcFh&l@2F
zfPX@Gi7^DC*fVN)S!8}ZpYlNitoyW-#J#&jb%0Z`fq8f58u}%!`EHw?VAwh*u1G+&
zLTPpTzW?gWx`ckkTVI<s3>&Y>&J9_SZufG79uYk4p?x|wI+6V~u&iqx2Q+{Y{~aXD
zq-g(_B{n_|C}U!&vLNu3@=}q^pvJ2DUq$F$i{B=zcLeNxD(OZ5c{JBcy+(d?mm6TC
z3;h$>|4k_W_W<610d{x#$+h-YW8sy${n^Rnx=fI7@_6SnpZwJCMYxY$JNg1iYQIPZ
zeB-rwHC@GMxmNJ&T7?JCLTFl5Uj!!}YqxW)R*%ly39F$hr#kmps2S(wMN=*+yle{~
z+Re{{xq;5|a{S77r2XqGBET~RCm7Q-z+7=!KFCXHmOkMpbqYJVB))PqdsX(eAe3=0
zrT14X^m_kb@qr7g#@~ZDI7sF>I}$tDw*M(&9#Z)|L(lSZ0(qXhGNNVlm>!t0_}-+m
zP3^r-zB=M@!buswNa_d`BfP(5+_LT|<eovBm{&KG*4#fI4n6<)r!4#S^8rD4?gE)b
zwbD*&_=}yh0;lJ+<s|HzJ_OQyP@BG3>)~LA;?Ekj@9jO@V)sJAoh^DW+!Ww8{z0rU
z1EV#4298O3ufuGD*V<YBL88wyojdrs6nl+|FWXjtB|hKQsk#31oz=^6Y-?sEg!3>g
z-)eJbRpWV?qmWIFIL9;L(LZ2VLg8N|7yu~bxkkt)tF1?;Tp4%Lx^{Qinkl?Wk3J9s
z$9>Q)^#46eU85(I-2EQ{Pf@jG<pu{YhZX!`W}}wF_=c5Rj07%#f@DTUZc>6~K-FlR
zS069an+##U3+r8Ms4*Ss7gY1HUhVGiH~h$BqBBat#_S5Pz$qMy8P>Sa+9;do81Oio
zk+tnY*DPo7-jaFr{@&8rmq1!3<AKGSz7d7qzg(Y&Bje|!TjMX=wvV{NH#SBp$_d$L
zj`xM>?48zwd={Fg_0`9i8EhNk?1vJRzsNc~Vny3lhSy~7aR)9e%m@?_MEMMok4ndP
zr_e_JhrnjyGh6SmOlz7IZYq3f@R_+Ue9^JYw7$m5*COA`;$dJ%g%^xqqZ(rd9`qEj
zr!K=U%nWzD&+^*kx}&O{=-_W#K)H*nHBc$N#ViaSb`QG?@SV+~)e`SCf?{9+ab0(u
zExUwawC-iJQHVg-YYdgJn-0$aC%4tiR|@MQ09%~8-RzUlq3-l#WyT*A5rxuc%dAkf
zo;LCyt7tp2%QsS&oh`OD|68KwzO4+zwPPLd?2u3cZlljDrno$H=cJdfEa;JX6Hi>Z
zWXNZxI%UwojNc>ZiOo`?QnDGK&e_as$3^5Hm*z@&XIsQC!G)4E@r2YbMk?=_t|q;m
zsWm4yw4JWadI*;CT8#eQ!j>`t2f25V(~yaC^k(CwIp6B4lz<cdox+;1!e-&u&aol+
zES!pJaXVKHLuFqWnuRq`XioM`i`N%^-6&n&3bMZN?lKEVt;5tW_ow-uloQa3vod{f
za=-j+xJJr^lJtOweiOxh1<Ue8CKNI}^IKeiQEloVB7YgbEV?k)fLBFmD(G;>ySjRq
zgN)tlGgZU0`^yGK2JWx~RR5I06%(f<z;5OeJ>*7xOrqCNp3fZPfH_(1UcZ?t`?g&N
zlo~??!F`lFEPF(<1)k7w#yrxFXpfZH@l@i@2wbDg_R!R?5hq8(@;qOFy-;P7T{UWT
zk<9<L-n^8a=7^=g9Vk?-<LtgwE<dz8nAC6#c-xF~ZQ0O1GW3nKuATM!hi?8E=A8U#
zp3)R^@3ubr6((N%;-VSfhjnfvCXMe9DEpsT^4J6>s<`v3{${QHYuQt4cSAZ=R(8-(
zZ!g~AAxKth!Et#x0g_o`Ipe2~CQdY6O?P{wa8?%*fL44dV}bIbdamHL+<Qyh{J2Jk
z_jR*Z9mrT@dr-W1JO7BReQn(_RQESM=PR)tZ4B|<aK?xd6hGvhz@bK&YH3fZk6N==
zO<!rTd*k};su9?RaZP!Bqgh<*Nojq->_{F_Wkrd*{dj&#4akvdbmoC$^x<hCtf1#*
z-W^eW2kUBtDbeHUpS`hObR&FtIlbpzO8qqHy4z&W`#X0FVEHadDR?3BXHo`|(dx`Z
z9}eUcOjdwWQ{$|#d;X>_M(M)s4T;1UF;;B~3Z%HM-@bo2_EcX2UWScJqai?X*%j9z
zsyCrP<GCnKQCtR6&<o<+{x=TQb4hg5BfLM;9B{9=KTEW4bgo?U=rN9i(}?WVOrx6t
ze)SZ2#9@ooP1J@)Y&(<>m9uH!oC_4$3Z$W9A1;E3<=AH51&q-^R_yd})M9HjC{10J
zW}dxk?@%^(YU4zerRfqu2Ao@+O&Gj%5<<wt!6Js$jZ4u|rM~fkf{8@VN<28;dyG<D
zLyZaj3TB#4v>ZhZDxaaEo*$Q9us)iLE9zmBc}M-{EI+)ao5aKomKuP&9~b4kfgf*F
z|I^%NECni2@RPDwA5P;nWgJOA;fGDrhqTCs#g>`2g%l$n82&W{di3Ez*?$Hfwpl(_
zVP+?DHhqFi_+#-<xuAuJ#gtczYwWc%SjcoOR;}<A{YO6MIb~Z#GH@;+v7uk9lf7t*
zt7srgKri;coC&U6KubB20rNbU<jm`jaCX!~?{%Ozs~!|l20Xo4{ugWS|4(oKr>p+o
z5cZ$5KIa4DDv{U!$cz1-Bnkr%XF34z8oJ~7iv!qqV2Qq}EBc0m6CbLEjg(eAVt`i7
z3tSC968g3zo9<xe*W9l9c;3NoVE*;eCRg&gbr5}Xgi*0h?~^?}P45trMbOBa%i?4R
z<R~+o?{BkaI@#lkTIQdpWt$;wVyrtLqkUdH&0#>%d-YC{mn^d5&9V(f=9~K|_gL;1
zqJzH&Pfu%4dYTF2-W)x7P>ziy^Q&|kFt{xUhsUgQY=~IBX1j9I)DaWp_80y7Gdgx*
z`u>XPe+WJwpQE3WnY>YYYh-=yg?dV1Ry9%c>}9vsUnCELp15FJccc$8O+Bpg#tjkE
zO&e`BF8_#-xOpb#`&ER<+;jAYY6f}Z!I4}b6O{Ul?%-nx1lzP4X>=?pCETt;_bjsh
zc0nCx)(7^R-I|?Iv?e6q&et~SaM5q(yAB_*+9AtjzlzBt{DyW)=muwtknbe0zUFjK
ziPW4OhC@FN^-MM;m0Q{{JzP2YqT`l<&i)e3u))a4aX-r6P#m>uw6kP|jTjuSTm1~}
z)w*Sl?)8lgP=@yMy-i!IEv2rT$)&RrQ_*THX>735TsH`VaFfXAD~&OW0~Ujat}&%*
z$s}oBJ*CSI{%oP;^5%YTmbT24zrKF1Q}LRho4V|gBI}^Rmk%m?)s2uvoXc_R8(D!r
z&ZW`A%hNV8X8m#DVvw=gPm=E~4==~`AgP-Eiqs(MP{dSiC9&X3YTavfwSuC#uPXq=
zm$+-C8la>}`nG5Chyz%C3_!X{ItuOZq3~W<B92bkGbgkON+(Ji<`c#Wy?zsVf;1=j
zWTGY#|HunPTMO}84n$LA-a^+(Hd&~xh$cugs@hI}q8qQ`PmbeLrr-KLQZDOTlJfhc
ztOWa2^h(&xM_Hybk1sF&U108XnUJJ6$Vu^{a->R&){X<O4!@kNu*>bk4_)EDECG?5
z%1I1~!Q@=yq2GFL8H+(FSN^#|i*2q>ILj7l?l^108*5%XfaFJI^6nzytnpw<Xqi}n
z<kv?@n_OyA&bk{0%w|AGO|6k0nvz}yq1!nv&au!(L~`0Lco<k+Jw3WNGscZ&nL0er
zk6pPMBj{SnC&`6j)eJ<{jHaM|x`wPur$T=l?3v<4E<E(80&%T$LD2W@GUE5C)`TTQ
zlSgtQNEl|kk^i={R4zW8=pZx3rtp0W{`&ZpnBrM2ku^jP`Aig6-mDM4;uU$?-uslb
z+RG`c#6*f{nYZw{W{X&vV{aEq>3)ju>Yp8+hhoPi7LT`YaAYxUtGokF9-R;qbNLoZ
zJEw2wUW?l2N?u)~dq;G7B4DL;V83n*H<f8D=b;x^pF3BTE_A_8PzIk+idVeLDmXd3
zko<9TPE^V8uFboX(vV2TXI`>IleNP;gdM-vjoi#g3UX;He;1376>tAROVS3BgT#9-
zby?%db{}>G+Li@coL8340{oqvIS$5`sHJIx07HcW3*V*f8xJSZ7yxKtFNTsv%l?Ya
z0Gaq<nKw3Wg*Q)~AidA-i2Ze1?Ry%X!L)zHj#I`jFow9cTyxDx-C0ChP&po$Ptn&m
z9RoLCC0>@S@%9v4yr|EyQi&6BGt+*s$rt0h>9op2<KC%u<Vw0<8P(=1J|y95h&Rma
zL&c4uunmztNgKPm+FTN6-2pB^^(kitf(s;=5h0a<-WZNfyv&sTlKQ;??%$s6jC<H3
z+;9GNsQ1A|=*^DV0r^z?{tH>o6Y{hPZJIJZ1H;_ABT^V?rK&JVi@5b!j(B+*x~|J%
zSL<4HF_{yyg7{hgk6hOLQO!L3@Nx`*y(Bhq3Qs61@ww~VUGM=df>oL+{tp49k*0!W
zrc7tQtq93KL+ck7*5#j|6AjC~GO*^pey4YbX(Kjy{MXvIOWaoSOd3s3dbAOIRK-cg
zQ4|z8UAf$QR!sdkrO2jGP|)AtaYylo*|G7}WQJpa=KUf1hp^I)x`qoPjuepYcC=)*
zW2yI$XmFT%sBRh@yc#$CQLlFVqPRZfbE8D=ns<PXgK)$q5N|386$Lm^Vbo!dt>K|f
z(D+=J<cwpow<gfl99v>-mz2ihLO4O`j!q!ntwWP39kZoV6&gT8Q2t6kPHX!V86PwQ
z5UBVzu7yx;i9BRG9r~%O$jU&Rnp!`J8lde4d0X&$QS3<ZucY_Kr+ET-&77fI22*~=
zXUJs6wjB`b7P>9~tj8^)_bo7YRZRNIKp<$_;E-c2Hs`eDP_%DP&rtA%9bUJ)O|rzU
zW$O@#zs&R@`%KCK)W+~~w!#p#N~X}3KbIhHCF>eMX9F4uAuFn%4;bz@>c_>*{0+Ip
z<!p*9?FB)je&C~YC@`6Bq@7cbaVc*18)2%9DQ6pq;8-41k+KUcLOG!@>r5!p{iiQ)
zi{LrZ_~Wy@_T~`hAnbnSZ$5|9X9{$Z<jRWlTc6L__RF|wI$62uz*pE9rw#7=%2%&B
zZ)dPKb9@K>f8pa)!U*BtQs8>oq}n~*t}oUA^sSsF{F0_qZ__Kc3Q+E<)qmnX0~F5p
z`@4n}$w&d^*5;&dg|p`u`vV2HTEEc0uJiP=q$lvkN+ICkYr>`sTyw07*dz<k^Zz4>
zlLH9$&~K=G^JtYw>Yv)vl~Mq4V;RN;L$@?sQh=?-L@p313ph?b3isA3Hgu#6+q;PT
zwXKms><~588mI)0A21rT=YszUCMp_gkPC8@ok{a6q8)FPgp6X3qe`a+QGOX{7uL^3
z%Kl;RheWkjg1CkX_Yaw|$lV_8itsB#-?zT0tu-FcRM84d0VEa)b3$$|C<*qt?I~&G
zsYwS>lgNHhtNfx8f{B*l<oxQu=}qnaUOT-~ukYCfX)X1iCG^{+-Pn`>^)t$a*`fqU
zZQY&M75tY0LD0n3&=RPC_)I$~W%Ygk@qM8s?N9hb8P5`Pg&F&-vH(*71&Tt~23O|0
zH|U5y{Q_AMrQKC^)V&c^zQ^G|fuPNgtK(uzAqi1wXC?f63=9?Gqyitc3x#r4QuVy0
zN|e7zg72+4<<4H`y~Tqpq>ej`Bewb{c9LCQyK>RW*D_AIYu7N~0yJl)abXNTB8oZ;
z%?`cO?;GsNLmy`z$bh<8`ICS&19M5qFD<m;mX5y8{+oI}Hv9MU-&@*yXl)q3--T#@
zuYoKD>ar!D%3&&MSnKMyeveSaVl&V^=L3RulTUiaE8|5>k`*3Qt=Bx|Ckhr0w|6=1
z;nAaTq&DSaRP$;oqSWT6WaL~zw2hX*UJFOkPi{-AY#6DG9ruB*q`p-zVLmV)lpE|j
z|E{g_ovgRc_w)MnHE2aeB3$>l!~banPeaNy|KYOzN@;XlT&W}L)92ZeQ9PKtm*Xv!
zL@fs=knw{j@vvodo%Vv%Qz#F&*32z9Ls4zLN^g~^8Qo*6dU}Z&?E91jr}WAyZr>&N
zu4k&{tDI%e@~6+x6(^I%0bdGVMOgp2cL`E;ZpFq6n!Q8|a1c0m<=dN@SXN#dKEz|J
z6>E$xJg{CA?@8R9Qb*;xCxwS0#UbClN@5Vy>ixKpGQv)e&gn7Ldn+1rM12)CB(<Qe
zxoT$vQI>bO<BIeVAH{m#v%uNS>0*ew5nyWoCCc~I)9F`IN>4p@Tlod`UCjTjl`E@O
z`8Xh1C}Jh3x9ib#_$Og?7SED=$Eojm>O>J+U=q!@aY<KbiJ)!$FQStf__;#hTBWno
zrJ+|AfT&~@7l-)+Cz57N`XK5Kg@mUYia#f~-JoLVZbhf}VyVfPZaU6x{;rJ&$cjgu
zW~2tKRstlqhMLBJTbJ%t!15$j2Gzzhpsh{4E^9*hJO(Z*YhL!mv@fI(oiU@pq`8Lm
zlg3s370&)M@dsd8XRVOhXw6IWwN@?Guj{K6CH*LAm<@6`sWCC@F^1_^xzLK&V={_8
z#PWCXSgn|sqf#DiH+%<dp>~Hp=#^g?x;=jaXpYlm7hipj<kl_wrnFWhY^bw8Z7#?W
zbu>>NCL4wrSK?pxXD285zIYu|Tqm3<IGp~SCOB5`9?47Vd9<U^)MVWS7ZveI8Jp_Q
z4Cx}E0p)}q_Qd#J7$wYjm8H&ScY(KdNTG~G+rk~M4~mF~yvybn2HSoF3KFsK@B7Cn
zI-J7>RuaJ~H*!GYV5_(5J)K_aE_bey@CUt)8Cs(BdSrC8b&6(eZ5|@C-5})yQMHYQ
zVW!Ft-?H;=4p3%4{@j9l5BVRL=R}LOpF@5`j@b#9YM?R~QNosFEH!GeeUI444b3qW
zr>}0c)!Z&id%tb17udZF+kJe-{qm|ag#jX?$$~^zA|mttjw|{rXHY(VUI1*0&oT-#
z^QJHe9)bMCOm*%my*Ch4F?Lg%8D%NCeLucnlvgU*)A~@<6lYjc88nOO8u@xlq~bld
z8BV)<-g81yGGOAZ57e6{HRAXx53#cQ5Zp$m^A?^UvY9UkD_({P8Y~Z@xcjI~i6YXl
z2)j+5yPB-isuFfU>cjZEKDbSx+h=%GHOEQ~a0O#s&Fh!_OurGYUpe2%%(Ht;OQO$`
zKf9wcFMPLMa|}CHmFmFmM2w^L(B>h0i><Qzz$|5;$5(8E2u=1Pcxyt|L#00_xXH;%
zTXdWC9Wopk((3hqma%u<Y9DrW=*I<hm?Z2j7OOJ$r411(51OAq)BvPt#}3G*G(s;#
z3kLy7D-#>qbbu1J-cWBh<<S(2&YmHEu|G=a&`Vg`=&)vr15M(*b=rg_aqf8i>|$zj
zZCJ=$krAvoujup{hJ}b&g}(WZ#SR1d4j*^`0#LTIToCcMMOHjOEV@y+Ouh#fcMYB%
zDe(h2VocQh3H8$a<`etc3wTw&Cd6z03;hjK?Y444^~GoiwT7Z|u9)}$aZubJ*@yYj
zr5q~@h6RMuX#RLTbWB1kf(C+bCwuo4x{Jy_T-q`?zLF#3B)CH!tV~o(@AOb@O%47<
zWuAw#`e{*N_38#zB}wBi7q=08I+?C{<NzS1{|(cU{8u{b8E)PIYUh8h<$o{U0dE42
zw*f&d=q|GZ{$#0-Rr<HJ&JJz;atT<O3jb?k{{Qr=fm)j2{~qoF@4pB8tNd@!OGcLI
znrDaQ7IijNp{l*%jIceTyoYio>mdV0{y;gddc-(|ADl+PoCkMf7OvkI7Mrvp&mtTV
z1*!CZE7{)|U5*bSdv`s0({Zs$LKd_kt%68&;uQbyzHl&AM_}TWS)k15>VZuY2RbqE
z^Bdc)qVlRHp1@bN2vuHGsIvJ@;(_cAc`+^ai6*Bvzd7eO?~FWe;dS+-JX+;ss$SA%
zd+ZH$UDA=k7|(a-SWdXM7FXH<>WC-Z=PvBc>xb9Sn*`T6`kz-)@%DZ_0O_q=)I3m2
zu-@B`UK)u_>dJpKFF%(cI^Dn$#v)NI1^^&tQgUi72kvo6{`OHhtpShJh+QZ*$|~DM
z^Qee1*egtexePq%6H&u8qvWz-O<Z{+b|k8WYS$ORiETi7+Pb9Rk)dI}1;?nf4*@F8
z24`wtFu6(6U3SNdS~xUUYq+XZOZA4ceBrA3{Bl0*&;y>$y*A~1${HM;7}!qfMQjYV
zWWBB^nS0$x`TVuxQSiFhB~3@hy2hI>uJIb-06uJ?R-Wx2BDYNrktubLm@EHC8j;N)
zPE4JX`a0fka=z8CR9a8D@_;R={ri9nWVB9KP+>9#JuvU#kUP=9=tDfMr2dS`CVb00
zHhWR>Tu$1Me3^lmb-vq`lJ?E@*2Hm9*Dh_^(ltCxp<y(S(mO(g(K~C5Ov&x{8aZo(
zNH^TKyQkR8$!x;X<bw%QvT$@(;}R2A#2mv^A5G~POSOoR^!0H`Ajdg`Y;Lu-s~uua
zei#KrEPHREW9HDYMbnAPPn>e15}uQaER7@0ATS*jy3;+=s9m_SxK-4zxKZ-)$&w4d
zE@5mBb<8`1ASd@$*WrT%5Pg3MN=Rgf@>>96vPzc?(o@+c@EDG-Y7Yjg*|~|CW!o(W
z?-Qi%?!X+suH4JwrK{#0owvqJ?9z%_lZw9!O2Vo0^(05o(LDS=M6I=^IW4%MQrD9u
z^j-4vBAhBWr%;NkDSD2+dT)Leho8TII6WfQJ_EO6d~>7vD%sYXaMHV5Z?#9IR<rpl
zkC{r9A6x_#QH#C92|-1znUGzK6-ad|&>4~iL3Mh9q&O&54@E?ijp4m{R7)zNCow`%
z9gZ3HeQ9Hwe9$DvG1W=1bgNKm<Cc=r{LmON!)B}f+n_P)!FzKbbK)mq16+nZgSwg~
zA*uJt!_2Ua+40r&n65q$YsD+KP%}Z3^5ow|xoF}s-EP_U!~G%l4=H1sj@$B>U1zS&
z%JNHbhA#{l?HImeUK=<GBEsaNdTz)k&VAmt7l(Ez5Ejj*1va7azIuah;1}<}KjE-;
zQ0WEC8Ncz>xnQ5)B$(V{-ddO;{|Y5bV3oR%5?I}{waGOaWndFa<L<?wvMZ`LdY@V7
z^5kh6b>%A7o#Ew}%!H?|YI_hDvNpLd9DjvwndeRBhvl81HaahVj0HVlwZc7oSFy9F
zCtk9P8+j)oQZUb9uB;Ym;jCG!l4_DqplgQ*Eki3Le&C4>@~}G3dR~Dt)G}aPG63CI
z0kScy{&KRbjEiO=(!4(Q3goZ!cM8Bdrh46SO3WLWu`3azD9($|!mC(wWadjZL4c^H
zu@A966-662jzv?|L-M0;%>jGC->1#Sux;M#FJODt8Z(Y_jzBt~$n(`s$NFeVpFX(d
zd!3h6sYSCIM}&8%W=3gOP4#vQcC+r`JmU20?`{o_Wcd?B|H3_V<+PeC*{!-o8;HA`
zC<Dz)!6OJhYV2Ua36jlINWGvLkg>FE1Ymu_0~Moa1=8iLLPzx+tVlM$Ww4Cj1J5ez
zCwdPNisRk?L}fw?Md>+@Q>{(m#Vm?-R+Z>pMR78mvFP2FPUvzXSc%lV0j*VP3%-be
ziVO6GX#*+a1vS`AXnYRHB`F7?xc#6wT~of7=G!P&HH#_D(9p55@Q%Esr~G}ifr&AL
zq9ScdomErlx%mFENuG~DR|+lar5bp0w~5^Mi8O2PD+B5-S;(0gt)`Ksj#!7SuZya>
zuMX!evvgo`a88!GZoq6KAbgAOz7Wy{Ft*b-WH#!J2Ns0C{iHl(kXkX4CuWW|%LDCc
zC<C>cq4B@PU9v==y36_oXbJs~OAdf<_u`+d{I7#cax(8<1J{Ar2q+5&&RD!RA-0lU
z6x$%+yjEl?|2Jj5RsU}oyW{#lK1KQuL2Op_SdD$!VLAZ*c+WK};NT>e-ray?BF^-4
zJ1=~W@!T;_wa5wbNhIz?50Bq<Ne(8Q!^ES^?orqMbw%3GMb{mHl1s7Y9OmJLP-2%H
z$6uR#<=0hLnCmV3wvc_@=7;n6j=$M=Pk~r^y5JNPhQ3DdG;X?UN_=Bzt&kZJou}tY
zo2hnvc@7{L2xY2=XYju28aG36bu;~HIe+hQt)kR#h<Fdw4o_8=#+<po9Y=@>$fna3
z!&-WdmACsnc--PLHQ>&>70&t7LZkf^O32zuhL6qk4a~szv7AR2%=;lejh_vL42qbc
zlMgaaPY(dRFP7DeXKF&}*SGiI@N^AdPj<A1cI~k|T@#sAXFC-I8xna<jQjmR=T(Z#
z1M=UOELD4qH;Z0d9GK`IM4<cv0>r0HoB9v1?%a!yeoXZE(WtV_Hq@{r7)^g4R>~nb
zw?2!7rhvSwyy+Z}N7BYJ#f$e5AUmtyV8zZ79`4h4!wm!2kKL(yAcNVih|ebB?jP_I
zrCt#U-Y1(1suRvDw*3NLSkD9{XY`{_dQ$aajcyyJ#0?z_{29g)cJ+;<a-|!azgHK@
zJ0%fCP2pmB>T9y`T7L9bh6=<Ji3=yTui&!!rd1x#MomQLbQ4UHxYry+-F!f7hkNQj
zl!F%3YgZ4eMFz#*OZt@t>Fdc(kH}-Do<jTfDY9xMZSS^Efd+*sDrA$9^D1KEo!V;R
zAAvX|n7-FD;<K%q2FxQelC{^79ey$?L2FJb4`{*lQ)P)M4_y{IC2<^tF_P`Q$Wj;C
z09V;`Z9!N(cY}-AYU*MZE9;8b3y!tv>=BkLqiW*5dxPZPYb3~ERnq!xV5`o~7Xv1Q
z%EH?;rkU$MeZ*(uw#)^seOX3ZmYv-)TN^{v|IU5B`#VNSQL(pTcHLcQm}S|y)dwYi
z59!pzo4!O{n$7`t*pC<YB3Np2^h}MkN_Zzl-a^myP(4?IL$PO;M#L3XL#421CiB&<
zB+D8|R2<)_iCC<YW|fBL#eVlGXZ<DfE3C>>=(?F_=2?EDhZe=vf}vP9vZCGC`x`$P
zg7w~^BU+)$^Pr?)_V#wuK!1>*=T)1DpYZKwEs?I40A;J0J$$`vwESj@d}zA)Ozzq#
zcFZ|1zW;)EMMY-A+{vTuEJ3=|&Lv5AygI6^e}&vBUAX>fzL6~~H=@L~7S<>YJydPZ
z*Tn|wjcFeK8LbBLY^=TVdS-W2lc@nx%}5ueYY?=taZ`=&f>^7rM)3EaC)#S+_SOPD
zDffjF+%dxpV6-TC82bwbs3g*lI+q<7kfGEW_>eQ8@YXBb^r{c(tJy9ns~MW9c&CF_
zb<5xmv!#B%q3+6w(D~$l2u{`&DRZH5S~rOxRS3o<RxRg|TD@aD3=dE60>@i#7|4Ui
zoG6SoG45!u?*fY@-RQ@2+$OqdBcs`oUwSt5E(}W9n3{$+<jYnEp&73(e){Fb$sZT5
z$k)!l-s$8pz7OMom{sv&M-gDTqG&4H4&~)owV%vv+2mGK%I=()H~B741B2<2wfku*
zOaqGUQN@gb;E040z-mPclx(JEt!{}b>pqvbO7%htAWbPy)-JG2>1iW{#IgR=FUevy
zwDIW4UH*MHvuTl(<ywO%TZ)*Zc^L~~k<I?kx@PqrSO+fHhFOVfhE&$#3CHH)h4=X&
zw`tyRWb%uTILx9_!V3FC&=!1!#=bf&ZH*xjuWcQpttY+||NN-f!UOQO4vjbCnJ{19
zhq;xee)*)gpxpyW`KC`H|I7Z|(I?6dxjcG%rT5X7vnS}7-+#PMkw8B4J}`Q^l<qZ?
z$f&vnu=UfW60N+?01IylKeHV-?YL>Qp#ludfc~-VKurd?gISl@v}CI093Rscrxu(4
z=yGN<sUR0n3*v8r0?^UX4&~TPq0=?ADD#+(ouUeM1Iw|Pau&R|CHJ!)An||faL7s_
zF486!emgUyIoWF~lOm_~S0&}Me{`x)xyBBHi0Z`n2ihnu+>jIefk~VkNO(bR=&Q5@
zjFGB%FNTum>2GpA3<h^-vTjDWq#V8J{S~y0QxycymGq1Xe;YR5{L1Crvs$i31lV3K
zLB;^%;SPr?mDlegXn(Ku6&f@!fF?||>Cj67HSHR17@Dk>XaRVMP-9?uDf9=AEF?V-
zKb;k)sSdtP32-7Pr9*|8Ic$(~j6-V^E{oNXjF}C}YTqQ9!-I*Pt##n5AXb)*jFXeW
zkJlnU@$D1>1jpn(*F5LJh@hwk%+tsnY(PY0J}M2GTa-Tp1p4l#7L@Xr-+_-V7~cMo
z4{{|Gbvr4Ys@<6zi)Jc>pt$GHplK6?^&_<uFt*t<-x7_hMn}|3=x1!v>N{Z2;KePH
zZq_9RXzh~)z;Z+h)Oevse87-<SdJN}0RP`a{!e8F@^E$}2fheccX{Z;bVg)W1(<um
zKTn9qs%ES9MsoU@faFC6G+hmUYkqol=$0DxQ=}u8$W}YTHl8Bmkp0b8<4-3<!};)^
z9cdd;n!n$JE#~@j;Uj!5S>pL)yL>@{DtY`}A=%}$0G<-M_n8KoE%CP=ww>a#Vrq7_
zGrND&rstb+sPE9oYG6gX0ou+}2kbDTY%xnmWGVClf)|F}m;WKSCX;Qs^hX+XxcNRk
zBe<z8e|4aJO=fp<kbl!z9YJ+P?I$~sT)EpdIK`NKXsSqYryF^)V=Pe~<rz(`Zf?;=
z6jR#Iby(hEQEeaNwZ}*AFdYFT)q2mfUvKrGGr!KU9LQV<6T=SeVVEDG7y4DCGuU32
z>?`u11AL1_2=D_)j9(rfDW6Z=S+c-OF3f*Y)kn5@E}c8ycwBECKGm7vS!R^#*S^wF
zEtTBSawH(FjGKYoRC&m~kI3a&TR&@WKIaHdu(Y$>$A!&lvKhL+VayMtKd@f<92v|G
zJ(}N=Kx13J9Xw4rd~r&?R%zx}-~M9|Y0o%NS#ULZ7ZIs-$+Yxjr5J4t5xzJh`3qPE
zqMx0A+!VI8zbksCB#W{YL@JLX{W^cu3spcbFI6#ibN?Zj1BR0{T}ft*imCugX9KGj
zSF>~<rGwE4%Eo`-LkAD6R7dZ618lHnzAm7yYZ^MoXE+kZ6^-=rJWA4E|BZW8;}kLT
zrYN`8^0zEEF|)uL{GNiqj`^P}1Ggn_ZS2j7`YNda^q%G9*oSjNH&R|gr~F}9s(LNl
zs~73jrM8MPx(B#AD?e-dWi!6yh>G&eTj?tr^GlEBw%2rkNNW+!k|94c*No>6?T&g@
zh$u+>au;hs(nz;KT2DQ81N<a~CH_O8kgQ!7vB+xw>EyjmUS1gk7S;bB0^)AZSv9>M
zj{|5h<uiW~s-PRIowUpXX#NSZC5Udg_@bT&vxwLCqSGt72yA%17iDlnbLdz+uwFxm
zzJ0M_mM2Qg6n-nxys**LpT7R$L*k3fM#RQ3@-t1264f^?Vo-;bGt~0P1cOiP5~d8d
zwiJB#v0HFQ$2VykwZtFmN}j(E9<xz^KO{{9!yu`7Ugq?*5>ebV)vErN&$WMDDaAU{
z+$Q58fbKsfa4eWRWw|3Oq8ao&w|}@q$yGk;U^*Ak;eI6O1hyL$!mB9kxJ3n%=jRlr
zwXS^boegqgG%*~rt0oOv*f0|bhg#PdsqF(P+1kFR>^duOyVG=vm6kH3HoEo0z3o$a
zE{1`j3n^<aPlUSP7qaW9MkkJ|^-b-#$<c`UM+u?&&VL6RRq0q23BCzx8=diRWqu#-
zE*M|X&)1Eg93NH2xap`qIL+d<X57wli($8Q0{8nuN3Qsb8Y(sn85S{C<01T_VWEGg
zan(V6Y2zCFB?lRt%kLFjmALe&VbC(eM)kVlHa=;;;>12ZCds7ica#n<Mt*`5P<D!R
z2T)UzW&b0H(*xTr?8DS;&cE9ym!PO?Op&gJQ=GQ(oflM6Bi3xu5i=vJInlH*j!7Z%
ztkq~pqBUp3-)ZkrU|V0GT1ZWD3?Mn?Ytta|?A@`<^2Fl7(0D@b@e~2?x02fVZ-*&R
z+*O0G^;bxX<cPl2G9#RL<{AtR1<HZ7CSjXVOFyb*ULAj2RXl0EZ7T&?->>3Jc>e2D
z%(_fYLs<JfL&2!I+>^xDz3HlMrgrsLdgdD<c3%6}r6F}TbmNq<9utMh6N%|HGI6w&
zpI?RC(HH)U%J5Oiy(6ux4LL7wXGEVxc4ol1Ju}&T9$&|}nVRQc$@#9e4-Xv7<`-G6
zl(jFzgGb^a4B=8qXk}!@*-jUno7;NMMp54NVacIy7Bc}+l5}{mGeCw3IUp$m;q^1*
zW3PdjPtMH-*@MrE+>O62ZxXrIpoji?pe2vto0fP(J(>Mfqd2@1>C;hR>BjzFj|2PX
zw<mgJi2i0bsMTl*OEUzdFtbP$D9!j!{5-c_yg@And~vQCd@|5KHRBQ1Vc^6e`BIw$
zsc50@pRtn_nYQ=(%?aD>diL_1(S=z-PNL<8hYQUlo6`;{*GMJpMlxi4aN_7}AQ1d5
ztg@Fs&c-Rx7MoOW`OqYVd6^&N(3$&b)p*n^L}L9G8!!+sn=PF)%lxqhUj=%v_u8O~
zQjFjBXAdBd>Y|^VCPja|%1VTm=L!dwob<hyNu}iq;&OvQb*mM^jpDf}`H?;^w<T~9
zV_|ezIzgMXUpucPf!mFfdyQlWDdmdD-s|7jbKiSdUioc-?MXDdFP@?e;;UHzlzfN$
znmq*IVU!$+Y(6yH<WI|49<{1ti&!=gRL3-yq$N0SD6k8wyKskfopF)c3D?k1q%$VX
ztf8!+>_h!x7ux<LJb#IX*TyTC-!(#)-y1C!nd&K^yVG4UWY`(IxF3mk#uI>EfHlbv
z5$O%oksgeZ7cH2P*{1jQ)2s$J9AfS9W|_fqU~jQxMgFYVXlENLqBBOC*n%nW#>!R)
z=*xa4qj_xi8k67EhtOW=;Hv<2ZN0hm4*g`;P5p&TPJ#;-NF16ObrjE+($O`;6o~fL
z7LKxPAPB<$O{q<<F(o5*A6`z=q7>LP#x5>AbY!_gu>^9=G-W$rmXdP4*$<&Nr<|3<
zeeg=Q*$wO)Lua|Sk8{O&u?7i`^t&7Evc9=&ydhiuyH+z^ta(HS!X&16q=CSC*fawQ
zP@GQ?kcFxnXEI*ai<(CM+RY*o0IX24yJaSEz}M^7kbP6Kv-xs+M0Qq3T$uEBDaE0w
zN}N)7`psTr5beJ8Lfs@@(Ze%Olg$*~a&EF;tH5MFo8_3AQ|;QEA4I=^dfpQ)qg}5<
z*U&#d<azQ@h~S+e=z-CMaJoHF7tO&)siqh{ymTh|yWV7a*=?LL?5Q<8JdazaDiz6N
z=V-a}5*nH6<L@F<H4p^lxl<D`MjK5V{fhDtL!4F~Ufn#d7ib&&x8eQ&Jj*2f>wFUc
z?d*T6yZ?^F(p!pT`u_&ZQUq>kLof1~={KU8%6tSqNW{0+K4rXi-!E)KIx_0xvSJLj
zDC9zV5t`d{r!&XYEGmb_zX=XQP<|89O;W!0EnEUK^Vyg$wz;~%nE@2zolo&D1~>+F
zGs5wa_VfttrLI3ttFaV;YGcC25aYsc4bfpQgSyDw;ME0Mn(MLXYo{yM2Rz;mryKtv
zaE_2oa{3+Uc4g#?H*0caJ|Ov7EqmBG_Ret7aQavTZxg>Op5Ey<cs@?JS`{_-uC^`2
ztN2QhAuNhE_+X>1xr19>r_Qh3%cMu$OF*mZ80(%K%w<enqc-sLY`_PI*A+Nwf5^-1
zG_Kf*sn+=3dNmfHJTolkEwLfImZD`d-M438VH8L8tAefOolM#Ux5!Ugx~1{#ASajz
z?7e3x=Ep@~%*Q3Y7>E7<pG(-_IgW!jUTm2pBPiUitN5i^uc1$)TBQQ(!&Xs$LvEJO
zGp6+T??-zEO~afrq6a5PJwJCI1&{)a4tbBtWhBnwqxy!A`!;q@M>9yMIWSZ^PCF&K
zaBI3)#KyJoqr<bGy%eGf$!7(U&VPb7!zVl6XleV$uEmAkq27kx6{`ic=z=Dl&ALMM
zBJN3#1e<N<_qSSB*>DRH8sl9Q&Ox?)&Zotv5zA}B&x`3f;V)BaL^ye(6FRh6rFz4^
zcVmQEqN0mtycgpdi#Z7DwPvi4zrwenb4$8C>6k2{&8`~qZN?`{GzCc|T8p2>Z!#f=
zY^}%T+UIQk^V5P;@8XwyE&SKYgO#n3Phnmj#ZCLYeY87|>1Fny-myYRyVqs8Gxz0j
z>%_VH#-S`HgIJ?88g`TDB>?NUqVQ}fR~lnEmg7DF;X6v4Sk=PGy>mi}@BFyac|w0=
z=FvJ-0W#~G^eK_GyprziXBO8#3qs~jr@7}0+<rtJbL7ZY;EVjfn0o7=sNeUE7X(E?
zM3hbiq#G8bL8LpDSW2ZASh_<I0bLrTmhM;<B$f^pX^`3lVF~G6y2H<Te}8Ar%$Z>Z
z{$ZJ6cb@liU-xyruEv8)(T0BMM{?d4rVt+Zwh2#ocKD(j<;+bDy`_X0g!eY~Hbv@I
zPcM0sk3eG|(n(S4<W6bkVM;<BBqOUDxCLQgddIjqpd0~I0ZED3>%87+XC>|SSu>^E
z)Ct;%^tYBG)G)DRIL;&As?*}{ge={Yx81j1`zyz~dk`4M%7Xq{>6P=i2!p=Ob!Y{A
zU>z<66Rw%+i5r|bZf>pGA0D8PGEZAP0;qrRw8-REQK!bq(7pFJnUeVQz_i9w!y(I*
zQO8-$t=Vp>Va=N~(O+<@>CO^mYxq=6;j|Q0LdiS{{gF-Om8p?Sl>WTZ9-nIL3g2cw
zLJs}#+ud{IP+SXCC;yb1sie%u@lv(WCA0&nQ$#kKaY!`gP@Z;8W!rS=4G-$2LEWFB
zC85GvVP8|KUQ1&?mp=5=KGH*RrAPJWmm?9CKPI1rQJoGp@)irAJq^Ez&ftx{E!*_t
z;IJ{A-y!_71|nIK>95_R(xgIucA*k8F!^ky+A$dQdwj`*q1<-nRkaf-!zb6i5w=E!
zPuK@fD<dmbVlnK9I#gpO$6<Ld^u3U~l79EG+LlRyZIb}RtBC7)Rf#l4+YghXjRJF|
ztc@5*r&bHzIRy>x*e$HR+9RoU@p2l@QGmpgG2Q&~G}AnieuDwYCj81J)-dR|q}i1=
zy;~6=Vxqp1`)sVNyryKY!KXRl#)ZWAAcS@({;vJ*IDYAvgpr381YLY|G84il?(q0U
z(mByLNBR-R`4GNZw3XB|s}`MKr01a)-;Km&z+Q5|m_Yig{b2ZDLA94l<{k;P{NRr5
zLQ%ZoOs+*mh@?gNa{P>k=XGzN6tL`m|I^HEZ^`1jKW9i2yPQjT{p3wW@gs?}>OK`C
zf=|8oag(v*y6Jp|_TZk8A3sE3?PR4P3J3U+=|9Wu%|Fsq?e1o<EJ9B8BEpy_d~Too
zTzm7$>QZZ+>CNgiWA!UvB)!2T?<VqNsPweOQnSjZ@y}+v9-=aKre!oOHMVPqg1+1@
zHDS9aZlXfFi<CB%d2`xd*eYa+etTMY*#x@BdRmS|js@rc_J8ZQ-k91Bh_TQ%TfRI-
zNQ&M9XM1)xnJ|}UwX%^dJ8Trx66#ae*dtYCvH5#$+6@SLoifj40=pc9Ifv7aH$kUm
zfx%av`qXaC&_DgrBvV;pZkCgojlShgQ3;5AKG{@=Y6Wwoak92P8@+{_VNRTmOfZ%Q
zw~*04>kwCuNHev-c|Cf7$N#U9u*S`U-8L342Q?7*1FeAnEqX|4?k)1Jd!m)6$f969
z|8lyfYoev)5sRiRW5e;45s!Zpn*3-?l({e|%Z=wU*vQ=9I1xPuEGBg3wNecB&ctch
z)ghTm%U`Q!@4ApXkY)uNTj-(WI-xpq4&ia-t0#C}N8F!Dj<<>r?a4J-<1lk$M0!A<
zGC+6d46_^~1~UQn=3t3gxL#YVKC#LNZDWDw4!yi;<ywzBPk_}$1w8uJMiZeKKKtPr
zpt8J2Y0hB;0r-XDFeE=IQlsGt^CXm46@DbSw_T)edIKHq;$21##M+{XwoOy^gTdxz
z`2J;-@I2UNKZqje<6vDa1gS9CLJV&C+7R$h2vYTye_m9Mkc!ak=Y)EBo|vQ}-?Tj4
zjU=aWd#g#HMLyq8EJc(xk#yv+e>!DI*H4Pmnqr`jZ<6>8@ponsbICa>%jk>XZlpUy
zVv8KIwZ4H;kg(@PTm~HU*VFW+b?T>LpuvNnpVAXNli4=83a$QydNX9e+O<$PP{f(p
z_Gf}4Oh=`8vMHD$K_2q~qJwdq&1}v2elm|~n6<)gGkqNYf4d3EEdXVN>P{1|VBmfX
z{6z47D+}Nb@jn)OXh8m*gtZ-Q=VZ_eoD-4kbkdF5ZD|Pg!9S^$Jj0LMr-V3n%7p-Y
zk9Zd4&hQVa+Ghl!E`$Gg!^T-?E-YXw^K<@^#Y=9>G6|b5OEvJkCdn^1YM>Ex+~8KD
zVMzINg?6$%?6E`3Aw)-2bkvi7nz7C?X}}g->`&Vv@TB&Pq?Z+^_X}}iV>d>$MRW9U
z-%@OP7;+wi76{gN8PQ$VD21UU%)6wIcM?9+di56nX3Qmx$EvjZ3KMGxp2%qtBGN%)
zA!(9AIHSyBGexI!N9t%vT(M&PzJ5^`$}pu*$(Wc^W&bxXUAt3y##}*1;Eq8-(3jA%
zK}!Gd>5mU&jcFyk)JFwHY0K|b&$%Sk@%DXZRW`4;splEmV@-2b_k6O+9(E*dZ?kt^
z*^}WhNFTbb#D<?UL)4t*=lRe8b-%@R?Q!u)xMP}Z2ma{#QNwPVEvc>1jarGQbGrX<
z;8`(SSK9Uk5j%awT=E|dxYzva#$(}^i7aAU#)5(|5^GzE4Dw#eG|GEdM!MfBs~-EF
zR$<d!g9AoX!kb<)dupx-RigywORMxeXy1~yb?`m=c6G}WHr6Xzd^q#|>V(i~zA72!
z3znbfZ9gZuJN98dyfo5OMo-gL*#5f6_c?W0*6&r1P+o<lk(a%d-U|h;mCGMk207LO
zSWbzagV-G%#wRk{)_WBgYUe(xk7i;FJ)5Y+Xj~^G2V07X=~=#x0OL=ec+7FI@n;IE
zQ=9aMV%jjMPT#<J>0Jjfthco$D(EZ1Kb9XN7Uf8mvTs>c8SDuOqs!Vw%*XW|@>lBe
zN$t?UBN|fVCqtii4zd#0(pLrJxk3V^OitunIwqEBA#%l;9O7@1Lwj9Prcc#<+D@Wu
zmN#A2n6)(fW_$BQwb$0_kL(8Y^|p*T>m7b{!$7kmWijQ%Nd(_d<k5)PFnqft33V})
zFH<uob8x{dD$}Nd?;F6es58G)_wnA}=J|y-J`{)^Oz-&&Z!Sp_X7$5Rl)}<fXSjuO
zK-rO9ETq|J$BH)cMvPrmNHAgkx?CE2%cyo-I{SCGdQ@wo=SQ5Pt5E(9n8!;k<F|PB
zntg5klAJ(E^4)mZ=w8z;>kc<qW1}-H8YVq0wZzQqa}sZa@u{BpIc6zwzGEp^jmZ7<
z*3rK9#aDUAfjo3`cL{8{T#x^Ie=Z^G1QlcH*2^21SqMiDHIC!)5RacM@*>8&if@zo
zRA=Ro@u!otcJ^!yQwq-WN#|YBXz-+aNqzS0P+~`%OO<xVYypOUYL7*g^?H8CY^X@8
zjI6E55cE_~Iu9D+*AsRIu&>`6I{o?K;Sm~)*rt7063u=&Rn6(&7-{lZ66*Msb`L&b
zrfeLhqw}F~Y!44!;P*ip(Lufx%+Ho*opa+$8-i7#LxKyKI<=$aPtjg!M>$ZRM8k%+
zZN>pO%tWomrlqp^p~XT^WqtHLY-=o{A=YobEWF9rHbFC;YGj4a)grRiWaZ{xs-|=;
zBlUF`N<T&>4fKO}GF?DN)nvzz4e@|lR|fp~pH0VzT4PD&ouA)u;p#`JmX8`RWy0jM
z4gQ;&>O+Vlv-MS`U~t3TqV<wwYJubyDxlS(htfpQP2BWI4{d$~ufE5t@~Npr)5uLF
zGj6cUQN~mInmeV86*9ImHYwY&=nv0{_OE@VB^(nGVZ5jjzml$$R{~A+e}ftC4dt0H
zI0j+X><)6!KL#ur(u5v{=BnKP9&V-W=Lqw|X5c44TPdcr1>;P{(?*ZTpVi?7dfc7U
z{v)uDz4M4fQvmOfb5<O3J*@<wPAsBtcN+Peto4*hGui}HCJwpgr9eo}@<5SGZm&M1
z@%dEatWIS&>9X=-$D2W1-~eX_R2<cE98hJ`Fix|4N|F}@zj9A!-V(t1ss!MW)l;$^
zK!`xvP(0Ti;v%^HL8A#J%+}mru4^KXc39F9RMAh?7UT(ssDiAmO+oyzyj5{W%b6m4
zdU^lX7%Vnm%;DBB$OqJ=6pu8mvHF#yfOIu@@46r$IxBplk~btEWg}zrxf(K+;>;{A
zc+?I#`y`oJRK_qQ&j9?OjYu6*w4Q!)H32JUJ7a7u5~b8#nJR=if$ps%hRqz(eT5ro
zsqQ;2^SqCya-+&BryHP~WJH*#!j`iXfnL2L0AyLWSN?B*KL0*$eb?FZIM~qLT~kk3
zpV&}n2oaZ!mZ}5135=%Vgk%`q(qcbg{jW$p69EQ3NA`RMV!YL!ii4=5YADwgrslk(
zlAg+K<n@Q@kw#>HZ)wVozT|LAF*cmq{<5PJ4KX{zW=cvPq@0v9&OKGPV+^yz1`5vy
z0{Ox!^&c%ZsiE6#f{OrGB!0~zlHa4<WIEzTjI*usx&1HpiU(IFN=6(JL?1m7XF$%Y
zlac1^o5({g=0ANwzm<>Vp8g19v<P5RVxbBmJ((BMJ6fq^VV->LSq8!?v-EtYj%|{J
z*RdUaw|oSwVxZ%|HC=3g`u8|$Ja?cCtvJ4ihve#T&GeP%WU-1Q4mH)5?w+Ak8|Ehp
zUA#KKiW}DEH3S)jbd^{-Nlb=RCa$UIG%WI>0)4Wao|?#Pa~F0PefFiws&M-M84do=
zbin*GPrD43|7yY!C!W<uH}Lx<(bTtg+Z7$I@3Z8;QAitQdly(qt_B@pUun=7`VUK1
zR!?I@qN?-;Ixw$?Vpi-2>w_hC{0MY`G2ipdS7vpzblm1{*m&N7A-T6cw)&)opuZ8}
zHy=|@nP$5!>aslgIH1k)&@X>BswfnMU;A;)63Fo)>rjqoi@B$&QU3eI^mR}`OwVk`
zZBw-S>O@0b4;y88m(!_KO&Un(^W^rl%KB5O0S7-XvlL5oX<nd5rfc0_zaOJPm}?G$
z*^1}>=gixdGY0Uv3eQm41=sekyE?Cj!+zI@?J|Gk-BCVL|4d-Bhqj+1M2q^fgjvs-
zIOJ$f&WBU#waYldM-YdLYwU;?PmbW)BSiimXCA_*>VCU6>9`JlZx}(ZMC2CSVKJko
z?9f;i|3@DD@1Vp(e5LaU$M~h<ZEK(@!Cs$cf4pcl**Aid5uSE?SG6OApjarZC}Guc
zDA?FN#QTUh!P{)e|LAZ0`;LXkW`1EtK9!&ApQaNy$40;J^48>}4a~zoYS6+0%V2VP
zL<@=0fw(&L)zjj*n9_-WQpb;>JpOkr;lw$8tf{q%?@XBwuO=i%jvWv74OPe_<;v-)
zWKz9feS@lfjr>kmx;G!!?toz6FqbF|o@~Sz+*A^IpH#3V-s`@@bN{1yj|7N(3Z-7;
zu!pt6)F>%JerZPrV{{_j&!FaF1{%66>^zNy_5f=mlfDi9xoJf^nEIDekdeh~DNg%n
zw7i1w>+NFOrgIxq-D3R*J=GJ%ZKw;qY-W+|kB9efc<OkH^)fL{N?RJ+(U{wN2h0Z;
z+2<vL;b!gk)$#p3-mFh#Sp}<py4Z6nO*G1r44mafOKUX~ekAP6uin!#;Lnk`RS9vk
zTQSHBjD{Ld$}TER+9zm7@pRK9Ov76wvhPF@yh)g9#mx%@-Oi38rq_6yau-q~PX)D<
zW7=yk2i*QJN9+Brcc>@79wyzcLG{hx(l(B(Og~+|z5fHWsB5Ywo(jJTEqX}((ss6k
zqFdT}{4p=#o{m?mRde!RYNi^<3I5*VxklswFhtcN<!zw`o~rn`a<eg(Tr76W?=Lq=
zhyB)>?Yoa<Y>fiCQ~D&D8!ZjVs-H}j(&e_XRd+3@6z|QMe3kf*Kq8>}bB4G1Q|0^M
z@5r&zg~H5-!N7SF^s>j7{oWqCur+4JmW)+1r$At~=#Ae}GwK~BHZ!KQFXX;8bC6_E
zed(eGV7PN=UP>>~DgHU+Gu*MGr9WYKM7}~MyooK5!5K-Pbw+-EAJCSbq=&<sGgz(6
z4y+FKO*K(f#iRmV%xbh^c7Tg6!SR`GifMaSI!H5^D-+v<XP-#1itb$1d9;U#d9V+W
z3G5LYs6<B&11qLyV5w)+RYQ#hPLd{m;w^b}FL4XGmGjL=QDt<Uw6OqQ-`Kta2C6V!
zs~jKDJLVQQ6jGssrP}kxdJ}I$husu&X5PFMKYfF}^NpG<2d(_YIiUEM0;P&->)@Q+
z=55I|sdckszAThAeoJIrSppqMVoT(uhn-=()xqC*zs8SD6b$|<r;Qwd)-uvms+TsP
z#z0CYuM=^5uG#NU-L9<pCE;>N)micmqzcQrK&nYks%IN8^=t9Bu+{XIcLkV~e1#5$
z!^f4xcyxL!b%ZpGFC%?CdmFa~t`P94$F&E%r5ynRcuk7-70~8Hne8tWxPJpg*4T=9
z7#pz5ThgQDQKGtcYU&9?ZwXL{Bd`@h@&<@Vi=m3r%c2d$yCczAoigf&LmghgEy1UX
zIC{3=Sl-g1;$bQWv74NW#Kx*CL;A2tqegM-uP;ecpYB=IR@Q;cQ`UCOk0fRZkTMMa
z^NyFgVKsovw&@F-3gTq2$z`Q81zK@YfRReW{a^k1EpPkfF-uuw)Gp?!Fn{0FC!+ra
zdWtq+?{!Zszudzw2A`Df70F{}U&qaME@$e_F`^_Q48HcR=Pn9}eq?R?mQ^9Luj7E5
z&1zl|r+L{-O|$Lot?i!n;VHsRq0#v(!N>^S0Eh$|06sq0Z4ic^Opw>8=d~p%@Bo8G
zd!B#<x8V+Q<%D#5m38lteL-2A1@2jS%lU;Q<#yN26}iAH%m*~|a+fPaP`^`=@z$@)
zl;3BC<d{*dasW$X8&Cbg9#tK)B)$26tPqqK%+fgAVjqx@0s;T}a7?SV&>f^kLQtA0
z5Z&efhb#ZyygBpiq^3lc#CElZm4v7yz_fcv<i$!0QGDu2*{8Q;!~xVXwvhTUpXQ&v
zo!F`_0Xy8cc<K6$;3<I|&b$m>fuHqZ*nK^VxLY2l?~?{yCEB%7c(5DJU$cdobe<q)
zi2A8=E9y`OXE^;vY&8m*p*i&x=s84Zsb=@wOK?+ZtQnwI98azIX1(ziFzl6kMoMRC
zwSE9VU;jrvx$}SDc_axj%O7Z>E~#pl)Hqg3Ogk|GCpdFK!53zQvG$pt3Br62_VHrZ
z!Gee;7Ul!C%$n>#z4DZ!zCAg;B^}ORYl8^^w}lGOMY5Z_6MC~7w5{!M6J~ju%lbdL
z0o0dPhizvmhogO{H~$eh&Golh^qVkC(6E6>wx()*QECQ<swmEqWyZ_uy;aw?k>?J5
zjX|Wp+^mPR;3{7x=Bd`~{??X-eb$hhF%cpe@)R8i3PF?g{uQQ}K$YgaExjwaBaHhX
zV=7r0aI~O77C7RSzc<@Qq>y848S={Ivo$DE;!)f}ZHo4!QHL~!viHgTnJ-(BRkx~A
z&+iTg*kWBG5_+eL9iF3sw`^_Yu)_2g1l#SP(H-jS?a!ohJu1Cqxi1pzQx)?$(+}_W
zvhP0vYs3{AlN5#U*bO^ir-o=<vi-ZWsHqYmv7bpmw2+AnSiJ@Z>`e~;kTG3Ig^d^g
zrl%=>u<$tPW6*hd6X)E21hnVS_Vi(VpJw#%ioaC;gVMHZMMYz7>s0eB7e`qmY{<fL
zQL+MxK%@Qr>83X8Sy<|06Wb=VoCVdxoUlC#$k)crgYQ5?fgq&p2~B*26dBu|B5lN?
z_++ERii~sM&H|!nKWa@+SxK^%?=BcP>{HEl|1d3AM!{}wdF@7ZPZ%id$zN5^|B9Dk
z0L#?(Y2G`gyGxp|G(sThJV$cgib<+w|1D`-AdVkGdk(OCg|EHA^(sIJj14W`vPMcM
zAFbyx^$n4fzK18P=m`N5yTEa<kzHSjo&;@vsT<jOOTN9ojP&tIxh}gUJkOcyUXdG$
znLtS&22v;hSVV?7fmixOk_-sYWs?u}qsiFG@ffv++eHJ#XpLh>rsX0Nvc3asa-pp4
zVfoaY=ql4B5k7Aw%_U|u>TF+VF0N7N8oi5rs+sw6B*1+)#eY43)WYE%PEZ8hI^EgL
z@~HdAE#GQw(pu$c;}6Y~cgD@=#q;T-D!yyOm>)98Dum>G?7R9J0gL^|V%D~>GdTe%
zUEY6>8`o-&DVC~z!JT7Ssy{BLZK_nM%?WlEJM2Oee7Emf2AkTw7<vjfSWhY{aF%f}
zq-EXruLi8>k#9*mdfv{=)}+N93ofLNoBtd-_N$Wpov~FUEui52wuuQb(-iG>J>`(a
zVrv7PM_bx1AhZo#yyk=>El*Jv!ogS>@X&`KhH3OCzl|-cZxAN0P_oCRdxcfKQF#y9
zP%8HwdVG30nYzJRe&u_pPdS9t1ynL^#z@xlC>IO*<39x@xHm9%#5fLz)?sc2|2%-|
zR|cEZ#>LLH<P3)@YnewQgjwpBTGO!c6huuyTM&ix+|-B)i#+t5n9~!~1B1@qk}9P5
z^Y9&9hi+|DRU6m|Y#wY<^^hZ0IvXV>U*vZ)(<78>k_Lk6@n0Fzs{q4r=IT^*%5UzY
zzLGnb-j+{**ja%;L-jVDu;v)2O^+EG@M!kBzS-;10P^(*O94!n*UJS(#(TU=9nfL-
zed9D$2uGGvKZwR*{5+1ix7fL{de77Rn@yA}4Yi+0v)gt<6W=?F;9psLOxVa2Up049
z<#S`9i8A#aT@yXg7qN(MjJ|3N@LUj5{r5Sqc4QWNiMVpxg^Q+4=)CjcABuejDXEwZ
zgRv4Ou7sdTbVK1wwV^)EiI2jIet9pN#8P#{@3XO_qV9GLA)*-HmsmJb_CdI_f1C1+
zoPS{6vC&a#l3lKQV<h$*zi;&@iw+kH?iE}I+kGYfoDMai{UTu1s4c%YXR<AfgS#G}
zehDzfM%BC$96sczj%V&&y==b1QZLm)&Q0+49&eNUAGMsFR*G4a&#vs{p3b+n46GU?
z!7*Qi`63MTCJM-#P)3y2$G5h&9Y%A|o+*tv3TlH40g>1&ez9Dh3U2!!rTsu6C_9;Y
zONsYZz05smG^g^{5%|IENBD$&FqYozY`IDY09l);Qgf>WazMY-Va}lr+sYgXYXGCh
zG$|bDPmcuH`hMq?bs_UkkLNKzajNfQP5|~v#xO<D1Fhz`=LQJJ3~nGw8}FQbrT-%A
zBcPuC-ZgFk^sX+Uj(ZX2O-4Wnz=^t=BcmKRC#lXZ+Zf`*zqBTGLV4Pvr#{eLcxun9
ziO&+iGZqDa4@~~OCJ>Rg(Ox+>VY~6ewk&GN5aTBseo9n;=Z+$SCk_~2O$bdvWFbSb
zMtAP5BQ;F)23snTdk^vprRV}*Ul~6UwoVm~Ighplujb#kJNd_e;^f7|7%x_fm=hAZ
zA@}t~qmZJM;ML2{eX5M#*Gs=-8neBErpM)}^b_+z$BW%jJZp{}!RJxMI0jSxIG&=W
zxGw*x2odX3Ybkk4BMqU6+9aTVO#B)<6oNK^D?sX5nR!WUHL;DS$l9(Bdo{t|Xjy-j
zQ`2v8pf{M7O3yjcRC{vcJP*z{3<Y-cy5y}X00>O!aU&yIg$;|a<%TM^i~-=yO{O1!
zlzazh37y!>zYA#0z<m00TlX#jlmE+V$G*kK;_subh+7+*ywcLf4~AD+LP<JQ4#}2P
zV-8l-()Fi=*c)Rm$b8-NYX)m(k!SVIhrMV|=AMbpbfR^f&gUi1?S)f_fVpTW&4%sa
z@mckB7_T+dB#j}q)im1Q70&(7aA!{?q_#RlY4Rk%&!Q_&Oh_nK39yYV?RB!bFqB-4
zuJco}+v^LFFpKhKG<$MPXo-l5G0(povFADC6>ePbMXn6tbvz$!HshL4d%Q<qV3!j_
zEw4W6@_IJP?&7r=R`w|d>m)VU5L+&5-;$bY&wH$+brOb1^WFBKc1=%w`mZK$R^v!3
z{_1N`8l!RH*~W%e&#v<+dJB@@%|kg~w2_i>#s4iMPIsU9yjN^prDHlrwa3gWBpRM>
z*K9t0*Hb%QXOWLdz{C`%!A>E>w0D+IhD#ahI}GKqAZ^|O<M+(_gf1vesNefHAIZs#
z5q#Udu5Os}!b+Gb_#c+jxw-1A^k_U<G}0Me(uCV?X~fMt5f6Z|?>`$ACAzVbz-yXr
zXVYQ3ZYxtC?aUL_diT7yG%pg?WMduSaH)^@xd%N$8T9qjc@u#CyNC0*?#$UX!>una
zAcpB%nT+5xdWo<reJ8n+t#~=8oqGfCy?3N1l?(d6WhR&3iN%ZTC=M`C?f`$oCdFB<
z2b9soUWU@*aQ;x6opN4I$<W`X{wDOpmY(sY<<p#wP+qRsK1{?(;XS|a&vxCHOFvFO
zE|3~bfk)N_g9pB<HNCan(dUEV6HqdQ`Zp*<gwxUXZqmPf)*kRfd;MTctX<4ku)+7r
zi&041y4xmi?xf=9+N0$(kc{71>2Vv?#UFj5%ossuwv<5_VGq+~D9@Y`(e~nu#0mM6
zYnQd6Y#Hto&?yHgSNA<)#HHHV3#aPp(;-uOKNVNM<<6=J{7p=Hc>U4s?ByJ^Vdk_Y
z_;!#V)M7VO%QaiENVN7eKDB)eW@$^Ej^wQ(k3U^%w#ps0jwTEk@eI*_B&Vn8J`>V*
zpzfM5y#+aI{OFPoa5YV88V>4IR}=)?UJS;vBU?&}a9;$jfN3zO$;~Z(?p^s7_niY_
zIlj`mXs^|GMFk<h5SFgAbf`by-xa87mN`FQ76H=SQLf94A6si<l3?R%gQAk8Q%#18
zySIEIZ)azyE1=kE=10!{;wy$RDI@05Yatk5Ga>yK+kAL(HEk&6K87X>Hm+<-e_8$(
znN&Ul>7gGec#@rj@Zzqp<%1-7CIi}T{xx?P-9vCAV?_WAZNlJ2<xo%A<QfdttM7<e
z)73wP%-AnZA&2OaBv#<Wy}e$dk5n{^o1i6+t{rC=y&8<DE>*k~mPcgVG#ZS~Cb-}I
zWI#++&R5!BLLnBUJ@;x&@aEIfZ*i&A9nC@dRHA>E6XP`sc>k~=bzl&A!s=cfL8<gI
zkm37>BW>rlY<WT=&@#cQRoOyRBD%!!xlKDXIEG%pk^@<rV5@%#&3_DVIZ@1lP%UL8
zu^l7Dp0IO}>cZc9-3|?GKCsh@`)FFZ?OlVtE#v3*kT%%D$ym%cHKV}b&V6d?_J}cT
z|KL=Th->6qoT)$&-xP9>8p?I2a{77t+c&)Q9D;K`rQ+SS<q@z==S<`&h+q9O6j|8#
z%cU*9+G~`u+WZ5GEcl1#BD}%&WlIp_x5j^|@5&({>p8xF!Oi2<7W1Ky*Yzza-TsoT
zLG_E~s*`GTfLAwz9%t=#u(%u-{Y!vF2I-t;dB#k5B7eHGz=kOpKM6)=eF3w_i}rc;
zs`4%A&FK9i<7`^ydF~;=iE#+<5Uy+k7Ia%yJ|!jb3|}g^=B^g(=i08P>{r~jOy0B~
ze|_(R)tZ{OlFr;5ysm2oX{v(We)(T=xFAU*+YzMdAYfALo=&X^R5U_DuCNcXe9dz;
zN0aq}6kqT17AI2#rD<PbO`+=1VMR=nrsNAr;ZU=pho0)lvlq>mZwY)s;FgMNui+wD
z(g@!>LA#(%sN-KAT=R&SM}BQpPvE{m8e0xKsf)KF>a8eG_5_~FAw0HpG6P>Kf>V6Z
zJ}EICby~V7{$zhlZ@EX(IiW<V6F@jR%q#Mp(hbvgNF=LJ!SV|Qu#7xV?~sP9jyMA0
z>e=utH|-;epa|H%>`*7dsvmYe%4~4pFz8TW2gcjFkv>gZb(7ph-t~gy55RxepW6If
zZJK1usNQb5&^?HhR|65HnXB7eZg+)gdc!w>s2Vn}H?e9NJp>!|C{pK8l$STXzuNBR
zhN5`Sm;^XUibCXA`PJlPOxrI!qtkMb8~f3&t3v0<O+8XvLn1&TSW807#g{&w5L9D@
zG}huv;)VZL!hqP2z<?Qp$GaRMUrNZp2sccO=GtOI@MKYoJUVMbzaGD@{-mXr`Nt{J
z({ithd*d9bKho4eOWDTMfd6-z!}Rae9(dHU{nims<d`=^-pfxZ1jr&|Qh>+n|AtnQ
z|4}%(3=+K+$=ZlGTiieAQ8u_oEzq}5gR+|ORrqP(x%cf96nU7n;vd~fb^&v<fMn8=
zQ2u)9DF4B9l>)DDZMKugB&+1#x@RWN;XY^1ms6mo9gdmAnFv^aUc9a7av3pNKK}x{
z8`CcB`kw#(up;;>XnH8<Q`EMR_(W!@oOy+yP_Blk`AjI?I`R@~$_0uoU4<46iWI-_
zdq}9$yVuaEKc9^`oW;<*Zd_}$?Sj-Yw^n}a`4)`%whl0L!RyVJ>Z~8|ifkIyulgdF
zkaBxr=R;Fj;xk+N_4kTt7$#u2ecgfbuE>Nny*Y{Xj9>rj_)3cF3Pz~7%~uSzrHkLc
zvsVn3`mvr}MH1wmA?bK8v9~VLYC8})HgvQnd&`@`Z9fWi^`^7ygT8^*&fZyB6OV06
zzpae3dtsc-WLRysAv<QzA-EKKQd%*uIx!TrjKIf#C06+(N6y<h3k$^^6%Dzy^yg3f
zY7d%BeeL<z3(El9-px8T#gf?HTF@0r65fg8C1*dHVS?Lo&lY>!z3ZygwlTj@n(zcR
zN?&;xe5{zV{&P}MyQ($`T_;MS91r#Je)ysEXCG!#wttFl#$fmt^GI9l*>WCZq{&=<
zE`}h6F0kdLO;9T8C;w~p`wBmKpU3_BCL3gg5J?;i@)@~L#4JbQJpm{-7wX8wn#>0R
zn9%Xr!sbY7B@-@oS)(WZyxgbj`Dv177$!GriG5owF|pU`%)F{bR#h$Kj{3+XLzQEV
zc+8rOs7Noc1mXN3X69-gGzNz{7f|1+7b+nzCr|BB)6;}a#$tu#qi$n-DP0bq%^<z~
z$wE4tYZ}oO&To<&U(2L*Nh>%4a**;;7rc|XC3TkL*pJ5g@?}q0;fNzMwwdSpyCOTw
z>C(#E5x%0$y4kL_HK{Ddh`{@!g9}l0z9sh2B=E6hdhx9F(6*v|gB?oLs=49)8upov
zQBTdm-WPGrvA<SBbznMWRWWTa6^Wn2I!{#p^ol{I+Es0R|C;#95WCqvy1#}_dHC{_
zdUC7oPCje8d;=Mw_<LdMcfH>$D~KiJoK?J^@-RxkB{rN<i~-t?#Ex?&p;fN5BpxUi
zf4**twAPhJ<?COEc-zuJVZdrN7puyeb0BSkZjG`uX%zF~hk!2*xeW9n-Hp}CEVEm&
zu4t+3)IA+i{Ei9jnxz3%aHV~Me?DqLgyG8`uS%Rrk}cUZ+yQ_Zi0U-#i6D6Va+*tp
zXPw(rG@2(4P0~H7iet6gV;KZ8Hb_u<Y*E+kMx^-AZj^rOT%yEwQW(@V0GDlq)tp#k
z?)yjqbsy299WipF7VslW<fTE1;9zBabSFsQ(T~^s*}P++eM;Ts=)zEKQ7YD7dFILI
zL!GmdK%=r`^+-cYD65IIW4mZ?dpKtuW3a(W*Z!6|zKFS_bUGO{H*zbF?%V1F#bE)^
zv!Gn0x8!u)cDmubHd#NudA;&BTj=2FX^pgR$%s$r3;q`%9aW3mJVzvCq0~g5qKmqM
zEc{0CmOPL(lSAZmbOk<9I|b|7K!%B@@yLy75lpPdHpm$`@J*q9h@93)ME`2?I9~)#
zx_~KPdbXysgKh6nUF4sWepur<GR6ma16~ukd7V!GN5sS}B+ygH?E4iq^c`76_BZM_
zmi#8dR-=D^k{;HFUreWJt&p`S#JEqC&s}_6_wCkP3@_TD<dq6pJ(V*#{*~s(E;`38
zAAu<Cl=zb8${u(|bmP*moxq<JllN;KeG+)qM0Bq%;ERupiIpxdemlswvjsHy!Gd7=
zZILFk634N%P3hopzqi%DFP@F!>(x))vIF1-@CM!4@lH!@g`kPB7SkJtdj1U)y>`Cn
z3-qDr=y!RBj;_|6cSkD&bE+;AF=LmRM(qiYZ<{;iDIQ$BvZ2_jkEnp^veB{gb{zh8
z$-RyKM?1H}!rF#~X(_ufvZ)$-{Pr3S%U;>Bd2SPsXLxaYm<hWhH#_n&9xL+@aSOc=
z`y0^793nU2<CK2ye)@3A^{yJnKTwAZN>}=eris-fuMLNsijlJPJ5DWRbFq#Jns9ui
z{zYB|4tJwDDL=|~d1OJ|tnc+|BcZp<IizPE8_WthS>8)pM<bsnRW@XnwJ}6#V`+={
z`f<>&agAj;aRxbYgiqQpPb&@YB55Cnb!irHq;gxhzS@PhNkC2C>NgFk!{%R?G`VNV
zbvU>6&W%q@6a7a}qa?y826M&8G>@>0*uqsV)0yQP{w#Cur&G=O?`!EC-(`!E<O7aw
z4D8ZVCrnyjMAA4=ysd)!g-v<y+<dtMxvq&rc{#UEZ%U(6%DRHkhE~&B4*>|UFhv8$
z6r#^*gaDKQQcLSQDYIQjD&`vvdeHSY?$}5Y@koVTa1p?M!=REYMWo4@&sgWF2pJ`y
zv=vaU>z4fctiawsxfrGj<)PCjRse|IUz<FQB!~1R*76G<Fs_f7CRWca)f%-Ts(zb$
z0_p5KTxT0w;MJ8yAbmIUAA#c*c*Z<z>t@UHo-j_IntzGlGC?y<XCW1P!S!>$!b(lN
zB9)of{u{GnHwAffykxd$_(aG4>n2Loe2=?c3v}Z&vml~GzLwf(*H975WT&l9fd-+v
z_dNA=xS4rgv|M`dewuV>*@LBfX>gX`%ton}BXrO&Aiwb3keFe4G`Lv6Ahc*91JYaU
z4}$TW`&UtCxd-JkCU<N;GAfE+1zEsD=XdUq6`1DdMu<VBpib*7r|C@5Q&qkHPG9Gg
zW$<9hb*#S(|9}2*W>uvT4_5jxP=Np6nkbH`5aobpuW-*3SjFQuNl?xqIu}VLH)03z
z1vT6C#2+Wb1Q!V3$UF)JU5s2q!Z596eWUa;m&q1RjWsbAp)*9cMWc{l=+BW&_#&DM
z4?^Aq9fq9O(=<%Jy`4uFc7jPNSkQm%wR}T@-=*O1Vpb9@xO~MtdtGvxxcp$Lso~!+
zm2EBm7<9_$w}3L6p)G`B{~6#awr3^FhDWKIT}2pgwD8Wca8q(L>HGl*IG(4aVriw{
z=GC24QuZCuoQzXBeL(Z2A0}2iBqN`PFaBix6%z3no6J^jeX@3GO2vzOPb1&Ze4cOj
zPT1M})%9dL9H@*9OC=Bqv+DtUGc6N8Wps*W?J@o!P%7h>i}qAqTsATX5{KtRBbH6g
z6j&8uyJclA2)(iqdxb~9%fDkZ?0|hF@Mz1wsf6vDU7xbL?TS$oi1()57ZPbl+We#9
zojashgWqK>+xHx=(c-s^psTwKym30}L$Ps4vn~MrCx&WtW=F*i8PtW_#YR&!;3sbK
z=KIE9I-<W{@bxO<4J*_1L!Au2_0xxw>T(Ox<#dRJD;0>U?f~}G-#`}xJ59nG$qnzH
z%zwkSvF^ls{L&k<Su8{!=I-s_wRQf~Jgg=uFSAVbKz{rr$kTX;&mvWo+jfSlw<v@q
z(ehA}=_Yk~ESw~V?&@bB1nJYZq%>D%`%=q3bu}bmjCvHxD<v$sYDA-Sqxg?`ivhAE
zo!C<8TTUfzE4Td>&c1V{&#}kh%Ii^3R5^Bog(_rvDh>rPpR`(6*m^&8OQwAyL{Cow
zy22g&JK4!48LzRLGC|DHrOKl(smG&UvX_LVhV}gu`;Xuga*4V4iDqfA&>R3pENs8A
zw1^wud}FsFk-071nOo1tGYaY>P9Cr~CwzbdZPCID1(i5kuJO!*ZZjHO^5_+_tCvl-
z?25O$>_jj06f`&SyFEW^Ud``(ZPE~zd)?cUMza}gYEw3+#!$ouBOil*Q4!O1VH-B-
zQJwsOr198IA@(%Nf7+KQw#OR)1|I`r20PVfFG1W-j_tcTt9#c)R6~9<Qw@cet)$!G
zqkN*p9u01()Mame>b2=oMxB&Wv1ykLqOrtPo`Z%lW7x=#E1l4}_mY}L?|v$lmCCT7
zD_~uMo;}WkOam!DanQbnYcxdaapU*^a<BmU{ma*Zsgwa~_K*gBk*`-NgC*d3MN2q5
zZ|KU%+%FbeHm58Bw-wyOT=_5Q$i%v8(tGfR7A7<Cy}?FW%~#H3QlzPAW}5m#Q-EKX
zl0!?>nlfaZptQ_-4C0@fn%Fgb8sNKYFqN5={?1LyBtp4X@=3EI-^M6YK%E{H^vT<V
znaNu-CLq8{CC|qE?!x)(C@P}GsK@i{_(>&L@lNwL=9>FVIKz}0|Dwg$mW(E#tO{^=
ziKK85m35f2azQrg<VZMkqIB+y9PqYto0dR(D|aHmH;Fm#HCMA;f0r9i{7Ni`p{<=7
ztMBSPCEhW_n^@*TTVNogbfdfG5pKK^Q(wvVUw;$yCk!`KW{5}Uv9};=XsWaUv^A)1
zo?MH$a}O)-R}i*|`c&LX%?_$9`8haKsRuL7N18~!Wh9<_Jy=*fvlYN%?NV0TdX%af
zWJ=~(&YDiY+~6>^f{rW>B(03ij<E9F48Wpgbc}i%TOVZ~kMj0@Zi{)scU*K<aj_-T
z<4*PdjdQxL&qb&CL$&=dbnsczZAnR!=~zSCw~3eCfsQ31`v?Br(xY**N_3HSEY9=d
zf37=cYIL)G+T??&b}wyP&ff2`9=w|%2j3`rKmKZg*9x8gCVxKF*?)mN%zAxcv(!?(
zFIX6t>Q`CsrLlQ8WWe^2@AYl&&HWgGpP?nDxnZuid>O!l{yk+FHdn6c&=GE;9>QqP
z6KuiNPs=a;L?kv0)G-m#mt&wQvIaj-D17gz@L-!oGIxJj^2-w2im}oeqv>z9PbVBK
zr(3zA2T9?Oy<XW~`5P_VmQ`-L?V~J5>IUxGo+rdAN*@UD37Rbp(tnQ>a-8YK_q?wa
z{Q5qrE2Ojh!$QpIL;30sr=xmPC1?NIoH#QkDuj3aY&I{R|1SHN^><s5puKmRW{)#&
zp+a}zRD2iubr(AHEVJ=;ajGIPJnT%zEK6XaVnBVwqT*q!GaXH#Zf5cEUG=H9xPL-W
zhn49L9ycTaQX|HG!W}z`Pq{5%>?UP@C{{}sD)6r7t&_%Oc`bh|lt{|@U&4t>B1J)h
zhb&w*EM!bgj55P-xyS*hrQ?o>P{cP~IJb3d&i6W(yc#3+ax6QT%z#<ycl!bz!~`RS
z8H*sPEH};tF-6oniM3jF)Q%>QZb=xXdiWN2jd?@4*WGRes(o&%i=xZf!wd!M9~f_5
zR_+^#1MpM88&1zofdVSbemmp;jqZR4@f#OVi*LQ^-F{*trW}v3{BX&q`dte>iv1Sk
zvjskvANtoz(o+YIc#y1!o)7E_J`9RF#hXXa^h4DayM62Hzl-E*_U;Vfmj2xeq!|So
z#mdd6KsG+6l;5*vXrexA(lEazM%wK&T>fK?23J}$wPf1!7*LiBVku=>xsgwaBk>9&
zkXGy<pgrfpZh+$4{_+Ky;#`6=tih=0^cLbyQ50OW8DscLBDrFmt~uaLd0aLCGpTF{
zmY$>!qKBWOxP?eMyvIlOjg)<xs-2Tu<Q~;>o+ByeLS0Cwj{S`wl!J5*A1SmeWPccp
zTW9v`0Ms5~eWJ=-6cXYDaXmkuEI}8OT5H%ne^tKcZ8YM`+%CXh^#o)!JRC%N({$VM
z-i`AxwPzz-mdmGGE$*ZBg6jK~6|B;z-e7nTOX#)xzv(6gx*G%e$ik7PZVAz^aZ8EI
zf8(n|RYE>hwAILZCO}d)5O3(-*fM_DLb!x<c_Ri5*B!<aco~jtNvXY@vmVVECote`
z4Ad8SH$ZpEf3MAUub&jV^t`^*TZYnCKUz%xceQVfcr^sv81Ha2HwU&YBkdE6v{DqU
zFo_TPs+}c?Ci6;odljOAZfug<yb?}vP$7695z59}@5Y(hY4g71dE7g5PaS=eY+7Hs
z^HN1*-ZI&Rz84Dao*_rhl}%qqSjm&@H$rkAO8)FJoeIf-9{>@@7O0(fX>uF2R**k*
z>J|((oOOIzhL;fmr|AjQYZs}4q<U2}>%>&^eW<NCx*gG4WOps~iT=4mU6WGhajYn<
z`1dGL4dUVNALV$GpV`0sd#tBv7jPL``PxW_-~xmHYquhPsyOCYa9-N|Jl2u!(sM*`
zDk+pE{Jt6MUy|Du!!Oyrm5ghw-TL_X^NN9UPwf$0fvbqwU>9HfgVX#^R&8Sa`#U;u
zAK8#0-zUM=svhktC3m^zTHK!lmGOQO_$`Cso}BmAJZ`HZt@r4NP-b!S$IrtQ6}q<)
z4)fCn6U&+Q=!;YQt%{pUbO(#*bQ{hi!CFPDaAN(Gy@Y)R_E9Dl=ignvb&`)T7ZyHm
zY}c(D2s6Xl=s)Zm`;&hU@<D~_-l=<(9XlndvicWO|9Z%vcD6_sE?6|G8O1qo${^Mx
zMe@gpxyrHo$hV?k<Y~1?yqE?V(JOk6Oxu<K>eH^>r_23SAy)4TMg&LEJ)JMJE<W$T
zFBR1?Zg<B15d@T9C@lL$hppV_GJf91^RjAN4Ix1LDx_Zf)*La`G}H@)&)%$3en;)F
zGoMgesw%5IyQ?>qAMD8}#jc=vx%K+0d{m2$p*Mo2%&Fk}LmBwGtt5hYfT|6_(#I^1
z-aI>ST=PuQ--a+)zwTf0fSpXdTK5<Bnb%?V7<au1K=pIRgMUgaZ7sA&Rl$7d>%Wb-
z?_j2;GRCY_UwqVlIvf8wA~T1|43<5W+U`)h+rDYR=pV|bJLNW?5$epwxa-tM*9(19
z8EkCu4fg@?@D%~=*tXuen@P69gLJltKWO`8*>H1zo8~CFo2L+nbXwB@^>}9#kNKuq
z6qY@U9ppVE-hbHkvSN!CV4hSm)Hh}}hTpHdx9dE9>s(u2rbbsPvVyP!n93(*ya~(2
zq7>XKAP$!X`<9BvxaP7z8$mt`vt;fS8%b(sV*xc5*}tAsWjO8M?EXeg28~Hx-?KfW
zC&=nNh0;p0KHPKZ?Gz3Hi@drr$$I8Y=e2W#R9I=klNE=^M~XLBzmxvd4I5b$K9|S-
zrIBZ2!2alP{uK9YF;$Ab0vS+L_TWAyXTNr0M*aMWJJ<Rt2s=C6Y30Qb@*$i1#wBRV
zRq#PyhX?zn#YF#>>seC>xb%6!c@@&)_lgshPQX7KYBm1qnUaUN;K>QFoXbNi(Z&Zz
zgSiAquWFOf9<{^bzBiyjta$E(@64hs=4Je=#v8KMK%8+V@|jm)len%=;5N^7wPPa=
z_ZPKOjp*Oj!J;w&cK;4pA<c;zxn(=rAKxqn81Dp8;ongOPptCcB={`$_ArTNRoY}s
z4n?ilf=I8C`W^V+cALX2&4Z3MbkE$|Xr-jQ4KRaix{+4><O|H3Nj7H=kIzLMwbaL~
zc)n|c*^98=73~R_b~@ecg2ls);~E-Zh_bMD$@Ps&;T#+KsMNO>VIw<(+h*a#vbbbg
zTjA4^cE-vyxGI*HMNH4%xH$_$!{iSs+nR;MlRM2j?uPOd2kfX`NG@@}pT(8e`Ca$i
z?)0oo^UIQyF$Sxc*H{VDBA?>n=Xx_WXEs=QF6j70O0R)-<J(WQ=KazUx7)K`jnT4C
z;4tZ-BT4x?A>;sOl6CoWyVml!EDKA#&LO1n{rl6inn?xX3%{YCttyRq_M}S0WVKI3
zJ|OY?mKzZva<)r35{z2vCNGP0vP{FRLfgOljJyhsWs=F#k^n9R-c!(l*6ey}RN$<L
z&z{qwSx=?8mb++Fyi&ryj*#2xSK@B4eXZ1wqYnn-*VG4-c3n34G$I7aLq@NipvvF<
z!srfN9S7|@EXdNW4`0e?Tmh>8)6G%ocB^V>S3tIU<sdLou_@0M@oQFF(qw<VVtXhS
zTtO!U!HWR#HLGHBLpjs@7=N}D8BdKQfCVc7PqMT*%|5XSvX}Y?{$755&2Z<OGJYJ9
z;EY2A1DSHzehQs9!Q|0dsS)B8W|myw?N@_5sLsxgk|?_!XXbA)!@!T3H!t%w=T-Ms
z#J|3O#3QNiBzfVfJOL|c0UUN`AZCqrj5f}Rklx>j>~V7ieIo|ekaI;aLuU-)0KzhS
zr-W5A<$=~{;HKYu8RZ;K$QX=q#X&WwjA27tisMunmH1<TA?wUg01G?cQjE&w?OL>a
zB#)Ny2G7?{6e_>w_YVU$^X%>g=M`spnz_>|t=RiVZp9(zWQJlQU;F1mka#QFkL^O7
zwV{mtRd1Pj&r%LXO);f$_Ki-Defd4Daex0KxC0~;e$`ffAVJyl_>p<LS{>wm6j+J_
zd!JVC{T}XPRpg#HDTy;CiZ$s=F7UYfmTTCj$DA(Z^~iT7laMo%m5fE&N#(J5)aG4p
zacGv5Ez3d!76@0sES-05WWjHtMktL!f5=zaJSgxresouIh~;Yx&*hy=4tcD<n+Qf<
zAa)>s-q<m!aYos%a?)Zi-%;vC=R)d}f%wxJxvq)2z2{`+$HM%x!)if)nwsZwTr%Kp
z9Xd|4b9hS!v|dJ9irgTayGH+bK=z0JfT%K=7W@4b@Z9hm0nb85EzmX6^*8^0;}a*7
zCIyNRu229Iaf^tY(^mRSu|)iy8dRQDn{*uzhPyYhZF4W+sj&@6xv$*%yLb2u;V9&0
zuC*n3bAThRV{Ub$;=DqzhO;#(bFN;DMsO}!&wCqM0h6;hsdD6Ecj}#uvvXVId(xnC
z-Rien9kxGFsilw)y8Bdzw+vD>NfY8%ZkIT4UJ;jSaJ!<;&k=XD0wPU?ot)xgvsb2)
zC0OV4m3fAKuIKcCyoO)?op{Xm#*xZ#wm4aV-wj3%C?ihX?-G&Y^}6n7?;#HBEAb7t
zm=9)Adl%*378+vOYI|f4T0pPknXU{&DlmpU$3}KG9r#ytn|~w^33Xige3vtuYgEcq
z3$B1yhT&iPv~yWQ^-??6^vC8G1)*e-%*{vxwWGhs+Rc_~O~_y)y0j_=S(ZGMnKdma
zkBqK(HjoAT-hbfL9kAP?U_ePCAE$9FMq~ppTaBwr{H+e%{_Oump`}gMxbhW7zn2aO
zng3L+lenig^`Vs_Gle`BldRG3;r<`#M<>|JaM*HWY+(D5Tsf!GGXUyT5c-=}K{T&H
zgGF!gmYXAJ?Q3>?GZpO-#?67GllR5XYBAht$vj%=O<+FRxW+MTDf$}h_1lbw=hvZl
z-C)sBN|<cp+=G?LJN3)6wca_MXjy*x44-+G?>;FZJM$zP>eJ7^Nw_f|>R*O?Hb2!i
zZ{jPz?kka*BNH_jZ`}9z<Yv{OE0(8O#mR)<P$<zm8hK!1wcXX7{fM{S@RG&Bn0X+f
zH}MJdVVA9%Y)*5v?^ku49%QjORs&jgp<=6<ANT#ugy4$+vv_pa1L6FHb{DCR%3lw|
zY@q6ezlNH5`ZOb)nMfOOnI6K>J0;fTc4w=X!3hgU2}ue!Z>DYaWXXrc*Gw3bZAl7c
z6MAq|kRh93-;T}h#`5!J_fJ80cLvr6$3EcmmYI>`D%JRzKG5Of&C|YK3eUnFh``9g
zX*@XKx%aseQ@D+@YM!8{HOL8~HvV>FL`O`5?qe4_wNtezheHX1c_jN?Nz|S3W6NzW
zJyEeee#14tRvwnrF90*7HpzM%zH;H&<7W+-Y>#9Xg^tJlww_Y#5*QYEaEgaC9$Ad^
zr^;vcIBVE4{|c!l*#Ijnv7$5F16qJ)wItJ_&|g`~si?NWiW3l3P)EC{YupVGjd3Vz
zI>5tkR~@bp9nQtG7XJ}cMk@U~+_C84h+S>TG?Xz%^Zc3=A?@-7($$wn&_iuBnz4n5
z+E2!pncHuZIH|8SuQ7+oUK$fy5W4jQXxEHhLo%wU{(%z8B=s5crj~%ov~<_VSa_;%
z&ZcW+yzwW|9;0X-otYDWvTgB%-^_vQs6x`%qr$o{Z#y_>Y$ow0q=i-Ztmg4;se+Rm
z-`}oJdydXYr#wcFf8?57lkRqkhCn9epIpcvgE25e(WD#ruR9i#HIN&hi<v`j<BK0F
zWGHX_Kw_PvZrJEiZ<iiqyGRqSu^V_1t0vU;RH$c>>a^=K3xCHnh%r6*OO^*oVKC|T
z4f$b~!NK02;`9-M6E1e@)y8WRrG(y+%ugXJc~E>d`wNG72460Xg@@N9KmQ|WXiSps
zp`PS-TP(YnF8Xuhd2ot({%;mZzo!!B)m|1{y>7xxuVQ9pa>nQfq>P2^7hEE?zja2V
z9{z5yb*#8DazBh#O!Dh1S5<+U-Kb8~BuUt)r0wd2*==2|7u-nJ3*9_wOzD-LPkHdJ
zGm&uYskzp@R<T0*qOOjoCllmQ#u8%9x~KDxoI({ppWl0VW%@mg{^CD^YNx}>0kKrp
zd7!7ZE<oWdyADfjq%RMolOv8VO_>hR(kzkuq4d!Cgwowp5%Fm673MZm4M$H#L4VT^
zN{9X`$-;*J4@u`4&*uBTeXW*OjSke-)+(tj)K+S5k|1Wq45Id?s;Jo1-dhl=#2#(c
z7K+$nkD{m%o9{or`+j^q%$e8qxz5jVyx%84+0#enjS)9oj{YB%FP+~u>Si{}epzji
zeLYns`$X1z_$WJ-#GkF+Tr;(Qm2d;zJY|1)j0aLiY>UQ2=r%3Hu1m#&=xh(|O>BBp
zh5^QJ+9~;fPXltd<eI5+1N*_k^O?)Ucx<xTzj>~;wm0!^H^76uczL*suCcwgoD$Rj
z45l?~lQQO&XCLj3_HCU1BO%#ng{f?2$iTx+z2q{_5--hapS&r_h7}PPnKhQL&$4<i
zMq&gqs;fC|Wb2Hh%lcnrm@Tn27dw?ZpT+Duhuj-kHh=WVI*xY=WvVW69hb(ea;W<}
zPI%PqTH9%>rR4fJs??!j#O?iV{k=3GQ401cnDZ?Bg|#Hwg}Ie7^D=CS{qJ0{FP(LB
zz&ozrk|OH;K&-*Hh#q1Z2vs@Vf1TV3lb3S{)Lc2sTscDgSxO6!Kdx!7AEBB3)39~7
z{>AH25MK>x+@ePRKfkyIJS^7x8N)LpTSlsV6BV=BI#qVN>oHB5*0_`sL{ItLl4s>_
z6kU~;Nx>6(lms&3@zZu?H_yb6J}|m(Lak!3HZ>zRXNq(78w&o_2h#gt68?&PXlYHB
z?>-Qfd3kgIPrOWECV%#v4M|+e`Rb+K*-FS0&J`1%8rf?AUrP4l7+qDWkKldYzj2o*
z+Gn4<bSZjW-G|ntA5CppD)tK{D)BB&^@h~`9qqS+Nm52KE_MfnFjahS)lt}=ZQn-x
zi<Yw}O?GgqEf4&m>wN+w+lSkU&}Zlp5x<TNCt?o);@j~a0fvpBV5|JQ{PlK4^Nx&-
zs*@SmyT2@EiusWq0u$XT^S_PNx{3M6`2ga+sHJe;Et9gwtv;)Xh;Bb~Qy6O+n=_@2
zL^5wUkBQoay}_`p8mul9$mrcl)=J8oDwEf+`9kvfd$l%Esv~RU>^VT<oWi!cqhyPU
zx&51DS8r*zTQ*B<?1#f`E=0z@zR>)1*@@wn{;o68R(G2_G3ed}jwrMnOA+;AC?>{v
z{POS3d>n&LYnSUh|KJ*;p>m!@w|GXbsr-)P<pSOqq#jzZEFY0zqwPu6#l9>H7(dRV
zoSRduq_}tRr?Bj|^AsFL{0o}gEjPZP*i6xs=5Xk0jwee<tj3pzECFlm?PR-k9z8*o
zp?}sW88<T(r*=;lJ&HLg8e+rbPOTd;mfuY5n>5qIE6Mh?IpsA*kZX<pwJbj>r3<Un
zIiFvE9iLljv{W$_+2A$0+G85DbKuofPTCB>JP}xl-5xx{q=Sfh(sAc)F8p(6mNzD=
zLTg;YkPMW`I+5(-Vv3EZ@X^~SOQBrbdp@U5StW&m&(B8enuwaFZ{G=eqyx6wWDULA
zBEC+IZXL9)oFx9U{|=4ltQcXjW9a`VVa25Sw?eeP#^ETQ>AF$giR<jva+bM-(jzn9
zO1*gw6|2C{-`7WgXGbf#k1fygjN5Qt(*20NeJ;n%RHY;7JpZj={wuwS8$RD1&;eWH
z`>CfsIvO!+ip}TxK_99gmM4blW^L*7TE2B;Ai$L9w{(-~uYJqcu4l<CA}addUw#et
z@B3|npq{7ZE7H={aWs$n>r+0px30TMjL>X!6R${`cUAupw?iBGqz*?CAvC9O;Lq`f
zZ2K4KlEn%k)No~&D%Ev;VspWQR}!C#d&Vl@W2>_`nbG9Z3ssxWt<T%OV6qi<Cr$$l
zyJn;6>_bgrQ8ewup(n(5Q9?t_)Qe(H#qmR{iK(lS`oOA>TzxEuw!qPkvafiBZU}xh
znPE$<^X0H3u?uw9vONqdQ|`PwusT)*p*-K!h0Nxk>?=fZU|+`;y<SPL`d`U6vFPXZ
zY(a*}eeUJ)g=)^|7lCC!i|vQ)FFz_6F@P^XpBrgI)93`&mDg;8Dr!`|qsBL)xy$cQ
z6_9vLpHxzmxv~HuaZ2N7@RjP(R+-;Q<`E*S#E5FfkV_&29Of}wDQeqrdYqr2d?e*Y
zDEn2CMq-za-s*nNWqJ#F{gUMH#%%m~)?ki~1gy>zWpiG1?t4<vxxKo1oL^jR33=S0
zOL~cMLM*0xR-I+J23{gZ$%T*elj>u9iTYZ!D*FOforgl5;kaVc&#WR_@9tTrCo#o?
z;s~7XR!j2B;r(VMu1iZoND#Ywk4qVb5B#>vf}F;5@sgjlK)i6$lB^DV+?uv<W19yP
z;mRB?yP6Cvf$i`IV)@S33vw91OFOJ#Z0%SXrHKdV&cD8=t`>ZkziO!0w@maKm!aP1
z!(Ewc-h_#G&+8b>A5Y8JSu>xSbts-!>Bx$Hn51Un^)jUdwPE~ZCtVwBm3+jNNeNtS
ze#Y$8e<WrsOGbYiKLl#KtP0-WYQcPodL{9&h_!?8<8vG;H<vniC#sZKn}4g0X=VZM
z7L8?zDKXw|avsrG1Kfvuy3t)uSaabGDk64`23|0%GP@z!gqTCSjlmBzM)xYY*R?a8
z1?k`qpq_5A?UJa1RkRRSqYF9=vRckzNsKP#=*+4y0@REaN_6}uJQJRT83;f!_-XJe
zquQlwzv;hj^MZZ!8`6IjZ%)QeFuK&6gbJ+gsI#oQwPs&{#j~vXfe+jGV>}wsaC<=x
zJL&M7e^~dthc_=B<*AP~qc3+pg~J$c5mmrUt8#*Y@5h1}|0Ip@4>@_nRQyxFb~*}M
z>FvQft&}#tN=xobmAuA(mASQ#_5R(}k~s()?1piu;!UG2^#X%&9iqM;s*=Q@S~Q1>
z?d<bYVHT>G_fDmq1|we&ddvJb1;6YO2@}V*cj{GR86~#|E`pkpbiOp|Urkr~zm+Z~
z4b2QjIdCKd;9Q$WJ#rlytxor~Q^k!>PSG@+pr_<iAB9as2Ivp{JOYd~SpZ%%L`O)P
zWlGd>by(MFUfZGOyO7xe*YX}9D)xk6>0b>oHcmXepktX*^>gk7R0}!Dd7TuGs;7^j
z;13(08d%qR<}q(yXhtHpzDWi4<Tn$8w<Jy!2b_&*JSittX0FZkJTLza51g*pHJA51
zHtl_|xB-kCv)oP+Rp%U5-@{jQq*a*qh>8FLAZ`9@7om?FCu2FwzHxc<ziiD({89nO
z_ssoa)@Sko8RGbrWeJ!{KkuCl^_#x_>baktWHLr8TQ3}ooYdHB2K#D4&U`KbTba6P
zz&$%RT%V|MfbA2n$m~Ydp#}OgpA~slvlp|_!Mm-V?)T`9_kV`Wa8GH)0dh=&N8NI|
z^Kyef8MU0*f4>Nh2zA*7JsOP%E<n5JCsq^p%?Nx0ro`Dz#;SdH*j4eLZG7Z1Qd(^_
z7WS8eZf+~)3BoAx4?KG`w)qaI!QJJ38LFC*Eyi|l`gr8Q_B-0!e!vA|GjfvT$+k%4
zmPg-Waskcb7~Mpb)kJgJMWwiUY1U(g9ydtJ$4RQ(k`$tw7>oiwXXp<ZXyrg{zOIL>
z`_@+bSU2r6_@ek-Y-Ya$m8yMX-mp`5=J=7v$h@7>_-jeK4y1L|iT;9lofnlWn5Aa|
z>bO;tAjgLX{v4)}`FgdpaOGDO&p&l`4cH(aQ**Jgb41sx)!g|hs`Ju-tL@mQunBqF
z>&N?=wGwig+$a5p3mx^p?E<17{}SeCg#Tj?y>Oil6us}w^Ly`P1h{WO(*X|0xdV0z
z(RYHq3@k2f8$CuaibIgHmhXw584ST0jRv9#G2N&blXl*-)6&(2L=K+-CcE)Qv3;@e
zz#nxizwbB~7_n|i81!r9-?if>!i=<F4Poydgqz_4BIdm=3}=W1Ou^RE5a>Opuf)$6
z`A9Jdta|IlNxmCXrv!#5vNvH6)KFnAe`#fHN({|RS~=<tWyr#1Xf#NImqKPPK+bb0
z$9j?l@6~pR_hozK;xTp~D8f^++NCok^zL{)u&sbd!ynHXiIFTse-zdir31620Qzj4
znW7@LNn)qtQt0=yW{VMUwwe3@O3O?Q0y}xf^f=vm#Sf!g%&e~_NE)pEgD#r7tiwUH
zq7d>ct=otc?ROGdu@AbnQ-xG|u~}ZtE2}e&v0k+hrQArg!_QAZC@Ya@&Z>4QYjCxL
zwohXR9N5CY@a^WeDXIYBjcafyDrlx!SRlAR2-`8cBDP2n%!Nccy434x37<;9Gx+9G
z+bR0qUKp6$ZpA@}3T*cYs8&a||K|CH0pFxZHGli-3TH;C$+g7DEs$L!-UKsUJapZY
z(S$=k;DJWJD+^wwk1c39`X2@A7bx=~9>BLT?o&d%XRN4GYDwdJkQ=`e|9JY$P`MdL
z0Hp7}+Fd>3Oj7zlzPiv!48l9H&x%~G=Ka!~dC`&BCgB87Ei-k0c<ah|&uJ#am3@v$
z_p^V=$VVA|zj&J2&I{WCcEX<2%gw#1E>Ux`^t|+w^ElrZ?S69R)c1O`^tpS7R5@{O
zSIxbHt!P<YMxcRXSAdb^*Zxv`y+<f`_D$nDx!mzqhu6RXq1=aw3?Q)9x1-m9KQjx}
zwc~3lik<1X`}{s(bw}H*TK3No`P7c~=K4wg-B*e$iKeuob_p~;Q~Ig;cQl6RGxD++
zXsaZ^lxlxwIk$9dWt)oi8xlT-o40Sd^*E-xLSbJ=ce^Ev0POnOmxHNd6}<?mJ7Ldv
zbw{vBhxd-t=SK>^{a^@DtwXE;^JMImJ=q7C^?B{?Sn=GE*Gj}s&&@3@bF4xX?A+gt
zfSOR5XVb+fRuGF)mHNv>k~xGYIGnk&<PAl|L}f_`yok*Q1|-_9h^l5eaam~)9Q{Ae
zMp=p2-OtnDo%h2{vhWz;v_Hy@v#>ZCS>C8tYN6OrRgwvD^Vp;iHbQ{}{?+Saa<m;b
zbp%6Zg>Ll0P|kH&t`K&$NxkKvS%ZJ%KmWetlA^V_QBMCBKX{w>$oA=-wF&L6%fduA
zK}Dm&O*TSl>&eo?;FPF4DXq?PMzW4km2?Oju0BNASY@wx_jgYFYIs1(tG_`3U~boa
z(lfpLW{vQlq8}U)wBkPA<-0ghdor|{D&|e^<VO|ge<U;*?wb2?&d$$ZVsE>!H-Ao1
zKwhTRP0qbJB*{jZB?1m~X$`8F|K2zf2hUMQNeNd*A6D~yY~cGPV(ac#aLYyQMsN%b
z6^4AjhDT^a>+Btcd*D%twcli$hnXikwf-Yn%Ddrn@mRjo5*R56dk^dJqwqJ{$eo|d
ze#vPxnfFg;k=@LZ{h^?YTz-2;0Yh?bw~B`eDB5j$jz$3b`I_EX)8BAXxbco%%_}Rf
zmH1cwM(F`YFI5WVFd}>xhSlg^INVTydsMq@oHx<?5iT!a??9i}o8@JqhrKi4<zJ`(
z(4mUp);z{P8{l466Xe<TqHMTZ_CUt>M5zg2A(03pB{Mepcrkt&FyZ`r+=WmUa4#pz
z1(I@dTE5mwY%envl*9FBe%WNS)w3EWLig;w<+fd-j#z?$%OT924zWDT!#Og-!%2%z
z!eiJ%a`eyU4h|Q)7MOACV>t-6`Fk_TIaN+aou&ukMJq%luHwydNpoUj!hE>g$95)$
zU?)j76#zrQo|1=sPv=8m&1e7hiIW2Fzj49Fx_pJr-LVC0NBg?b5Bo$%s`;EEAJB^k
zZ!NTcX$B`AxhE$RQNI^TfrH800Q&8|D&o>59zpnR61E;;=CK1tFq(W!%qDh0z9hDZ
z*{S5gIhyq2QpD?AFV|UEz?kdBePM7-R90et72{ZgaK`|-1wiUVxQ>5nwqce&0U^Uw
z;6@>0MX2`nRE04yoBxO+&&6;3E05D!3)_WypB)~=pFQ;-&J{uDeu?mcn&mlT0NIUj
zo0F^6b{k*+e_edlYE<LEX}n2sLUEtmzrIooeJ}4iL;C6MI#={W!5nf+fL{%t7;`bn
zzp`rvhITdQEKk&#YdrU)RbBK7a{MRpGnb|9usB9#=7yxO1BmlrdoTIdX;kk1m0@1j
z1)<Gg0Q=~5HKnqgZNawH&P#O<j^hxW(O2kaU?jtY>1Fxr%4Ln3JJ!uL_s-&2VJZ<0
z+>ay8e<7zYt<PZ_I~E+?^B1`mHjX{r{y@{Zw$tu+{F7snOQ$HPon?!gZqAc|;S>bf
zo(?4I1mCLTM;A}o9ic8ZV>`EF-1mpLHZX!!iZr}=O{92F)*c}wYeX>Zj|jwU3v|)o
zmI9?8DyFKJFL0Kv%znicP@H17-_0{fs1c3?J$F|l28GIuz9%EMVpU;Dphul(->{)%
zHtW?L3$GPft8*Q=wx+?+(`QSUQ+HRit<IgBCB?F_%j%qN3nFcg0ayQ#{1YH72lJ{6
zdJP(nntW;MV^sl+1saBg3T$3aR||e0WmWp6E7SKk^z8Kwc<NKxE_b!xd5WdIdgWfw
z(|^};-6b<9GeFS#2&=F~O$l-H@mBJAgzvtE{PHA~8%3nc&Xwgg{3M#us--c~ZD=6d
z-)Pdk()i4<Oxhl^08i}OHFO;F3qu4$y(Wo}RO?Y?$iU?_Z)~&U-VY*Mg=!weWb$a7
zd7bDh5GBjk_9k=rpAvBP-cv7?D(NF#A~!iAOf~pZYXt?p;(%@m_cjG35tm%_o@iNu
z$6Gi`=nd0(=p$ZZNd*6`WHpWLos=T-?;PFbcGS7696`t@7oO86?xhZe<QAdvUL+F7
z7dYo1a9u)d*ab*L5Q`+^5Q0bOS}vFUrR|s7lm7ie{M-n439|iOR(rF*qt_@#f?p{l
zT9!~I;2<~|GZEJ$hoG^9H<wy62NcvNl$O^al3z<#yciNsdHte5U_~~*%fH+@40(J_
z?J6gnOE}sO_j>kCU_-{}o3A~XGAfd;>oq~@*_b`TYQFh)9b(rfXm&JW(=!lcD7U={
z?**BJ0_&(#z!_<1gjVx$g@|I!Z-~yu=V8-GOK3amV!OoOuCXE~b|w9xLGM2fU2ctM
zm5L9NcZ~VM;W&2@ucC~w+EG=n)^=!|(vBIS{RaWfu;-*m<&^*g7U8!*MR^NVRdcPs
zwW~zbQZ6V)QUx`}Ei(%px(pobq;KaCmCIZCSo^QMID^l<h+=1;gcrYF{Ey1H(qy;A
z=Iph!wQG2Nh*-eqjr?fQhGimHcuf79Y!TW=!XrD*gjzffL&_nvE%$}U#KUnbN{)zq
zV1L;}teeQrhiD}Gr}WE!G>=#pyA|%CcT{%gCGf@X8a@hT#cQbDo8Cpy`S{Q=Q{jYI
zQTOk@bQMKQ_%Qw`hCEQ{@{w*etUKIDi(SB_p9^b0mi7`y@_sg7MZCOWSt(!yK3f?4
z@TOQg_OLfi9T{-5C!cPe)kqwH;pIvOC8d0cQ^*judL%`>hfBA-FQLf&f+;JBqQFR2
znJ<-9^1?uHD7~zb?3ooxE=56&;oC+D<bi%qrvytp9Qu{fdm@cnzYaP^bc9zTQ|~qv
zyENnDPvOe%2o;G9>pgJawFjpjinj^XP6kunzYs$00Fg0_OUT6TK1eIlT<Ay+(J_&`
zD84#Bu@f#L4nkajUwjZ3dHL)Kv&RJ}Nocy*i?;Tq@SY62|3|Isp(BNdX12J*S(M+$
zs(xOumD)ym6@qx|HGu+i!AAw?OFDjbU^t8h8*>BfbS`uJnHUc(v;lXVTfSeL3|@Wy
z9sOV*AA`ikcKRv2{oUYFnZ$=37gbI<??{0xZkz{9?8$`1r)|Bt`x4j7BQG&orPE=(
z1cJw4v%%>nH47UbBrh?uXRLe)dj1Clw7H@zI(H&PZfqL(YE(G-2=rTME(QlU#6<|@
z@KS#LM(+#>M_$eJEnYz24HKu8kpD;ohiK^3GFacfLNoEX>k`4QGd{nb%F&!HFVCe<
zgkc3|Vw6ynB<#a&Wtz9mBcH3Dwi+2N^IPvI91KTQL2EKRT)Y_d=OWmbp_<k(0Tb@v
z8B<i|<kNQ_75etEnhh1ch!s;T*-t+et|U%(`B@HN;b23X2VSeP(MLbCa-D7m`l<jf
z{^=|B`H`s1fqtFKDXKUIZDi@M)gFl`aIeqOmP&-pt)G@vLyBU<yowk1L3b8x)ms<J
zi5gO;oR0cWP)7*^v1NGXJXysd{_eh(U6@&Tse0h;P06YZ*V_aHAo5`I*R}MIi@;>O
z8LyQ{Wnx}$=A`h(^ssK4A^LzPy6Os8g7x#`dqXK7lpY6I(w#LtnGS-kxR7><h9`SZ
z=8FElA!}Pa$-J&0ShARz{mJB?;|l$c#CV(NF5I!8`FHw;sk9?$@aWA0zj6AVcn|aX
zt(j8ai5>My{`hw-2iL>V%gn5@#R1XLKkZu|OLAU7oHI3m6?oIYf6enS)>ya9jS66U
zupbBWvys%q3x038r=8-yQO!{u7kOJ2-Lkr5=v5S{3gZ+hT7R<6m74ZHlJBjPJ^7Cp
z-1`T-t2KWSE!QMh#A{ps7Pej~pxF88&wMXBxo&6DE_P`D@0pl@c<4B#O@e)FyAuzp
zW&Wc}`H%ZM;OAZ96KmrF#jy8FLaSwBgDL>=gQ<n)j49SA`%ibjS49<2F75`WG`gAe
zi0memezjKw1m&%*-Yz#++E5-LE)y{}QEKb~$5n$(jVeM4SV~H1ADef%6xXx+^?=2C
z<0^v?pW98`y_qDdK-`(XA-q4iIY|4_qMq%w0==^p#pb@1owIcd-r!cfTGK31QuDde
z-VF^E)ot#Br~YrB3do9ce6^RVcB2C6s~*=2g^Dl5TuiYCJu;!ec!11x%0+i+W+GP-
zaAIG^_ll`r#CUZFF#+ESqh2!*YK9iWJ6>gVL9Ybjz`AChHH}w*RojXBbS~9f>Gsu5
zKf4Xr(?sOz1kt;^hwY==|B=jUCe6J5i?N>46XBMiyf~?m*M>QV1cY2-D&JL#y2Yoq
z9gVAcQ5`&pInqfzbzd{$r!KTCculDMJx_nGz7ZvU3@l^Sc0de<c-M4!K1HanG<P-#
z5pToewi-5p<Vv;U4dvcT))n(`({T>QYCp{FBFS=g5Z8P0%*u?{C$F|MQ=fcv`Z$TX
z#U1RtQcvJlXi1>aIEtt9&E)5!cVtz{&7MruF<4iXDMob<s4^csF?CqG<fQCy3Ze0z
z{YR-d)_te*RjWJ0#Z)SBw$=1mp=lV5dUEz41bQsXZXCBYCZfR-kDv;I&?8G|1LdS2
z?QVM2?%ptGvZuavYxAEfJcIaEGh;c!-_&ok@%{Yc`1^a{{N7(ti@{~5_-}u<A_Yv`
zH#v@L^x^TG$2sClop@A1+p9Ud9g`N>HT@nEdEqik^#+BrXGASdtxEJ2o=>(@z%kYR
zx{xwXn|MjE)os@skhr(AtKEyhxonM0hIOl#+hwUNk4w_(*`@TIvM&68yRaBp&qgoF
zxbVK_V%e(J#)qfz(AW;+Eo?ZURILl~^0p-N#!2aaB-FenJ_bZMgBZRZeVA>4AW|yC
zX2DinV%V-N1xnWiuhQKzZ>;J_uw4BuMlPeZrOR>JoEcf(VQ&Flxu41I$D6db9@pCS
zYA|YgT!G@dYz*BUZp+}LJGAi?CP5YS=PIJzq)fZ|je|P=;prF>l_7B5Y|*?dT%C_b
za3)sM`;OW89u@k#bcdlu!Lqr>sYCv{)8H{$BuLQ~Y_{47C*<8b^V(eoOTLn=%q;CL
z8ubh^$(7Fn25bJ<H%axr3p-cgTT3l%=JY~Gpnu~(!j%)*3o1lruVz>%)L>pa#&tGW
zq0knC&pzH2zx(=6L*+G^Y3JH2B=v2aTw3x@V1~WOlwbrd^Y2|j4pu&8J7~j02{^o~
zrAmpT=>j+S-l&}e8aMHDW^VF>psE{0zw5XP!v*&>i#s+`K8XPuHZ`)P*qZwI>lIv#
zbMF@<{HEQ}Uj?E#-(7$i(Y519rS_5UpARTax|Qo24fsr+HuzQ?_Fpdm>NgpBT|37L
zY^_@xvi-*zLLv#pM<LPt@q52XO-`!Ckv4fR)<Q1I?eO|Btt`1TPz9h3L9C)XZF5Uy
zqI8dvxJ0lLIB2B}{XzU0Y1&^Ls7G6yFSeGxvTa2MFJ{%yG1Y>Hc|URztFoD96tV7u
zabGsfarcw-df^+1W+I!%rB!(X_r&gT7`|~PoOfz?gw$z+`4y-kubO23BWYKELNy~M
zN+U?RZ}^l0Bg8e=>^^VcOdIKDCPCX8{)~A&79Z?7mSy0=9JDQ!nAB5RVtf82ZsKtt
zDU7VtG80X*{%&#Y!T@nztIgm10de!CvuhvpEEzeNTfgo+=`y=_*#`Q`c>5sp8xnp`
zZjW!^EA5ANyOLyY6@dS+OtmU7+?Cm7wZu~#HlV}TpGegc_-#ND*63gHi@s0aVCbic
z^CowGY{T`E_MQ72cCuw*V6GtaJ_PZZ6(F@@Ub(MUTn)*H+SgNUeSaP&J<9ooOWiK~
zOO_eUGH`rrXT!K*L%I*^J{ckwy=x9eJ%FoS{*XjJZWa;2kiEJLi6~dSsgd1YKcbqc
z(3m6kh2K0;nP{kKcRY7m6SRc)HG0|gIEA=gNaZ#$mZtQH6v>N=OxQjYw9J|YiM+F6
z*c2mJQybN^KjYm7q~K?3ZNp7O7aI3!*Op?g0rcy-cort*U&W@h1lN70IA!Ttq<%f_
ziwPxDSvay0W2hC^22P(Q%2)EG@YKs1=Q_=X<1X+n&lCTYx`^q~rgvv7Q}P8Hl26#t
zo<`DFe@qJTt?iE9*1<D@!fjcge`H}n;u2W<2X1>9oLE3*#C7(sv=Gv(=YU#o7dg_S
zkXDP&zf)?sO>^hjTVl@XV~Y@cxP;8mB7lL3j+)YOl}VAC9LY)x_s^3GQPt)a5B17S
z6=kkL0e52B0zPPZHUFMZ@`YF|#h5TFAGe(afpZ@0$m}A-lk^6h`ECb^;`)J)nea|o
z`>Q*a4*hj#?)(Q{KV@R@1|P5CK3kvTdoigRh576${nVDvCf}atKzfC%@k71(K7t$k
zzV%~vu(|)#Mo6!2TxM@9kT+reTTd@V^PkR0|5M@%hPgZWEMlk6PeZBctAX!I2lK(*
z(UwVD=djCpl8%&d9XvG)1lesovB&z<ZbgUGe+o5k+`ZIE+Kb?#s@Nv=oLMRkj%cpA
zKRdfMkeRb5R3xl|pWE63wJEQt11mQYJM7)!0s@SR$O@}{$4+C}seju1B2ESLI!<33
z?GC%B?^hctvVp&ne{^abbDnkbeBFCFvH11~5GB@QjLe<rKl>cTsp!WRiS~*{VBZf2
zV?8W5>g+PS1&oa}r`O^(A|jstg<bYk*K61I?U-Ch)xUo!?;8v*=Cq5#+Ie#19ZvWC
z%GEY&{8TI{3%_G29K`~g#Bh8{SLlUq*lSsY+s5>>iAaE);7TOw8yNn-N4Zj<N~ad(
zUU;0L1x#d{9sh0bS>~^Fj^`|!F_dhCQ77Z@udl1&>heByG@X8u_(NEX+Zj`2K_TvE
zM=ZMf2{HOS`!)Tu9yK}oO%Qud{fGN>oNBOeqpv7sR<ozKxBF0+_&Q_%Pd3C*+W20>
z5?e%?o@ENwB_Ml)Yqogw5D<?mxL$ReY$aokAU3&y(kh_gTfiK(hS+xo!$WJ-XPlH}
z3|q)knoq=bc=1PuU8vd8!Ep^s?n1S1fL@w<ivhZkZDJD_Ddz0|g|Ee+MQ(KT$~eu9
za8>zr-0Rc1TuA#^hyB0a=v0T57YZFsz(1LmQ8DSWt8v*=(FQpg$GDxN+-0Y7L2{nl
zFpP6=Ii~~l^{)$~#H||+{`C!=96&+XFVaCXq1RQCCi7)>-ixzJ-V(UM@)R?~Rb{xj
zlV_eUXmsap@BuJ&Q&mI?<=Rj!YVs6X<n5B8bZITlBRJMv^TfAhQ3>N{ljuVejM0n!
zQWb+&`zc^xJM!4Kl>Y^GJyow`W~0OZKq8v!K(0XacF&zlOxCI{>-<^fL4x{COu(dE
z=9~&>mo4S^wX;-d^{JdZbFA0(w6n{bz%7x@y|{@n{!4)wtH7_l>>6uP^BJ;EN}7pf
zdKKQ1ge{BAlm;0S0oe2vph6Bs=_1{ss+)x*gwQPX)hhI@e~>9h2wTdXg_ubNRVscb
zJE>9q*ZA2s?!}|VC#R~=zh(TdU|iD|^2*JA@1jz^{jzIi-j&jfWkrbiP#m~Yo*^BA
zhpU!s4)QckT%`6)v0^<nouojUj240XlOs;f!(LLN!9fT{-A?7FQ;XIv-Yy|i0@xT8
z`?{J(gTBV6cvqFM6>9(OZ!SToI5IX6xyao6_(>jxkrXMroR-W!ZPAT<$XB9NT3soy
zo{|w6+u5bv+s=16POVG=jM#=f7rStL|MFJs;W#fdf1Lkl%Lj%TJ1ld!R7U;ruEM<{
z_)5|La<ICKM9-A-Z(?kfnm4azGHr+!kGn-bU&BdX9#2QCeWnxb4@856miJhliy2Nk
z^(6mJk~o3x$;;QLC?@R*`u)~jg*{+%xyCNahgx=<k@f;f5|!lxgS_%Q2LkeVMH^=;
zDpQ?}3eQM&wa8~Bh*N)U4tv-YJ^yn28jrm1RIy_)t1xAFr5xk9_xEfgOS;0ICan8_
zk<_zhFCo}hil*<DXmL#_J`-;`i|R)#qzup0MswvKm-3feOjL3zjTKH~GW2=T>4=}1
z{EEk9^&}7WkIQPXv&Fv$AGUbWaXJ$U2beKDGhLo5p(#ZP7syN@K3;yLXL(x}`$B%w
zpIhaJ;YM-GLGYhLXoY+Hr~V)VrMePR!-jZ4SqJzM7-EYmO7|bXO4O@oiwHHRr;R98
zX9)f>ao2Id>R4=OFGS8=crF|4ZDk}|VV}0%4Ys35-#7pCC>3zpVTl$_e>rjx{H1EH
z<%m#jOrL-M>Bee1hx4v_lgeB?6!F9012o=6-ghoDR%d^ObVn<psKKGcGD9K%-3Als
z<0h7|gWmS%ca2`&yBNR2p%y@4UZwJeEGAw9ZU_9HuT-=(wNbJ<E~d&*eA($8`WWV_
z;p|N3?ptQRjum+;A|PdWtFs#kg7LhGN3;LPJr-Q}UQC)vtl1(0PK!(C{v+AWzfZeY
zqD<Y)669DrCg9^X7FsYE@xK@<?!jTNK~<<J15oX4mdZNfXG#X-)Ad?_DiJ)%X`TK^
z?PuMds8Ml+1eL*e{@F4x)mVk;S9l|}|LThbrUv1tr?)a&UcTq|h;w?X@~G;}%FQ&U
z>#rXeKFRIVoW?WA1@vxoHrxYM;#__)vM?pm3KUD%W)3+%@T&BoY7U}l*S}%G)tMT@
zsBp*SS3e9?j|@n4b<Kxp(4HsTUuQcnHN(tq(hC|cja2!{W=P0MCXv^a`yRO+RyyQB
zefOR3k|=~bGvH#JBb`TIvC*K=6eBN)-)PaKH^N9qDtCK$SNWb+N9$hiwxV{7oF3}=
zDn507NUnu*n6Osm8}9TA+OD`N!d*}dkq>7XZZKXz=z15ih`u4CJILM)RXN<#CsNvT
zb=_~<Kym0Dt@4*??~~ESp~3qYkY#F>LQf`X;i!b_i)GE#$(G4*&IsYS9LZ`4)s&*-
zj}tR|Qa?7>M91o*qU9Yr?V>APl$I>RGN0a6$)jM+EqL!VI^*MA6W_U%mG>mg;@{o<
z@M>FZPZ~GOXlK1qCJ6AiG}6V7@O+S87xJY!ze+zzD!howXhs6W>_l3?_)QUBoESKR
zcBIcw65dj){1RBZr8jsF+GgYeYJ$Ill$!m^gO4Q92d7j@VD-;$hrP8a%{BQ3_Zt7@
z)I^CKfAnPsiT0A8Sz1MwAc{nQ(VyxfYp}Z(605r!>IFW;7u&J91*B<xWYEHNwf~d2
zsF38Wh^l`C)~cm$B`<C&$=qD3=Q86y<*y~mso}4trmCLyIf489%@<Q&ue>MQF1J!F
z2XECKo?c9SW>sjfP<&cso7EcR^bI+-nGxkb4dho|Pw!jwIa56ip(wDC84M3xi-&en
zXP=Hray<9PiHmobb?fT6nd?vFg5u^><A=lV`sfW$o<!Hf==yx&><cQL0bAx|C7$oh
zi{jygKQ-Q)i6;2zEZruN?Q9lg*$USS=cQXWSnXwkzE=~Q<IM~6K3zpwr=q{vgB}87
zTFMD3d@Bvp%nNn9;NTG>aG2?&7M@XBqn0E-tHi~mFL*fPaw5{VM^evfqlHRNRO}^5
z?^L@7^RHGrO1HYNOBaTJLBjL?pd-Fsl4$b88)fN!;n}3NM{kT&<yT(H^vh{%+{rS6
zQEa&iG-`n_kegm=L!xB*Uu3&&IbEK<OH}l+qPc08T9;|+Wp$`)Vor(YXy7CKWiEh_
zKJ)5+W7ebP!JfuqU4!MxV35zTYe^hOVqDWcoOXlMy#LtGi=w|j3oK0{_;hC*?lER6
zpOnsPSg=rxZrkJ55gfB_3zb+#nUA+B8#R3F%GEZ#g`7TW8vc%M0UihG|Ftu`B7!@Q
z^Jsig_S@6M`dV9r<cG_}61by(L48_Ss2PSA#ZNs;mm>jNtlte>rd~Fe?PyQG`By{t
zo3JkovS%Rdn@40-0tA@j(SF}bR|1YJRiJgF%abPK&ph@au!%K3leuU9Oh>Kp^rnsn
zc_LQ#{sN+><ajbt3Nfp)KT23+)_;$cisZ=7Y<F;-<v8Wfi4>82A<C9}LUMd-KDW69
ztZ4Q)=6nkI*<20ZJ`B>(J`(mOz;x%<YfvXO&=iI-h$?o)%}baX`^VzQ$n`<kAAlk8
z>D9~?fS!zUTGk0VpujxaQuv-J7v^Vzbx;*iMZ_}S;%T?pJoOtH+|mx$bh-8IuES?3
ztN@HNw3LTu-@3J(9Q+V_-#}%!^>vMGZ3bjf;C;|tFFDp0H4dJ2PQv0fz{@EqATFx<
zL21k?bMT*ii>)R8_bRO{$?Xd4v4OSu+rdsM?s{>bJZr4~LE2j!Qsz-wUMJPz9J%T!
zj*u0QwBXnhFNJ?2JpOwJr}5rutf_yao(`bm03G9>`jJ(^x`>Ar&@EPY;{SENm$d9P
zn(fC)d?|R+;qAa{ISkwK)L9UK=f_3bNnGA~8r3)XEb;=h)8~O&()?Yscol?TI-k7g
zEK<8ZLc)vkN9&l#Zt=9N!>wj9wV(g%|5p7dC;5Maz*s#|2W4BUKsCE4DEso}m0_cy
zDqvsXLA<{<F8g9k?~ZA#Xyb9&4P&Vf{t85-irJDQwZD6<q%`LjpbsYyaFuk7=$VJY
zBVtcdI{|UoX3+OD-Hd|bIT1%@AWAoaQ5e9^>#5<p)=x^sp<fQv6L&I6;$-|jSRGk(
z{X6+}F0B5HB=bm~#`wBnS2O+#an+$C%3=iz9j6<VaCuY=p4rZ1d_WPGj$VI{-$@-`
z6YgE8hy1DWUvrll12Sx+FG`f?nI_@4y_#soYTE}jo6ywTmHahjbbiJ_j+gLTm#vK2
z%*tblb^shCGC8+x!lYnsEYX^CkVh#Vt;voxsfv(0KC5B$&r|}6@9iKffffBY(APfp
z0u(cOi7e3!YTb_*oc+APGq(f%_A>Zm1tzl*quKc;iguy2`qjWh##()l-vy|)FW^A2
z2}}3995mmI1rawW5SVKkEhA>ydp6#ol0wPQ8(FBknY!{v%olvCB1;&~8q4ie70UP#
z7i?9=cae<)eGd>{&NTm7bi<-wZ=h&R%tV?Y8TM`-@)p65u`f5$!_GXA;B(?S0wLXA
zJ`#LP%hp*UnCcU>igZzL`M|L`7I7Z5<?aKcjow#?tRx!9=o;N-^S<t#6d1bPetKgv
zHts{+*|nwFW>2Tc+A|N*zg|=_uul-ufj_kpMkdfFyX{<o5CxzH**~|9uOl4TTbG-}
z-eRA=>3-3;e@kAPiR8UOe~jTu&rohX!e)s;>^14kgF{7X3c7*Qdd*N9<Xxg(G_bj9
zC*pxo-rh>xUI&6e@+bh3)9RvK83VEH#E#Xw;6$xI0FOIUwf`gO+A)oZcrDx%x^Dt?
zhIxy~<y2a%Zk)5PLnpaaJo7|tCZtPe+dM0i17qIoniwUMh@TcJiI!#!?PS()7Nk8$
z)V;iZeSxD(%%_IWzpjWfa#`h(Ja^ksV4BsSx024B>(HmFg8A^2*l#g|6Hd@Bj$gX&
zQ;-U3;03)mR0Z9ho%<i@Neu2AXtk!t4-_ggqv#URRhD({^-aH47pZqW>y^U{I$kpD
zuF?2!djk2I^qc%AqcUjsLHF5Nz^rpN^Z7SGqN`#-{54h&@``9j`Jf_Q`(vZ7TF#)W
z+DQW<FYvnCzON*UyV$2TdmGM*u~P>0!fS|K`vtg@$1b=R`0CZpmICvIsYE83)RYJO
z({!nE<K@IpNB5LOQx=MSm%|Hpk&k4e&|lrTC6`=R;JoFKp4b;#c5_ZLg$v6bT>zHb
zPABfewrzFlFzK+DRza~Lau?=9sV$Y7>;rFxkLJiZ=q?2>nXU|7qb$kmr<(JyxL2w(
z8~tm$TX|@e7ZS*%Kc0GNubKmLmm9J$L43o+(^?}pB`c<vBvtcb@=M}`x{)iiZ^9I%
z0;XsNX1SeuK+_(CciZ+!G_vxgw4@r@;%6iD+P5u#X8ztY8+#R}JS%D}^`(bM(r>6{
zs92hm8C~{U8)MP2H#SrKUc-2n^9(aCkeyU!?|*V_O}b9ho}3*8vE?k!p%^>E+TV-@
zo12{P9)#&>##(#b@#+C!VdfJ=-?)KP1m^vcP2|4u`)}b`f3b<gC8IpCV<|k_OLU+-
zN2|cemf5_9M4bf0Ioq&!f^n+fviYrsv0X!4%xZQ`p|tJ;orKB()5Ni=Mf`)vc|W;{
zVaHP{s;q>&W}_Pt50e%p*-~~4Q&UwFAyV^O^^fWI?ffb)TB1@FCL)W67U%TJ$;r{K
zaLvn2MuzHeOq@|^nW!uG!u=JrY5zuG(P7nG_^<DRr03vhUhy8IrW?KpC;Dz2ZT(hg
z=5~ELvggx>+sCKX7L1hRHZi+9=EGTJJNITH?lIC0y9Usi45cB3`=mRWFup%v@5yOX
zzlOObPiB-0fn5E)8w%q#=}=#tB4G&tIq8xnG9E^Yi87SHnl)bABQfn#l?9@53Vb)t
zXb6Xqm%@#0Z4F|CbDL58g2-<XtbCdi=Z(}UGUWoW`Q9TdH$rh8B-%eNsG@t4b16P+
zvZ71$6*Azokb}?CM);X$g@LGF5S(x%f!#OxpRx~$6y2khT5}^*@DpRcSd*MaiNWuu
z<*5b}{T$z<=^#M(`$jyE83@IyQ4FyA_Nvj=^W^Nj#k1R?e?>F(5{>+207=V~)H^S3
zNiD;1wWx^HS_f^nY3cG!%WDv|#(b!`#NNsBw>!^w4LP1n=EwSYqLMPQ%%dUaJ#I&n
zDcNBN{~&l|I=Lb*hdsWZp(2VqZ8lzanXn6S;e1K_jB(<o=kmMk#G_xPYf*&tp&xN$
z+rN7*s9!)%k^`;sl32;#n8j9Awj!A6nw_49C+yyl&=`|W@)yVnJkW}NbRN>;G+s}o
zp6|;I;bPynYqM8uV{>=yqx+Moy>}7|{ztCh4;fuFl;qgf=H9t9>Y|KHGQ{|Lw~TFF
zwEfh%_vfz9e<YM%P8kKJcKr}D%1NGX7cvLL+85REcL`l+PfI01k*O~Br|PaDEGlLw
znXvhQsrIBa?=2aaB%@?aLZCcKJo!E4w{)DRwPa0E2kRMH@R$(0HvhxTWB}k^1aR<-
z5htt^_E#EM+ftILmRFR#Z?2+J@+P7>9W%s4sc~N@g%2-rVD9a>4-AVy;1`U*o~UMC
zni$C}ynz=tKmxN37;iMBV{WFM&m=ML_BUi+FLERlCzjtzQZe0sq}Am=UJTQoC%52*
zE-7gzHVV+P`H!Qr;>V(m?@i4=(Q+Ng+h9ivclzO&tW*|-Nsmg$QfIHJB)muzuG_b=
z6!X&Lp`I~vmT<ex9%(&k)~69R6nG)0sWu7jZjw06Y6tB*{;83lQ?+LSl(efQ?jX}l
zLus{rl6YkDF3)5g*t=KD5lN_~_x?3Yl=f|Rx5e%l3*lGMW)bH`llgT8NCW)#zyR}P
zZGJNw#zla<x#0dBsyH90O56<?!m@UVYIs9NMW>*NEH|=QycH7)uax=5gAtL--V!3!
zPF>K7s+iJ)<yUA?2JJ}K78RPLwCFcMSR}&zla1cBP)%iy-Wz>8VrFRDs4iP^Z%=mH
zE7^-g*kxdRRrMG((FGb^DN(KEKY11d)LhTZ6oZDO8Ckn}oDbNs-dLhV$#Djo?7KmZ
z;x+AFj8yZEttiLnBSVIznLkwgVyTwR7vn9}s`46&Fvr1e<_J;bOHOhzB6XGTK!>Uo
zBRuay${mWOa8u}KOfh`DQB-0{2+Y*D^_ZxXB;_D>Up!l>|KhS;y1DK|wN*Z6ZNcf3
zA%gsIt!a5DEjX7WftR<%iR>5mzNfN5sH)a9#^U@3RL?<kI79Tmm@-ybIK!L|^lJ?B
z{}OtnF~}>D;P$Q2F_lgsHVoe~{Cmn5!OpP?tu6%I*1CBgStF0{CNcB|YOK1pn<x5W
z^|CgcLAPX!?do?yekqD08)lQSnW@g>3NDV@bgHMZaNR+ht3_y$b=z|yxRJ{miyY%t
z2^uyr#hq4ZPR`Ga(A9)IH~x2T&R9g;f0Wxyjp@xCkbc|nu-Y*t@ogdsky7AtxOemF
zV&Y{m6!-OW-!GVS5O2POXO{V#WTn)BBn8cIdl0Xu@u2$SwcGIOwr&J-E|tqzkE%R?
zE-6X;ytdvh>V}}o=560@jn(FtH3}J$=i52<BTsm;m%L>b9yv5q-KrY4=K*~?RDx2)
z?C^h{3pHgbOm<5U>6l0WzxCbpddx-T&2kBrsMH=5A`mI|7%tkfiwXX+$!eV<1iW-R
z)!_vSn#h#Vk}9R(ymgpjb)RLb<WJGa*!*p>r=-A*7A+^Ss8-~0I~nk1B*C6pMa9Vh
zVLhxj0PFO!zBep{IU$x!$5%f5Yy3q%YTwiyQ68%A(VH$6%-uonIC2t}oP1>0&as>q
z#q+PD85Z(D%)oT04nX=l?q_}Q;X1FJk;iGk``6Fb*a)m{>}0_&OanE}LQ0&(cQ>BK
zzRib4u&P3LO+HQ~zdUSZYqJ#DP*ewR>O+_}D(X;k*)b5ZQX;o_-(<iCT)6?%5vOHV
ze(XZqa}Gnf?d$b*A$aF)rQ;VoD_rjP=m6Be0F4qznz~=0UE%_aE%Elqv+<M81Cgx@
z%HOF(%C$>PZJ@GH+$>~-p|1u@Gbhm#9e@UY>;UWBL&N9-D2mZ!foPeY?X5MQeQ~%X
z9Yl_bqQDf$`g1p~HMKX4$6I`a&9^rMI=>8KV0FqH|KT{+Vgj2jrxNq{W-88&#ks>z
zB9kA)>|TIP>1G~~2<Jhwl&!129|Ql42e3SupZ<3hcIl62>@#p{X^GW;>qE_OWad(}
zICFt%TlTr@oKjr}YI~X_wq10`u4Uy|lG|fSuihPjh(wFiKdRi^y9CX~%&ji?gMou8
zzt}L3ZIwJUyK1jX6{JsVlpTTxtx<7`D@~Snx;NmDcr2mE{3a(?g?0RO;<F;1!c{w4
z6ELp29{&IcdifCjokjR{wGYV1nAwf=H8FZIp}EZBBO*wv%}&k^?Q$|~4IM!{iy-2W
zS$~~Q0{=06m7<KW<oAym<6jloC3082*5Z6C5^ImeKD2samKQL}Cg3QJ8fY!U5&y}d
zHkiT(#*kki$blcLT{}0TYkpt)x@m2aYh<7}S9rEi`PV@6<unTuirc8fq$xpT=GjVR
z<35@4JAUlSuef(R27!x~{xzHdWXhZp+NZi2!#z=_f~yN@UJ6Nm>AjL==O2p;m922s
zb7@o;Aj+;rI_~X!_6@%f?>rW)K98*9&f)m&pACB(Rh20_VA@?2R3#ZjJ6@ZU;g1jV
z7==%?A-{?L^sAbl{(Oeo?TZqj0s1_udyqo$WM4vS^5{R3qMm{?#KfTK5SCL_rWcxQ
z*_I_?+uRFuNDiL70Vn!xZCOMX!wZ>>8Ec!oYM$5QC>l&NcRU03>;J~oVjYSB?0Q$C
z2X~^~Xv5BgFQ$sm7UtoZye&czRym@U(!TO~Nx#0MI9vh|SAZ=vwdaJ4NZ%P}@8Y=;
z$Xl|0`n=a5HpyV__Uf7cR6!sRmpl80>1JOp8@YL<Tp}A+@GMRQ>A#Wo^CGksSExcf
zOrTwyYUkF)c$bh7Rpjn0hC9)w>-j+DT4_4o(alV$|4rf!l#O+AY+IN)Wq`oC&BZFl
zt2XVLmV2CQRW`2zvy$2zE2yFh<BtDj=yS=W9r1w!f4Gv&<AYHT-O;a|Q)mmHm$1&*
zps+l*&n2+lM*PZ(nBS;8(sik5+Q=Fy$S+_9Xh8^Nb%3Dp)_IUA@nL8bm&y?5j)9sg
zT4gk)#PfOmNc$;j_mEKD-|LL?xMyW+*+L~$3+d**|3~x;>63cMt<M-gR=pKdJv7uV
zSz0>>q3>C4;f%Z?)pcPbbgs|p4wR_bc?hfzE5BW^K0$x^H<^E6V@mvbiu@@R?G<l-
z*^?&J>np3(XiB4T(n<XjBD`<mxGYa@HapnfcXglW{U!uGmgfQqZ5d<tz_QpH!Ba04
zyAw3~a<@T}?DgIEqRoIW=6!!0KmIK<gxF#L(-XW^3vxu3_t3XvsUn&cG{D?g0#dE>
zHc<)y3UTxeX*V32Tz1;?=Y(c99w3|2lqQKt1CungBXs9oOKpXLmOX+8E6wMBWJaj2
z%ssz=5Mxz;ewJq$d;NOe0?|6Sgp3OXM-Zv@H1JfU3aLFtPWoexwzk(T(ie5{w71Z-
z-K`Vpds&mU1$nOitq=D9I2O8iSWC8V<2;Z(@jC{usxXBrc56PjOH7n}s<>JJI|k^D
zh2tk4=gSn>c>|kqFmgO{Xu8|;@yA3^IhOx_%+(LRT>@^}+lzQM9CFT}sK1n3`piwY
z)Eg`9vrH&|SYO%A86vo_QM!`;r~=kh;xM7&S7Hk$l)P(p(<1W==~{>_L%<zs8*`x<
z^?$pKyXOuOyDGDya~Qs_&fZu~QsdpCc&NG@BK!A1<&K&XBiL;Oir6-x4eY>;l{nlg
zax5;kO#R}a@c3=90DQ>z@Zbu3J7^wsaMi_;>2RNesNGanU(eIm)(B>vEGG*ISjf*U
zyCLeg?_WH*(I&AU<euo(lk|JNvt`6{@jpnhN&Hm1!c+8_83nHPa=JoK73$5`TGs@A
zCQ*+&f>Bg>SBw5w6-(3p#OXy>PzE=8I6bdS8BHs}>{I<+y|<uhZkWT>zqHEc9sCKx
z%%{pHsd}NmwZy&!Cu(Y}a-K%Vwyl-vzl^$NeQ&T@*=Z{gVJm&ISq%`GTQ_5x_!iSM
zaBEBZ7S2%|O=LinW38Pip@t&^@=|}C5ud}SJ?G*}Jeb1HzsV7)Ev*_oDq1swt?fje
zQunKjUZvww;&Nl<t6gbMWIu1;Yr5$_L*A(4YB_AQnmlx;eu$u88~{laZSV84lXGMw
zWye?98B8=+LF+~7XB7|`Ri);Yt&Ql%kF<$={0bi`BhDRUi3HXFq~XrG6ACu*o>&my
z+oaR6OsO>(WF|6Hz4;!_RXEITV4nex&&f0MPmXJp$H+5E=2;FOx6SFCpOcfd^C%}t
z);GTy$jRPCuUHR1T61j>CRCRxWiPW!$DcYSc2KcSv0MWzRoWkZ`|7-cx^*G_AIY85
zv*we4=+1lQLyUIZ0p2{F#O&DGn5~B;a*^!HM<Vpfv3N%>kAEp$at<GWypH}C`Pu3~
zKJpb1;3v~polniYV>X>`f~qbn2v2fz`|G-2&J%(sreG@2T`BR&X7s9lDM3QRlI8jp
zjKz2!8C>F|y@vV=SCRDjF$2-N`H!0-*SF!ZYnupF+T?lq%QALoUt8M)c7Cm?FDt>k
zXNh8#G-g91b@zYN^o_)_y>NZ(J9`32w9&irunfGJ%zGG-vfR8|nD%Ng)G||IoOxHn
zG5v!{ow?Wqenf&!{+npOt3G(IW8H$131z`+yi<M{vE|g^^?l1+`DZv7_ML9TA*v(z
zuBTKHqOsKeg%g;n|FXWpxfUOw@6a_c3>TRbkf2VPp^9UQHrUHAp-99Lb%ndY()+6j
zfcVa4WPJJs5>p}-l;<j0U0#=|=5>1I&kiqvasN@1RW`|pRy^MVdrn-h3ZfrBme!3p
zxva*|WqVfVK@+YkSu1Knj|rGMM{>q}h(M>G&xn|+wGH8SS*4nHJ_}1QCN}8#;rn1$
zv;*8AGDAUz23q3BwoW?v<?{b1I_s#W+cu7)C@KhwI+0dD6b2|UkXDrL*nm+=v%%<a
zl**$B(nyVtQDg8BqY<T%9&E&Dq@@`R&-;7-**W~N!#O+K`Q7*T`d*(8+e@sA>k?zr
z%8v-upD#ec+8?^cIc}l*$lHD4z2maU@y)%R<t5{`Pj7G*x3zzJ>#5((MY>c+Eqt&c
zqhd$_#{cIwwz1(Q_^|CpdcvJA<`ILY9J`?CrE@Ub84ip(?IaS)f=rG}!1?mWg)yuZ
z!p5I9MFbaalZ?)qz~VWw2+<v$S=u9#0E=@PEX@=}vOJjt+tQj=8KB3$OzVPCvQPs1
zC{Lv!U?5^_6`Zex7_Y&h!n*k_o;rD=zsIn50HWFVqpsr1+8r4WYS2+>;Q}8KLREpv
zn(Uml#klBtnUVE}Aa(PuI`-R*9Bd6Y`wNQ>KXAXVU#eJ09bu;8Hl$cVIhWOJ?beXl
zgQ$sVub4DrnPi}|GWN&0;6*;azt!h}mv~RM#A8)m_=(Nj7o6aD9%^1K***TuDlz~$
z7oc8k8!trXo`GCx4mnw{tnxX2Ycz}7vxo^~?R)H^tFqIP%CXZe@ivA|{c>aI#+~u$
z&5qr&W8)udGe3s&m^)bWbY881TOoLUq}e!k`sWIyDLBUe+J~+F?A+u<mx`+${!pZa
ziJxzqR>f1>_}x<1IpDFU7E3pF3%Z6Z+QDKhlxnVrxLrJYCAx{`EX}2>L{+>wyQQIB
zkRfCI{Y-I(Zz3p<_xr`)G#7dCG-D!~E3;sY>?HOwLKonTnXPy!PtK-joN#C77dIwP
zh-cbZ{`-TQ038_5R@|od8+vEaz@ywrc%>RE+=w4iKVt-&^smR>{f}mbzv+Eda`aTB
z@k-U=uu`e?*(#Vy+kIEHG1g6Zk{c@>AR9EZ`*XB3*gkd$KpqfBD=w<qkr}|E{Pn23
za~`YGHT01rwK+0a`~tN04f}Zkv9PLR@yrEmlH1F(tv&Es5b$z2D9#yn&Pu|L^Z5e7
z!z|j$>lohy>(k-8s+e1}Pj)mqw~LH_!p^5~M8%5H)10JJ3AE8}mfQjNWs@gomBLiW
z^KI?${hD*+Ad3!kL-%3;Akw7A92>si+fHZ<6t?+~=7->j0;iGr^@N~r7MTaZ<yZ^3
zXNFo#`VzK^KXLo;_*0N#J#Zys(S>ek9x$Iz2=T^Tv_fZ@6srCVhMw(F_N7akoHh3^
z<IPqzr}MR(B>NGa%HxMBU~7t&=mDsS6T=jKReFvxXv~ojYGfVprN(Go|2-HcV#_eX
zY!oyIj;!*^GudtRX3d8b#uEf5dG`?*X(?Y2OPxvp#-LfFr*_!q07MpHC+7G|xW@^=
z7hY(7yPcq+?0B(>ENd8!&MIk$R2+lB{_Q=rP+!<oC0Cbt6Z=BKJ20J`!5U?Ms$60m
zwX+swPlaFcUC0bYxwJ}ymdP>^v4L|IT<JwZAx@~kbvwRp2rB2Q+f{2zO7ed+C#13i
zJjrra=zE9f>LboIDq)aw4Nol{XfVaO?MSlod=#(v^or+C5W%4bGO&6QW7pw^44R%}
zFT&sbYiIvvMi?d>{$psFpqHZIeeZVka(56S)}&}^B01d2w7L{*eLELpC+}C-6C}YH
zMcJK?62K)ExD@j#0$T7XfSnnnioGti51j32CQpq>+;av3Q!u@6haNgkCL4JZa(&HM
zFyqJ_Q4^fQHHD_4oi3KG%TB%lwF}&Mk)fB2%Y(#<%p}gDhTsXvV+mT(8thFegU?ec
z3g*Piw<#>Vz;XmSHdW$6u0HGsoW@@EFx#)d1wgo65`>u67&Hq~4-0u(y>R+Pku_X_
zp>RhA=EA9{@8MGZflfV#LBLAGy%ccT1@sNxm;K3R*X!<0RSm58eRCSWRfF#5C|!71
zGJz-juFGMeuu)aoNX-Mw{1oJ2r2$=DFpyRdO7w&4!TdlAAKxJNu)Fw87pZ0-@dL-z
zRZ|S1HEZG)B$!XN{q5Q?OUzqpaZ|V0cnx6GNhZjZgP2|L;k|<{YT*dcK3%rN)tC3P
zxu|vdrK@*C_Pk1v-d5-puNX6ip_g;fR#*dmRb~5M66<llvug0}QoW|hMe7z(Maf=6
zUB?adbDOr^o#^rM#~ux~X<wuZl=mbzM`A0ATImyK*xgOU`oIsmr8!QGb+L;<dhJZ^
z=GEz76)7eH+0id%GnvDt2Jdv2M;kcky);9YX>`PI-5m?Qg%~H@i*D$9+rZf?<3J9(
zWa*kvX#{lUjH7|uo#L}5V>`y_JUbn^G;fHKYs#Lny*@Yi5RvxHppQ=EembXwr*5Rv
zr2=b^&+|2z(L6s3P}MlY<?Q%~QH%-Bm2w7JxnODJ9<S*5-M7o(zG~LiuhGd;^fYQM
zfO&Vgip=gACCF*&(|fM|gf6w9yXvakT`kr3a~rRP*<ydLz?=1&-!*?-Xq&m!A1OyE
zl~~#x@;`RK2Em-9uG)n?5V?g7{CLW6wBcnH)?;mE>$`LB;mHB6h}%i0-5{wF<FM+Q
zEcgX0N&2~NxOx1FT0#kv`@O?mA9&e9<Lf`w+FH2k1S4Ty$%zon2r*D`g&Ik=wm42Z
zyt?6jdbqFWRn|TaE<tk6@cwbW4-Ji4Cjq(6(sR#tvje=e72V9-pOiarM~h{Vw+zob
zW3ZLI`)tE>he<owDse*E^TCT_?ST1W6FSxb^h%?5u{sWE_<JM1K@P^58qlEDUHTi%
z+t#MB2lB{;T){7i2=)_MHZ$=rdvSESbk#3<>gfe*PN7)8*6cqp*q2V(`tauWiCf|J
z%IorYuE{;E_$_-J&^ENSJr`6LKe4YDNTc4*|Fnd3yG8@d)ANS;CTb=6BrY}*R2C+(
zBO6n8PK-5PES3lS8~9s>EmYfef4eEHohPosyT}U|z|SM^913G@sFtplpJ%?AP=J?+
z_0)#cmrez3seks5jStUd>EoV;RKUbyqu)P7PF{UbT=T9bX943)32qxM>ptHkA)PQ@
z$D-7dnUoSAVW$I%aLs<Hje&FdldRI|ovS}EFCV<yFMXnC_qLOqMY78N&_f8bB<aj+
z6AA=BPt8E0y?zzbtA$I;QX%C?t`!eO!M5P3^WUpp$-Rjyv$j!H!hJXUw62;n_j$$#
z_H|;t(K$G2?lyom{iG8-W)R22^tab!yy2sk9V0A-NldWZh47KA?#i`RE~H><LAPdy
zL2U}2>0hir$;3K6m|XU#losF>l>~AVW}KN`g>Vw(=h^xrqcTK`ex>Xq5jzh*doYar
z@*FH5ER!KMPOgv;UW$DMw>4m*3OqPGF?_3j7d`T=UDQNauzD?vFY-(8VUh9jir^K%
zy8JHH6<!!KYCS^FQl_|NbgW<n&#wsf-!Iq%DA|<1>S!T}bR_&iLiX!HGN>`L5nmNJ
zR9vPH`!GGU8QFLgksy7VOKggti~gFaOM2w<`;X^3W=C84Lv5Jmr8UwDTKJDojh>*|
z^$uTO{G6~Be`$0XFFC5*mQP3Xi9GJ}&YLJ`!l#IcrKF!HgH1JRzwBIUDeVZqOkI}5
z(v@1l+D`O(Q?HKk>w2WvtmO5T%yabql%!4JQ4q9asY3<Xk&~$DAG-I7#)}t^?)>oV
zN}_Y!H2uQmPu2}-!qgFr%ibL}@dm$>WyrqnFGWt(Vzc(UFS_^37>;7|HtWR-ZA)f8
zPYWfrPK=nD6o)CSu7>w(8XSzB)0#o!9(M21u4*2LZkSp!j%Jgp9Mi4h#4$5_xffJx
zmtNQ@+t9(I6~WF4UaG_lf!>E(lx3{5DEF@o5qqkKJ$;$;%Y)gTouK%)XVpN6;uO%T
z9FiRUH<y@x;Ju&QFY&(P8~uuIMYL;w^#5N{>zbT?yrTkda(xt7bb5^d+l%A^{%);v
zczck)p%K!nP?Z{ur!kq#_!8f1r4bt3q}ni@A~4xZdrPAM)k>_$az$5HGS;zPt}B*j
z`igQ?xc16uDY*dtMYF?_sx$vrQEQYomYDQTB)rdL*F-jO6ylwN$qRoCjL5pJUVAzG
za(I%BfAG$3!+2xV_pKmNN&c#W(Z>XyZ$L`3m|lysRO1wY$&dWymCu<Pfd#f_0mzf?
z;yp9E4MXcF7}9&cWYtCT#8_WGlVE55x`v=DVBR5-cu@-yx^CnA$sK0-%w0zhHxXr|
z?lJx6P#UM5XdZ&WUO~c=L`$zhUK*!Yj=;<l*9G@};m?bnO&1%QoEY+Tsoe||+?O)w
z9Jw%3XXqR;5ZpXC8G0+RLH#u42b?0R9uF9rD5%S8GdHv!YD0U8n?cuAwe5r_(xyk=
zXOIT1`GFvQ`=TmUXL3QJcr0H}WN{zcM~m?pp-S+mM5l7?HvsMfmdLsV&{Mshh&{D8
zrgA&PG(5pmm!ZLRU6sA5_#1o2ByA#<)5+=YZ|R?m7ky%dA9ha<3y3Ie^kL-3PVxu#
zC~=+qO?LN_LE_cFj+db2TEC_?{%@X9yLrK3$i7rC{PdDkKy5zVlqN-DWkm1lS5p|5
zDg-eZM)x;@^kURw9#82mcPrAqWM}Nw2TKBT^W)2YCuGWU*0$@5f@O{Tf0z>+p!V_3
z-d91JEdFoVGWvXt*c-cF;QpOsx(26dcHr)bhac=D9{ZE+3Z{6fH$-c)`}FS!iwkzB
zZ<uFf^V=$7UDphbC4Vljn>V%ERlL8)#K%OiNcFVG+6#&6yA(Zz{@jA$*OFV&q^=-1
z&EQ96uL&V*weEMHPpAKkLR=pp*7iFt8Hh2hoXe8pc}BlJa@_Bl`q*765`gLl<9CF+
z<DS~s^SZt=4<L$5id{AOe5Yo|KL7J4YGDmq9Avd?sX~ZUuHSHuS>I3Ff4A#z{K$Gn
zVVXMX<O!-??Q^+%$3>GlA}WJA(3|V^xo$wqShMh{7qBk!%T~MY;m~9JN7f1f6hZwv
z*am`DWiaiF8|N4S`7`%6#B;7IXf1=hlL>~dDV4yp{@J~i?z{rtXV{0&#=CwuZy4kK
z#Em{zVXOR~Vm<QS|1+C58V3*;@CSfGx}`E<Z0NEnntp;$ZIRHWnzqQ`aTY=@G!->v
zNgt!)emqR>E9`GwG#EFBK^s%e_zUB4!zd=&w|~6KOpj#e9kvIeCFEc8OGYh8J`QB`
zBgAtHz3b}p!7%>%;>a%j)Z-SB5V9gtlvyJqQgmLFYdR$`$#r}DBw~1MnecuYKyWh8
zEB-Sf+_7YM;C&OF_Az`vkT5?InfEskB<-mz#_v+q{lK7mm*m7(qD`kkQ9*+BYIjGa
zaW!}5Wfr~HsZnA9KSklV?@_uPmV9RZ@88ofQMY|JSQkuPMff94ea_LwL;uk{kV91a
z7<5$J%zSNiJwRtvC);0IjO7|$ctRw1F{&VSN{z2xxaThUTYzwf?o{Et;RsXUntjB^
z0?nuw063@&+FoBTJGM`kM>5UZ+mGok!i^vM%psqYjLVV4gl&RfAM+l+$HBib6LSVf
zgx<sBj<!~fY_;u=bXw0RDtYE!Q5CV!;dy?AR}j=(b;l1<3A<=foqSpcZ^$p;%9L9>
zFcCuAU7X?mfBd@KKeZ}!^`%kjyy}evtvyU-UWffi7H`OwDe>9nYljfH0#|IR!Kajz
z^(0dZ8J^ad+G5_%H13!ok{xu|=3$hIyQ$vdUPGpqvHs^x8`*0=o*LQuY?#5S?VY~w
zikkf6LSmn;|MuWe68g2|50aBv*BolX?6a++3u#@`T>JpAhi2M-|DL`492a8FRmZ>n
z6L%($lytgpVCP!7X=5^~U0J56L1}RL_pC7*(Uzh~1^o0~He#0nhs99t^VNo(_IFM`
zJg&koC(TO1Ri`o(Co}pDh?}n4_l@k1wVP~C(xvvg8FM?tdf8dI<<bD94xhs!n}WJa
z(aa<P-eXVrsOION#Thz4GpNsVZ0*OVu@10`-R`><qRB-1Si7b-gG;+BD7#y#$iL+W
zx9e@Sp`6P2h?{7#YnyK9=rkinvi4(}*jtgmr5@t@WxnUBjXck&NK-rAICUeh{j!Fj
z=Ry5w#kS0U4kcHeSSH@swoK~@l||5IYHTEOYOrm(y=Hzo;xo=2F?81W4DRQKv}zA}
z(6}A__1V7v3Y-CFdefP`lD^#%-1gnvB4z)1T$<sVVhz%__ln(xq8&~@>}8A+ot{oj
zCrWY=G-RXkrWV;AOhE}RO`#KYeGF`6JM@v}H9h@K(AS2SyI-tp;Oji~TiMMW<crpK
zMHuN&Q#XudD;SYuT^fJLLhtcN>3bsL#;nM~$a4jEUL2@Hc2qlW>Z+andWJ$m2fVK)
z>Qi+|DT6Zwj)#al9$L|^d9EHZb4ye?)KklB${XW$C*!~Liyw{lO|z3Dz}K(L26_Lr
zWEtg4eExOZFzS{ixPoa}E(t7nRG{mfFfOJ_@ZzE+-_;sZIZ$i^CRc1lWv?E$H$1<Y
z7eST`6|GFa61wGexvUtJI!1M6fp_AZE^OaaUW#3}6h$|WC;j?i7i$JI6G!p(;!cl4
z!pwU5Z?TpM@9@O>^)3kdWB9HOSzS`#E5RKI*mYlv#s+dTw)C;dl+t?j1qB31?A#tA
ztfZ(-G>u{d+h?25@;2(9l&vQnrYi)5nTe<Zhjk#~U~;OHxTK)1R)?Zmm=hEviv+B)
z-vP}4Elg<%)KhnA>lN?WuU9ej_D&_UgS5xgd|RSx{_&^eJE+`DIN&3h>PX(QMg=-$
z*onL66-hjv(dsUoP$Jmh8-TxQi@WLyuWA;Y|7t@qyhkxl;PV>R^=KTzK~-uPjV&=0
zSS?f6_c73hfoe3uR8&vl+G{;iEXP+Fwu-oC_;<L4CV5nf(?coWTX6HgYvTJQT_Pdg
z&d{z++Thy~KkH&Q)btcz=?y1ZSJ8CiJxzUOf;L-fa&7JUJ+%^VPd%j#udf6XD=|t<
zex>BKHtx9+%_$)%gb<yr)K&iw`29UFwZ!b3Z}sXMW5)4vYTg&uW7&(<0vz}JeC1#2
zEkih1P+RK_{*LzHXM@_|mh++T>JnH!4B#bVnqOc2UZx-ZifgfJiD389ZcXre;=tX2
zCV{J)!HP$+p0VK`MXm@$p&c*)6!-i{uGdlp->$VHr&ewBu~$>ElTzHTUS8f&8B#i%
zo?eRgWi7sR1o$sl=*6{R!n#nQ+N?J{9l`xhetG)`s~WiaPhDiuGjoFz|KYszWr7lL
z9X}VSxJJY=HhSm3|CG*#k*ASNZ~jXxJ{dlcN+QOkf3Q=nO6X!eDDoM;|C(**u;|hq
zoK%3pvpxgG2c781AbreBDJM#}mse_0(Bag%<E8CIV+d*rgUfS3i3zqTJ_8o{ME;{;
zv>$lvxXCU<7TOJHLmTq*q2>$KW~2z(LaB(f5?Edi(pz}X2mf@zK>W;GgDk#yfWNK>
zEU64ub-0I*Mgri(F+B&5eLU#+UH3W+2ru_4MJ8wc$@l+KarJ2(#)9bZ?0Rexb8~BW
zWU{bawIDxx(LGJJi20J1Jd)qBGpW{SkS-z3E(tQ?xH27J7yALR?EKoLlN-KS@;2&)
zOJjv*NwDM=oXn@5HW0Ra$Ab1ygGyJ|@S$Yfid=Q}>REcwMy+?dP33*oLYuEslk+8S
zvcsb_aA&4Cz3uUVd*P5O(!|X3?$R2a=*5BUQE<WL`xl+$2H_`MCQ4%0G>H6sNkh^f
zZs!%?JrBA4Hdes7BtAv^4pRnicuTqpS$Z9Iy`>hm{6nysm@atUdS&Xf97#alA5DG@
zPGB(IJ%t`iF36Gi4fl)e^So5Bg`G`8+TkyqOvO6J1H%6XfZfOyC_Qa=$&I5^f=>o0
zMwkCrH{Mz+%D)}nIRO#i-A#OK__6xmg}ACShs&+=H#%Q6K*HDq#?BoZUc*+r;O5w=
z*Id`I-V<8W=HD_%ZB8C8arJnfcMU>FwIBH|;#Bqn=ABC;K@WPI_);V*T=Z8&F(+?p
zp?v|*fMLk-&MRc3m9|afTE$g`R=OX5e2(2M108q!GRW+3N4!KzU@rdEBl<ddf!%<{
z!(x^_m*^1lBJ;ljR+d8UX&DKN+_&Z4RR60PN(&<UB|-1IHOQqZ2Mp|lb~3A!hROW#
zUJ7$W>iSo9T!{O1O{zasf5Dai!+eK74~3OJ?JS)DWcGe3o2B`0#|^hE?RyTw<T{p0
zzw2m0jSe2w*gH^SF*ODg<E2Zh35Pj$Et|Al<E&+`oc#<YJDO~`r_K@o2-oW%+W(|x
zI<D)R&33n|tE%Rl{p+b`AxbTBHWzvUf^9^a-td+53V2YV$(9I&u<QGU`*-Bh*y4^q
zb+$XvW}qudkY87S_YRfb()cp;jj?IX@jn_*{vk6}e-5JZk>}(3BDwIooSLM<Uq(WE
zRZ+<}cu;Uv&bwdNK;`jc2j7+$^V`}Of62gL8STlmBjnGNX9u9Vt;)nLqmrF^vBpvT
z-@T*J7yKmRYSzw{Nrh4~iZNsWy}ExI`0pLnLE<wECGFtHX<h(k&f)dr1<^lj8<4`O
zyZ~nNwXvGi+9<Fz;G0bL9g6eWZr13JySr-U`&(`jD)}phUmB8<J$PiMy~o4!Q-n@M
z@B%BjB%0O5J&l8Fuv<vnd9E>Enddgj#OR8x5^b952kTiU-JyjFZ5(#jE`X56$EY#%
zNTMDylY6{*?^;-bH<uOgF?KqO<pT1BzG2&4!9Hy1s*HVwrol3QjA>i(I8Q!iOyiRN
z$O6dA>?4yZ9o|v0*_n*bqQsc-H#}k==dxWQ)=^ph4$mEJRGm7+FABa|E`yvrn$=E1
z3+ZJT6>y)UkBsM<n#WxJK8o}a-zcjHVnyH`t~Pp9OBHgbqe(W7xxFnT<{OoJofq_`
z+i(0FI_^&k4sV#4O$(T1<wl+-kG6(qHyCe(ZZyqa-6?BHoq33AsL}pC){oplo%HRW
zc4vQg7QMs1U3Lwm)2CUtKc;Sm$+XTn>xm|(EM_#VuJ4weN}x&^xV{2IejP!x_cxId
zBO2LGtny3(mjb$ao1eiwx{S+<$r}WZ_a{(n;JE<)W#!Fw(ddpmtLeZ!qlNzSx+N?y
zKo(;wcM4X-I-1>SO02N4ZuX_(!9vSIZX<V)W@Rbev#`X3=cY;Te}WPiI04HDY|e7L
z#~(KDkh>y!o+FhgdF5jaxM}-G-hozy`2GbKbzz#FWH&s{B5m;g#5iw7=DV>L!`o$G
zkFlH2QybxLnO+p^<aAcLP*HcTag|WVM0b7F@CdadJa<XQDC3qhbe&=)_u0MsLWDGg
zjIr9j_}7jg=tduL)2l?6<xK@;R-n&fs9K50OM_dT#;tE}eVCPXS^cQ8LmNr7MM_`!
zO5=UOc2YBYEfT)V>;nG$HqE%3N1Zp-=|Lw@&v6a85V>e0XHQaI$!+eJd>LqfY4^0t
z_JY36_33*vO5Tui6%ucEas?EXYZ@c5%jBG!Y-x3t)W>fKwa)EQ?u8zX-C1!i9VyMz
zDe>CfSvP!7hu>54J=awg2rp&%BQV+{GZOwp0ZR50Ae?AgQxr?79Rvg9rQ<1_!S%r6
z^oRg^LKff;Es31F4RoUmA16OUXDs|Nr|*xI+AovQHN-!Pv%r)+?xq$%%26YD>%KGo
z3P>1SGk#r9-Fpj`_b;M|Kl2IJzK0jH=lWVNT`hXPpwD`PSg2mm)l_`fj!kc^D~Rg&
zsz~sh$mqUnrjdTBaG@>{)UPdKUjvF(u4$f{E)kkbdbm_8YvF)=TaNb+eEH?rjHhZz
zOI5Wg$MyRT%w?;_`^5=aLgs6H^Bn92YX9aW?2eJteZ<sNy)O>%!Kig#xb^d`ef-ot
zV`=Zas$k#neSW=jche-8z!T-O)$o-$DL}F!PuNo(#55N=`O65~Y2Q^<ld#`tTH2&K
zcQ<Te7H4Z6PcY3ce*TzuIg%3qHPH_-5VhcIXxQoYR9rDL!On7IV7^aef<B=%_YwP2
z?&<V?P$pb4^u>Ihk_cLA>HrJuJe&kN=6`@5s~{X#AL-XH{=!Bj-!U9r%zxImp)+L^
z>K^e6!4kpreaU4iQk8#AOKD`Km7Haj;;9qu+VoIpVX<RDo6;{}yo9|FB7NwezDT&0
zh_ILunK|=sQZc8GVVjYhL7;|fI0Yz9Qgo*yAHp-5pB)XgtKXO0`#pHMqoE#JXd@;v
z>n6WeUj$Sy14u{kXO(R#Z_-Bz5>J9YZE#uPtt|rL6p;QNC8qkv<g2IFDg)>TIfkiA
zTTtV}(0d2129&wHJbjN*F!yV?&khdyRJ;b1ZT&?DsOn}(l*~cRr67;@-alWEc)d$4
zhFASZ(~xifu8;q`v#!u=(R?_)_<oOwp!TI?uKn**pOPdl$q6F<Lf}_rJ&m*U)AZ<m
zBsPI+>FOJ#!&`dME<a0MRjq_@&nK_py@pjQHAP-@;*ocGzceTH*7{6{Hb64ZUf_Ih
zawbPnbsYg!PIyCcujrR!RjqlP%8VrVwKj972*=${S2zuE_=fSTssv7Vi;Q>%7w?tM
z><#lEjID|a4b7CeMNquBMoOpj#@V`Bas{&8`vI>Xb<4z8gV{YT6EW5FfR8uv3Gm6>
zGYxvHc7VHLW1-%GR8ZkQN0G8UihnO;pkV>wzSb22W<ZbZSgp81a;wRhoiR$rtGMDF
z!Ze$1cwNchOP`FK!@Uoc4toJ*#CqV6*9LpJzAL&K{^!x@IES%uR__e|aAXr$4W$(^
z^EIaIAd4^HuXwu$IKlM*NjUn7Ez5C>P}gmW0*5*mwcnlOEibp>@918hs85#&earGG
z5E)S(0}kjEwy+OB&C@~Tw5XC5|2=3t(QW@`;(M`a>*c)vgb<WkNk@sI9zQx2U;CDQ
z1{*|4WoFnJ+|tnjQxk8rnsuCeTsm7%Wv-VOVsC`AiYgEW5J}w}Zz>l+MS)%*O$Qp|
zWWV+H+p7l^l`md`LygcH!fDsqquGVi?qmh2pihn4Gw;bcqbvhU*>}oR02xNBx5TVw
z?7-`FE8(MD_^It*52eF)$Z3aQ{G!VJ^7oJTBlo<ih0b#>wTS4OMD(cooYk$Gu*O&U
zPWsE|f=1U|&3#KpzU#8(ZPkqh2p_S`>xRdHonju8{717~6-KTMk_ccYE8LmlV~cl<
zJb7XD9}P4p@koL4>gdV#Qd@tyQp=-@e)Y0%|IxU(ekK19GgEwml0^?M#sv@^gz_@J
z9<!f2Ji*IMDOGEWNfyY`xCw?oQ}qmcE@L!`>_9%Y1@ljnu)Owim@G*NNd*h_0Ctq1
z@U3%0ZjBoMPYn&ww&ld7z}%~;PI4phbWz%!Si_pm2~W7O27SmFc|DNHUK>HF-aJF}
zHD*|}aGh3~?vz_5pd`7hYu=eX;A&+j41~^s$KCtHwck=CGyBu$!V>pdvuJ`W>k3Oo
zRJs%`Nu+w|$93*Gj9wueu*0->m8-j_Fmwys96Q0${Hd9c?>??vO8VQ<WUo89a#Q%M
zA&^psSB>_3zF*R3+SuT1Uuj=H`3Lo&sx8AECSk@LSakJWdK02KxujQ3ZWKDZA@>Ct
z;Sq@7DdWL{g&AFx9=Rew!T;EX|G~J78hBb*R`NgMOS=>N$Xu*H(yedrExB@N41?aZ
zpD}BkNc1@)_`G>O25@Srwq{?F%b)N7LLxDqa2exAqnr^)cG_os5x3CfnhN^nr_0k=
zwKl((blQ&Zuh`JPFW~G37^2Gr@0wX-v+dMZNC5l5Tk+|B`H$b~6Y4xouCvq0IZ8W)
zT^GE>hQdya<F32d;>Rtn8|7xv_e|d4P7BLx1Ho%YUPp7o^NIJ4b7>;{jr*E!2$L7!
zajNWU>-9e}w|iTbxSvMkPkLHVGYa_c$e2=}`qi&O%IK%yy_as?zX3VIoxbst@Y9{Q
zEJqc0JMSh&%a4q1WPaWEjw9PzC1@;!pY@zq2XR-i+?oDr`uOuUtBMDq6^<!Vrev=X
zi~g>woO>yVMK795No)$7-E=3%BTh#@1hU>Zc2e3bsQ8j!2BP6vG`_d;M}QoTn#dAS
zIsU_ei4lb;@eFxEd?Z2soA>MNB|$+fd$RMD%wHx+sVc7UIJUoD_cvKRF$B8=XR>7`
zYq}fI5JOLErz7pk`8QH8s7AE06AkkBSDxy*{q$KuejT<4OZ6(wzQ%h?dZLbF-2&Iu
zQq;|<^7I!<u>k<*l7191@x9V4%koxH8Su;axBIymP3<aXUzYt#t=&_xoccLop<ZWa
z)1uQ?(qlcNKY|iGmhVt%FqF#mb_3IRx5r<o`>UFKpX|pQAo*Y8!$%)V8tdqPD?4l$
z_F`0Z*%xDaiJ!S`l2d3YzmBPCj|flZKzo`*dv0?7qagU{RJlea#ufUj9c}M9xU+7>
zea5$QFVXN#;g}Mjb|{pS#tT#Hb88>h>)Y!SRUfN8z?4KOB*;9^<LPjm(%X~(R8~Q2
zK=9E}EaP*@mraUI12y;Z&he+*Dg8ggmC9day%nws^P};v)5wXbPP{i0isE5{Plw9C
zEq0<wU+%~Yf?xPhae}{)U-C-(yWSw-VNWgO*$G<s@PBLJOFsUo_x_`)93{q|%AqmM
zb%B7n^GAZP4{IuahWu4kehdMvUKFRRCp*?3ucNz3lCB~k$Z_bfd`6Wop|b8d<iL(E
zlC#1-iePHH=muFQ`3xx;>N&79_`4c<J!ksKmj?B!Z8XfT{e#t9JJ>`*EZ0Nf-7owD
zk(8jF0b-V;=<wyw^T~ZU*~s&CmHU8mHF5XS;C$7i2Fe-KHnmw;OpBxjgfP-L(XO)T
z5FSE%efJpbel|9G4m2#uPB5sm($B=oUhU^u>g}AsZBo`P>`1loS+rwhVRC9sRu=vc
z{L9ThEfCS>dyQPV1hZ_$s_=tHhVRzA-VP5F0M!_jHF>c{oX+^~V*#qumNL&-u0+aZ
zKn~ihAxX@2UF{?%RXe_9NCeiVHQ6%c*B;LoL>~nf$1vH+7ur#DCLCQch1bm(^egWd
zbnBxEFGGN$IaScWh0~virP1a>SdQm|#fJZ*2?f1XjuO|afhcYpamwr%{B%uGSHsmO
z9eJu<v8Nam0AgN{%3n5K%FDaF4&_!@(@&zcnZL6mi#wHp7MCuH6pQQJwe`lk01F+F
zyPs+${)C0~0v$#VQjit8sOmU*3+^+tVSZ5fwC-~F5FwiLszHt2J|67Svuadx*GH&J
zQe@VzgnhN*=m`%2l|NRixOR@bJXfRMXI_0zjW3@|U}Mk7dAub2&e!jUIceW7A-GsT
z=qXJ}#CHTzg`_P(cag-MM!&#^hK+s|VdE6GF`x~W85k5+7+iyf)17t{><|VLyHVKz
zgcX1Cw>P%#Vs<)$j3l{U)=KC)uRhCS{OBqGEuePw8CWTw0+jPn>49L&k)M;pzv5p0
zQ$0kiPZxPd|4vuYce~V86D4JXS@~<~nchf_6@50{TCXUci>CisyI6tDvQrIBx$gSf
z#BgDd^hpf$Z@RLbT!@bw(tgKRWCkE%n+kK!QA@tOs<kPYGs-_M9u6q*-GMq8l{FKw
zn1s=A^cfzbVo|kd(SvzuQsY7*9tHEim7!fXZ?E}XSDLBn+@FGfR42y0u;j>CaW}#4
zC*)`MVWnmC4paY|it5-=))nM{trg-)t?Ik#k}K)!xFJ3}JL!IK%;)OvK5sZflz4{Y
zmpR_CpnKC*3X(*=2Zjg92bsFb!I42tD>YoGP}}7vqD8`#eB@5euTcRTywhI!(=d0|
zX*AQTi-ePf+05H!Yty%fMO>ClZ~rWaN*6Vme^xj1>GaJH{*gH11WdT<66RQTA+$}n
zyzf3$ABc*N+85OJ10~<G3mJ2a{sd4=a&F$>x}WhpwU4(Pqis?X<vN)cT<@uS%N;qN
zYrpRiO-ZhFfXGvar?hl5Y00>fcgB!$9gf0X!I32!Jd0NXz$vVpOe8CUs?;}A-XL_R
zS8gG8DID)J2-(OY&4L+KIVJ-&3c9Ar#il=AQR_mU^O@dun!$iO+v|JpM(DcujF)`d
zb?@+3&a-@(3ZYP{2R+R$M2=SHh5gl|wJ?kS2#UsdS{e~=a8!oual}ujw}`GWd?aTH
zS(?*hoS(gh&b^&n#>90iF3wl6DCn?NjxMD_a%R?cXRtIVM?_HnG&OKx%5~ky3)cPv
zsllBZLL`Jm)I7gw_hcW>-3UZebXmC@92W_YT)JrfdfvL<zl`vT?sfo9zJOS9Rtq0W
z=E~geKkXZ~`&{QbkGD?c^QQga297>*aI<b4>kK_5P4~H0@BexGd|}<<xo=nxq{ss#
z%=Y}x$f4<}$ku$J3%jFU(^W<NuF$B2@{iFd?0DFg`C%A816S|ts3wwiSG}vNRMEV>
zeQ5c`eqr1oId!D%)BS{&G9l!M7saT87~{rp^W$`_Hrwz9%i6+XC)St6Ac4Sqx6Ayp
z49PGm38b=R>p9JnR>kMA<=yTs+N(fjkz-pSG<;G}@(9Gme?>LImM@nV32Os;5)FUf
z@v6w5V0`2MbPSA<@r{>VATJhJ0o5lxZ?wZUStZ}xp*P*_7KliJm!wKVz7;6-iQQ@T
z$9gA9E?@WAffJ4GLFuvgB){6yEMN|TKc57+jaA0h#>U9d?<UU)hy@g@xW`(K&$0*e
z>U?`~nl@S#3X6B#6=mlg4Pxs7;eW+&{Fz<)@nKteE^SU1viIaO^)aEAIcHmS-EwOC
zKN>v@YY1&E*VH<q99H)Lofs#t2upkK!cIGYI95F<*^XOBU)kc5ekt0(-MVh2ji@j2
zH>v{^oVPd3XTDXXE4E=c1~sdp36dcB&+izNx&Ap{`b$;3EeFj@e*&+qrVtEtWBg%~
z6V&Jla7)FJRj08iNI^gL%jTJJTq*nqy74{#(Ef|}d&6381|wvPTnVbPq1lqT@uNXm
z$MFV&sz#mNdi!I&8x)B@F+>*y3!a^q_j6-hIska`I}fm^8^K;$!Ca57iLlT*NRGr0
z1)#Q8P4t_aEI+;Dnm*dc-xZf-XI-HD>OuRK=4##%b5ms7+YmYY-C@l49}UR&1hn`5
z?H&4zgc3<{2VHDuwOowJsOMa|t@5qJ&YcZ&pVYXlvR+-Ri^+Wx6G0|$zHExx=KZ%r
zl~1op{4tJR`gLu(Vqf}G^=SXh7FR-pr}tf&f=_5zhcd&r#adZb-)D!Zcz5(a`LZR!
z*NV^b>sL2gy4Ga45_;Ccv0ZB|*)xShqet8@N7si?7hJ=GEi3gSD>$ZoHa42IJ-|O}
zrqOpWM&r5j8EHTBGqtS7aVPOfuYSM0Q^#KiB8-K>8^tX>fUbx$rrieg>_!Y%qBQ9@
zV?*(3!ahQ-MN+3D?1eLGHxB=wu@u$}{rJE)wWYt6kCGr;H?3T@lJFl6nA?IL`VVhh
zI5jELVZ_aJLwu)MOB2a7<=fh}PQ~(I^nGYlZLy{q=e>n&$;+oub&)H9PBq34y&wth
zZdTc)th$C3%%g^{2zQ#)i_DBDlcVx{on)N+0-7w>zMETgK|X|V9~3lsI2iCYDNR)O
z=ZAtBzOUPo6e-0-L@2rL3+dygpm-5#MO*i|N4f+i9zB9A7WXYtTb>&AhDXsMGVCoP
zzq(2VgG|Ere9^6^vP>xjol9l#L6u$e&Gk2_wx=W|7AHXLghR3uMDLZl&`pNt|Gry@
z_qkPb1uDHUK)8ST!-Ncn0T{HUgPDv<U;vjP-kL>i7Z@7jx1@sagEMY+Qq~PR-L5Gq
zvxqnKmonr`X{)u{mZBM2*N_^rS!fQ@b+f{Q-D{9ne_(5(RQ&y<-jdToXp+Q-Y<EZY
ztdenONpe|o&tV-l?)kaE*NfnVBGE0aNGFOisz#i(O8@KyHZT9-DiH_=^5{)Hbpci-
zeR-n!j=}j>1F*cZ)-HW-c#QUa!N!YzfD#h?&+u96BIy!b7yEMNX`e2yq>Sh4be3z1
zy8*2S!l=*_SlXt4saxyPr|7%+bJPEz&~RCczPvLLcg9nAJ$3OLgdu3h185aLLyeFx
z89rEwzg%%v70g0l*fZG3qwNH5Dg9#wM^v-Ckob>AXz)rm%y>x)YO*`if*3JBz+KXq
zm+5OvLjQsGthR|pqgycxyr}mDPsWmCO8l3)R^fV2DdD)w0Jn)WW<<8TFFb;3cT&Nk
z!;9`cGD2f&Z9Cx=e9#e&XHPDUH%#t``gGwLj%zsR`=&H>7AJnyCmXu}r9;F^L^??>
zSZtXN*J=F71sC%2t_l_h?7J2}Cu1WVh#1<k?Y}{mLlKuuGWe6tE$-ZSX9m|utg(Og
zDVM*lN#_m56o|M^v%K(J9V$YZ!mltGdq$|+G$FP~P+$sEhfvx%gBVCoO_WkPO2(Ej
zw81)dM#ppt1ftUs0$8!Fm_zz%$INn2@o|5FJ=W5pL|gVyL~p$zw|Nk#fd4>DtXhGP
zDb+8AhK{K`o|)`_U-G^(O={JQC7fmPSD$TfMCXe$Fn@`@fyNgoK%O^Lhc5mmJCA%!
z^=3#g*RM{DxeJ#hAJ>71E4>PX3H_8p`*trjEhO(7(7&BzyqK-4-`w<}`d76vlN0~Z
z<gR66lX%;9w?fG2uixjBzGaf$9H)lKs@8dkfZvltsPL}5?OeQdwXZF7L%zn}hzIXE
zTp~hi+MzQqc7y95;}xsF;|n9M|0M48JTw&^EPIlPI9nLnyfrptvWp(A4}ct{xX)2;
zIFFudb`DQh6|p>BaSDx#d^0{15PbA~l*OQ>;tV&6Ilk3~8J?0jlIEh0ovG~@&P-|a
z@z2ko8)qCS;#;GTWnw-BkkG)0U3LXqNlVAN1O?lcS!PTI*gLX(I^F$R4`xFyOnpYF
zig7W$hxo1NgI2^+S}j8aTdwaDc7GCfvJ30Yzlp|_RT-tDhjl@z5@4NNauFT9n>Y;h
zpsWc>*|U$ipKA>u$dKy)(QqwHmuI{DuyC&F18_81jD4KJL&>)8X0V-2izytpvff}8
zb=?9~;?M(*Qg*p}1LcYN$1eW~F|l*=-lkxC$IbpcPpz7tvH!}DSIw8=Yw>jmxUoJ3
z1Bgs|pV>J4s$q1+z481E*$`8yfvrAeU4YC^yqVVCcTE$dmIZA84J&jkoc<P4UH`Pn
z+0N`uim|rRAv+>aL+l}IQON@Hv<-KpcB#t!sql!XqtcP6`=<{ie||GH<G6&GM}21q
zeMfBY_fwp=0;66CsxF!<FaIvfAfzQk=2&f)$d)oN??zE^IP>FxeMI)$?qCP|9%nJ9
z$BJAklc&<5jk~<9wWVmr?DD^Twy*C~2Ed@wH8=JjlWb+2bu!<DSVG8kw%)uyhFor9
z9OVLVAto52{cLgUM~kG;4fWx%kBPjH-&V>shKcg{e|@o!U6e%rmgTo-g<RPg9CfvE
zl+2nk-$-bVQ@PfKEF+HG{%3C=U-)FCqv>m@@L7^N%N11r#6czv<y$1H`n5;k-_s=i
zl)D8lmYO_UZXIX}J#E?Ogsc-c{RsXn`(}htV=vE|YqXF}OU4H4p%=2^$HoIP-wlL+
z)p7qE_h`bAf80Mx6ox;Tf0k&t$FwkAliA`q<ddGVE<^3GQL2@#$GceyjFfr#OXqzk
zrxUt>l+ABIH%JWWougSL6g;XEHvGop`D#d-=Z1N=M=sIBSP`O#(6+X4X#3)nv)lc+
z|8JCTFxJli5@>h&){hVF(p)3EF<!IT`Dz9Er_CyrHUmfzmv_QWH8Yh%;ttfAtT4d1
z9cz`C$Na_IIL$Y<?00R!RXR*FMB3~gO=^%U0yPn|=0YR+**aIga08<l;)<Xt=Lki4
z8gE#+29V!4dv2^wmry4CNuNu-?(fb+Xt(HB&lC$mvCz!TeNg+WM^xNFjlHz*Cr^0$
zitW$_TA^s^Dy%2=$k+x+f1akves>MraVA|Fw2ca^qm~(Mx38<yDE(aCDJ|?aZ&6RT
z04&9@qgE#Te>7zL*M+D>8|zlTRnr6d$^iT*<rvzvK_(?#nS~9BKK{|Bp0IA7PO6Qv
z7t0`!CWCfmmc1P*+z+Rnx*V4#7PCe`!{50G8<`B$5clg!>V}q2`0Og&t3GA~CwR}?
zf|!=LCiuQnT{Ok_5Ei@wIp@x8PM^VUR=5~il5*Bx>;1z~^aL9){CGL}b`fah2jh-Z
z;!vA8jf{O&?2X-wp(TIf)?vx1Y<ukokn?hr$(|sc?$Xc_$)Y2&BPOizcVwQhzUyKB
z-ug?RfL@nh<kMZU7$b#uUez=dj<l&8TwN!48vk*@eT4oUxxR|Z8y^`>Fwok6#7S?g
zy#?h<O!0Q#vw>)6eq1+?Jd}x*i>`8jq<eehX15ECzU=HkgeVOME|}EaBZZH$FwOWv
zjD+F?7vSz=mENGBPE}zNg4aiuV#7b9>wuTFcv=Hr5thh4pBSncR-(ek--HNuYA}w#
zs$_y@=J3~D4)!mX0mf0>_3n!S?x})chf`sf_f{IgafE=JCDGhrD4IUqMUsj)nS=X<
z=^}E>{@N0Xr4VJ3Y3u~o_uy46$Jagu*6&k#`U(<P1k<YXd`jyzgWo~z%V8TpUH6Y|
zd6dprahy|wjIpL<X4I#mM!TTxKD8(>*N`ovn5pCww&B{!I}x!aqmoN9Z;_PA8)%L|
zgyEqZGBs*-9;i4R&Yco#@2WE@4;QwZM<yikr}Di@5_)lve{FAp3!&kx+0gXD(}JIu
zRcXjj3{_ZoY3F>>a3MbsIBF=!(1jK-gtTiE8`&rGib+0LINL8$1&3$f;13i4x)=W9
zQV0jvRW176E~xv-Ve!RVoOaY1$=-q>M7JezG-Bvi&{6dP;KdhQa2Sc=wL@xNK-8|!
zKscht`&HHwl6(okzJ&Tmw(9AQq-*4iW+`{=F#(!l)+7#l?9`D#XShZ7$>8OmI_|Zk
z+T`NpGGZ~MXbMx__?lTE;jJyw*s2lAacSMuz4V|@;lxm4oImelU@=YQ<;*1DF)o1k
zbQ9Gfct+2@xkJtWQ0}_>k1#QAP0Op`i_yndi=XvtlSMlvx-{#Q=Cya-p}HWU$UXZU
z6hf0ZSs0?T_5hWwGC_0C1t$PVvJXy3UoTMKFL{7uzoW_1iRx~79!R!Z^*?gzwRt$(
zpNowQiAiPGFq`(Deo9YY4{MC63gtW*b;9Fb7gVHW5AaUZev59-AmugO0Du2;GQ$1?
zI8nG9=DQqMn%>uB0!cwBy40U&(t~qZD8*{d*=tMtQZv9h=T`K^yN-3Un`J>S;nM#<
z<Q<SlTIsfg4ouVX66<-$9V)-2QXX`2)fG#H8(1t{9OKr08(qD!RMd-|6$&Qbnk#<T
z)))v}NNlwKL78mgQ?aH$mQ-YxDp0u=<0ZDH$|EG7L!9;=)T}HC`x^2vCtMF({IU|c
z+ifPJ<+^-L&qId3^|NV^^NtxtE|MeV%2Htpr%t8#q@Qy~D|4ALKvft`u1~AovMg}J
zD|{1CrEvf)d43>DPK&QYdEOdQc9#;#+D!tFU6U@Sx{iWUvL{x>!X9cZH3Uk|CAmHh
zGN<1Q<XJ$J$uW^8CMMNIdfhDoh3xzh@BHGl5C3M_vgJB(b=94cJlmKntgg|9tZRK#
znY6DPao(5mTr!KCWU#&L1gO_C2tw}1fr3>(b8W+izjqu8YD-Rk2=9s0?u&<Q<~f`;
z1|mW{)f+&}>6Q9jEqTD#@eb_q|1q?M2pD^@3{4^+KnWj@Ui!+i5(wtu;)|*tWuT{h
z$w~JqHGvL{tu*-E$%$ld-H9w7`w1eN6mNY8>YPA<HgQ>4B6w9R5b;!W37Wc<xImkU
zg@!K1RfqL<ryg1e(4x$Zh;!vz+L6KZB3+A0KePIx=gMr4oC2#S&F;obd0I+c2S1sD
z%=k&3O-#Q|{E7NR$OWr)R5a=uAf=+VZ@$w8n!A-R^fBX(7elW1emW>57#n=F`&hoa
z)P&?R?d+P?{uVU}K&yE5?9`LFzBJ(4DGU+zwSgKCWW>$o7bW~&zYZW?wauPD`Kf(`
zJ#dN9>p+7=cP47QR8y}vY38x!MC+;X{hc?6Yj`&A=5(%E^>@SMB)#JV9kfvfODYTH
z`oL}p-ZR@7?GR#nM}=2jV*&JSz6n%9xV)Yb{H4SZ9aAw9`+;75y=I3FQy-PiIYzIx
z4)@qo>CiB*sQYOW!c&5(?L@5aHyB0Uy~Yi1i6TS33s`%&*Ba}h3dtT)-mLRg`2_Kj
z%~E^#B&&CAD|eDlF#W;6%?ZME2K`PZd>bz!u25x)c6fVYT(u(e4<<0@QrAJ5N>pdO
znWs9BlTG6I=%c6`)DtpY^gKi6)uV6$bMDGG=2Kg4_{NzcmQz2ZXj0=sy7)q1yUO@l
zPoiQ%h4>^FX#|T;X*Up=?Ai=a1|FL&ujZ;!%tUatpfctimau30q`kL3S9O#!Qk%ea
z#LdUV{W+Ancf{LMW7-8+M3`tx$dX7P64;j3#dM|}0W(Fv=+?vA)5X7g$Na`dyWk1+
z69`Xbmg{*N0|kf^t2T1=p~_s&Th`9){k{O|{Dw51%KzBSA9H+W?J;tq#ZXcHwHYGh
z>{MRUgYgh3<2ZO@m6%=~@1pQ5$nnx=jiKta$sui;RFz2pJvc;>7-SYqSUxJEDD4+`
zYUf{VcwamzmH<gCQPoi($Qg>eXC}tIL#Fem6ha!qBDnEr)sV9{Lw|1qla$JKzkSc=
zb`OC<!4re$6}8saOTYFQ-j6P7eC8n%6gV5Pd4_awPwu)gkS}W-bHv`k|Dxq?H?_=k
z$TfE)*Ff^qj6Z0cB|nsXPeNfm;CGvm43^M5-uiSx7i)!?L2h(EnE-4xNqoS*93d#n
zP3F6ELloUDIW%eOrWOl(l>C~71~boh{TQ&wBztf}2YN)lFSmSM5*;GJTwkpOe3=_s
z4u1r-l-vv{gDi1zk`#wV`#8$fWiO!QBSok(>56iKlsEn+n?lsKJ^s<hoFu3H(uQ6R
z-@f-MsfB@2^DM$uT!fXAz@Hcrw``FbQF6@*b(lA!@3|iCv)22%j+%e`y{{wgBad2;
z%|{E7BrY(fial@LLI*=gP|YrplFW^{el%bCtSqz$WUoTuw3$}1r<Muk7v!?oLNjc5
zZP^^Y+peCFdi?L4D<fWEqc=w(rCtw)Ru_}5Kzjm(d_i@LUoBezAz|2vUhZ4x?odsJ
zGe|ZgrtGEQZU=C&=9+N2!`yQ1m7$=^enBCF+L$F($#+Gt5dLonAS$BxkHk*if?t}+
z4h%y0AgX-nr&a?da>XXSd4!FkG|Jiac$~|yQsH5X<K%4Qho(I<cuSq^Z4yW9h@ztv
zTw#zKEBrIuu?G#a-7do-sy>jw5onl=nC<UU!30uq_h68Bk%WH(6!w`ypq2jCBcFcN
zXiu(msa_p#QUB3=<2(;{Mwh4Dp#Aa^AvF_0j`}M$`!A%e$DNNcE;B*frBuQt6mAY=
z`8T8)YK_8`3Z>VzToam93F-v71XrD@^^!K<WC2Bb7wvY5%c|k`T=D=wgAQPNqmwqN
zIGgfph=wPy<yZ>rkgg{oZRD(T{-be~8$KI|cSU?#+#N6vr}?5XHH`UOSENYVFI+ed
zq5^uB1W%0;SSoKoZ2^<x<u+3UVKJlOsJMD#&$?t&*4^mC7xRs-RlN+s-td6>lF3r1
z=e$sg0h3eY?noGAI_9o2Yqr_FqgrR{Sm4h8Qno<j(|NdAm!|;^bx<;;<5GBUf9oZy
zW(ptv!9A!N%Tk|XOE!AsqxFM>ZdmquC+Mc2+0ur={%qhx9_^ojB*3b~ewYRwPqm&U
z8^zh{0RPQ?N?S4gUh1dwVYMqc@xJBQtgBY3Kuf=liKVbO2KddLkp6UHY0Um{*}$wy
zIQ9zPrQ^%gSa$KPv_WZmPsqGCyjk?n3#XRCFki$pw0B>AJu)S%v&k!nr|yoq%QC1>
zt<^~}f$<-W%_>7PVU;;OeLipWcz9%baYWD=00V5E;>yR+T2J`{O&vU5-nS)rF>HC&
zMA8FICZ(n3>s^9zes;Rc>;CHItA$}#M1gm$_@xmb(lm`yF<`A@Bxl!7u8Dro#UoyY
zx=A{4Y_5g_;YJOl%&l4MGp9UKaBIsSW+2p;(nKd%5}DAAi<|R@I_LvfA}qoK^2nen
zcI1-1B)j`06_1zlLe%}TUkz`Ks+Oo(=?$_wB>>7WXXXZ{|8aEQ(QLkdA2+I6vsAPw
z+R{Oa+M6zWkHn^}Rf(-u%+hLW7Nx}AWKbi-uG)JhS|s+az32D2f6rh3$vMusPq?q^
z^Lf8tuf)Q+V9mYBt{0+wfJZQYbuTElPr$Gs5$A6H{UO=+_8cx@WTCAiK>-ppX*XBe
zSHyvuR7*8Qaw4^u#X!rnSYKq&FNo2qB~#k2_H_lOB%&#P_ru_$>@AD-)b3jc*u*4$
zPr+z;D8s*{Qmf81tL2B4c_|x*C>H)>Zn;w^G)HuR#v1R^Eai`ZfBle{>GV{v8;LgI
zH-g*-VJsZ-03DpHIFvsDU2jZj6+4Xd8IB$`b+t4C7JqCMDtqcpgc{y0K5ZxiE8_(?
zZ}dBNBw~oxDsT4QFHmh^C}#14?5Y{3aY+3d*inb!q~q#vJd&LCGSecxsj&4-jrV<k
zk)d+>9e0?bEx5$XM9AU++dg*AOY}#iiG#9Wl8I*+=|WIPRapeilQ)t)?U-Eqrd@!e
zXetsY0r(TpK18>Q@89o$Q0~7mf2k!U|M1$kT#?2)f!^q>1kucLzk)0A#${z7=1R3(
z&Kj{qyhvDGP4A&Ef7T{)f8&KHXL2WXDfrLlMvzxW$!71%FGQi>fr8NAHJW?%i~mt*
z9-LHK=b%L<ga*_`8m9F>wtN4=0FJ7h@^aX-5&z3gaP%j0__+w}1!|o?<YC`K#Hoq$
z9VMO4wCF6@S8f*dGEC<}dnUp$QbA<TGVY}=A%73McR3)aOfSW6(jo|ED#WZ6(iE!d
zf(Z=|+>zChZy)(w^&mIIYLz`)rG@PeQjmmyc(B-COlFN#$1Vx^ii%*Y2ld~5nY9p<
zCv@PYl3y(^0gY7eOquP?ztBt2+s3X^-nLzTDP@s64%N~AW*^Tu$;>qxsjk-+)A%IG
zT7E)vEa>R6bxXC6YKetyUc+8ls|RRAb#c$HSQ5B8^UEFzT{#X68{!-=@mn;(NAuS8
z^UT-(qi}NC;FmFzT3{JzGTILLdNmD^5Sw%m*E-FvR_64xV;7TZ$?1t8(Pb2Ds_*nJ
z;VYmT<Gl&W4SK4j)E#1}8-J^3udi>|==!y`9wu~gzLR;x*zTv+Hh4`dcgycauPfp)
z-?@{l7oX{<yJLT_a3y~4)z9<Rx>#{*`}vsrTIg4)9}E%WH87Lp{v@x8SC_1ELGVvI
zF&Go`fD@NSHEol4E1Q>pbA*jkXVAOCqr?K$5|NDwvV-AMwRQbY>nBO68|@^^Ao}+E
z)oaj`pfyeIMWl;nk+<k~tOWHtpcB|osg~>%IyK@VDU$H`NnUy36?xyFt&xnL{UX6+
zj>54IV%mypQ(`VFa5?3AWClM<Tga<q1Xo>SFSzJoO`74o#aK)q76e|Fm;H~Tf#&@w
zd-H?ns7k@S<9cG%(&*FgZncL@S$;=d{D#*Q#Z4Z@^n}-d*R8d-w>nim8t6PUDKB#$
z6~5b4!}_T2_6!zr<A<269cF<2k`(MSS@Xvm!(nSR`J7@rA2uN;@nntiAwCVOAk<0X
zTYnW5^|C6JC2vBP>#Px-Gh%mkHM=stAx?8)mWnH8W?T*aa8dh!``uTDqbRL7oT-2n
z7O3f#ZfD}VIrFMLPRCP@4Yz!d4VgF7iMYa?2lbnli?FtUNbQGl!JbWHP4=GF3`FGS
z6!6)}MsJ?i^I_up<8R_fTdG?K%!o^lJ!xpz?v~VhzbnAK%6lTH*cum?J@HA!Rg_#e
z$pxE5^P#>YzUja?<xth(c<kSMCmsfpc=NplpZghDWErga4lF}O`W?5xzgal=mQSZ3
zYGlVabKcQ%%m#LpU8w;iR6iBclPmKZlnS%O#E-HWKlpGpz8bfvCtL&e4uX}2W3-a3
z(pbHlt_L92&boqYYMQStKhqnXWF9=zsKGM2-)g+FRhxGw7po$?)@qU6xYl>;>&0oV
zU5r2liP4s<y?rriGO=nwM3J0b8;ICHX&khs;Hj~nYpmU)aac*LLGwpQ%Es?~P;B0l
z-<t?RmWZ#po#>~M@yL478xn#fl*prr(iy}k=%qdldkyIi1RNt|haEx%=nq{jJpU>K
zC($eK4wM&dp9m#HGzwuMb;uH_OzfZ0#h1z%xBraR?O#FKy+8hgT0}q#?aHOgB*iia
ztg1oJOKgqm*@oz=w)*9d8%unKYZeDo1w%DGCB*?BtZ_E&sF^zSa_79Hch9h^ic)7a
zqf*ZmrZ_DuND=eH7vZ^MFxv<-=%>)3wIZPiCSl^0SkuB#*DX<ZnNvbwAiiV@G(b6r
zss!bPiZm()TQw0li(L=ITAK^*K8k%A7h?Y2cH+xU%>^=2Bl10(lHVO2KT__>zgKv2
z!b9*}GmPK$+DD0=_b2LK`B^%Q#3$YHxZK3#p4VnyPip??UBdo?Okjim*(MjJp9`yL
z=3lub=89S6TIx4Vt$C076dn)+dnRopMdzdBZ0HcmwS){(TSOu<#wZnJ%o|<v`Z-o$
z(VFH}ch7o)9#g9lC2P_sqk_Qb$}j!_;q&C$e&I1CCQm(W;(rv~OzF>^Db3pkj$}!?
zFCTPl5)uG!$l67Ve1$G7s;oD%A8Fnm)JjZa&D<wKY|I0lrItQkLGJ;7;iWAd(=aS&
zgfkNkCJEMNw0O$Ndx$pgs~&nbrMJU`iz`Bl2SD2^xbah}2p9C%csw6JYQbX~XWa0%
zmM?x1B!m`?4xd0Cld6KI5krX&o#@pvFOw2|M<pvgn?(i1BYWyPrruV@t6B(b^*<Ns
zh1#7$pny#L?*9zpjVR4$*?Z({dg$7%EidyrOD*v09W$0?j@1LKi@ayGih-ECBJX|H
zYaJ%^mYFpi33K{tnrr(%SZOR5vb#DFo#R5?b<TVurGObCPr~yS3n;eO-!V7<dN9V7
zhxwXkZ}|yR8^Bqp_cpJWWI^=3eNA5ECHd!@d#~qkPau=7wB!rLRW%RwRn8R+rl~tn
z+w<~m6)o3E6DD^UCMrT<e@(pqox^i)>Fs6YfcklBJ<e?Re0lxYDEfiz+=uTA|Bw1W
z?LMB&B>1S%8yC2sBir!b`FReg0|~D7kbf{C+yhWUrMiWSdb6aIwo5;S6lmfvwozDT
zS6Ci;zZ&eO@(HQ?opO4`eKU@K#{zf^!DJ;%2%|6`cglgJH>7D`fkD!ZG5SGm4=ny`
zX}3cBs)jQLQ#6vQkka6h3A<=`4Ge~6OB0h~n@;jLwsdgO3TAWyZk3BgSBuJRND-Yq
z2s%#DCle*nGms6}KI}JvupktdH|cICVxso5&yURIQdm5L-e<WFqu%P$O(UH=2m1=~
zCx?QFWDeMgn|8eO!Q%x|1~#cbNPaf0U@TA>HlJWZE4eZh#4J_eTid_J5!cW7Ys1<F
z%~;bOyAjr?X8p0v=D4xnU#Yacv@-ej8m{VhMkmSALDG6I7ax;6-uFzfuFI~o5eB3C
zx!U}AkfYQ|yzr5T#E0y^nQ+_R6L2*qRk};ufM$y$bvoykg4l0)u^{kGAA7RBFpCBk
z{sP{--QR+*eAda(oZ((qP@ZV`BFi_kxxmGIjoJ2&@Mnqxbkx{@1rP9G{u)_rumA1Q
zNDbax+qK~}uVOq51|8Y*5LU=EDgE4yZ46w>0U+k~|0w9Oe_~G`AIk7bDZ<Nzn67Ss
z#bbvd8~cPu6`K=4f8_xD+kRmIS`e-rzZhS$*x7IC_|>hc-M{K)B>7NP$V-v&nyY>6
ztEB6<*Ba5JV5Z~08xCJN!YS9=1OFkzPgLr8aN~Y@R&d$e_-k}dMI~wZ#3Hg>`xs|a
zL?v??I6d0K+N@ZpI)|b9)*HP%TV%_2R$CIBEhcv5ulsWax;O}Nt`6_?+FRTf_}0NE
zvV>xSa;LVpVzZd0HDhq}*O}yB2W?>k0VCp<2MOeNMK30GuK~JFz4>X7|7lP=w(G=Z
zX(4EI!v=huAoIGZ*8Z-5bF@k77pSjV-V4i~POdpS6TE|^?J+#dCK>1bm6psQnPkm>
zq<9)~P0bKMj7HW2JY<J=iU71DerhR~)VW_m9eyrBvBtKUo1k2v&uaXn;D@B78sVvI
zHVvQ|FsvpNiLiCeGZVitOjsn%2Cepp4kF&iR|wN^y`$V&crS_IigpL!tey$DCoVcP
zr9Nbi@yu};)$@PwX@%?a;cOnNBbgQo<GnsLjCUe?{6Ld_j%^9%-R$%Apiy0@>>`g{
z5SFLkNbBopmKClW&iSNwPnzf8<?M7+@UX*|T^Bv=J*y5h75Rmf>TRB1d)AQz%OzOf
z-oqNOsA16~tR4f|pj8~2oKqy>R<Cddwft7E+y`dKOq%7kxVTez9)mv}f8!{f(iRee
z_2-Q^?_!|T5V!;gr7K6;c9xk*C;~@~Eo%5&U&2LGW_pn{jYAY>6*ImV=C@WA2ef1r
zHvH{k%qb>Cfs*6ORqy}gO~?J9om_u8q>(YoO0d<b*(;p612G*;nS3T&TG=Y6G)B|P
z3g^)5H8Mq4W)DA6H`v{VqbonZ>qa5GQ%uLNL_>?{@MU72OOAQB|FWp(qSarF=Huy&
zAK`g$gAZLC6bmLH3w5UipEO^DGVwj87xytfYG3kJKQwvTTlrtdx1Ptbh{&Jc02Bu2
zCELO8p(`H0g9Lj}kzbcOHA^IStRq4t_~W4{9sk&filwo%GeU+`PJJVl$(nO_39mW!
zVZkPxN^AaN?Ngb?<~!Qph^@m9mlsOhud!ZH_kur;MQRe_12SSttN1TP1~5V^iB0gR
zuM<j3rLBHD`wz>4y5&|(c?vO|f+nIgCkK*m!5yyqC;FxDwcOEKhl7Y#i-_Q)w>L9u
z3$LOq-BytavVxeqSc^EGBvuYFV;t*_p=%MV%7g%R&!>Q-VNAQ{hk|DClG^E7>PPvY
zoz_8uU5`{n_e<u#iIT!2%b@puSsFdlZhqR&r8?z>1lZgV<hL6E5K7|?P-Dro``*b&
zv-Z%mG&P_donJ-5%L||r_~L~(>euu@AoY!QHA89vNGOMbp^>kv<x>etABEhRy?wMi
ztg*Sz9rnwm7y0g5`umf(M(8(W^*Usl?I~Z38rD3#5QZwYLExBs@a@-VyIyp*8~Xc1
z*Q^@aB&+=5ll-ADEi4fF*IrE8T439DB=ri+<WYfHgo5!0i4b3q96$!8Z`aQ9m7tUh
z>`q?S*Lm39QqtLlmrV1P+Dy&X`4f|Zds$Uu4s0I_heXA9&qi&zf{!Gs^eMHWlY04|
z8Hu4nzw;q@PIO0MsYPd9sMLEyb9-(IY>-pr>WmA=z%|fIH3<AeRmcHK2eugHy)jBI
zT~d>=Q(S9^_8KWnTsk9MN#Mz9)rY-&L~efq^9?XE+01C=H~#vJ^+mgdWH2Gz8zowT
zu0}zT)aGLq>6A#~htA|Y<)@fQ*{3a`zm&<?+qO|oLDNq+%+<X3%=^P{X-8isanE0Q
z@L|Tt<&d%M>Xp;cg77{XPcahsEd!9G1<!bFnHdJ}Z4uksUTe<WnAYsW8w|MzG1}hm
zE1`#|x&4p~wiGgoSKWdozUmn{^H6WlpBj=a=1>S}u;lzR`Zr}h=LnNx<mL`H((^dX
zO1T5M$yYXH$!_?fOu=>}YwmWkR)otGXG#97W~e{N%Zjq#w?{`XY(+KTP8;#$MD1b4
za`_2BsJ?!y*EurJ<N9P~Rb*`xV~(FXbwLovneb&VVz5)^c~(E1oAu^bk-TRb$|M6D
z#lNMisnRT2<`>wD$UwfLW8WF;FyeLxL-I~H-$u(5RSh@9Vj}#N<d(d6@5X-=GsqC;
z|0pWj+!Xh$*nXvLFg74h#1FGBVsK?2b~1E{s$=_%f6kukA$i1;56^Xi!vZ14jm3VB
z^R(d#A_Xs8Z48sze+zWS^6Q$OIexG@>WsVAI=A%^Nx*H?u_619sIvEJBD*!zULU7f
z#r^i%7?BcI>r;-%jaQw5Cg-Bvyd{(a;Gic>gfu=Hj?!Jtn}GyXAD)QPuHUG>95mIO
zk+UyHvFS!hn$X}&0||SS=KVmQ;s+-d6Gh%iN!D(u<<A&Md$yplW9UtB`|fja>Zg}d
zI=<F7K1$sE(4Fn*L52F2&rrN!@)pMk+rlqZW<z770DcOVA=GK3rF+BTS4*bH`<@?z
z=?Crb)huP}EYgOVFAIBv4`0k*tfFGXKe{zpp0-n5(T~NSwiQL?wEv2g>qnx>$S;aX
zsP^;+Cj5a_94_S!1BxTrSdv*_qX-?N{pxBRz$Lx(u&}=W=4w{xGqftJkK;|QlZ40j
zo`LpFe$&tT=jMt#x&$sVh`IJM_cW%>GLHW#RQC;<{gzwzxL&;{w?TKl&Arr2w59Cs
zxXXE?>hoy2BW9LYg*PtqBo~4uK3?ST$5AVKSyld7j93<h<*Gqe)UZzQtoo575gk10
z)TM4pn^~KXb%XjMtD@Gve3X0O;SE=QeWh)V(k(rkhhsOcSs%0>2ak1oLQ=w*SAj3d
zI(ICoKeC4??O-y|Jo+O<W*?g>WO(wj&DS<Ja*0!C{LvV+>~mRu<QYR?*MAg(8dHYm
zJc-*i)<L45>Xq2u3ym})G_dC{3-#!oy`twIHt7DAX&9{P5d9}-Vv!AZa|_~4@um7>
zuV4}+4=mIlc~)a*Pu>&1JFZ%b%xh%esJIr}gGA@SHq}l`?_y{FM4pVdSLe<hw0(5N
zK4ghW1OM&gAN5zxfIG>3{=u5=F;SZ;?&encO)<bIXN*A<brR|BvAbWqT#kxL{#@c@
zzEk;1@Ysl;kv|rESEp#|3F>9^CdE$GID2KPkU@pJ+}uye3{_RNZll=G5^GSOA%i@w
ze+Zv4ti^Gws_<rUU&@qR_?9*G9Gt^tGruNntIU$OaT6S%yF=~tPhZK_t)X^)8^e6d
zswrphb8A@-cADp4nOM>nyJoQH2toy7cP}OeY>-(29()H^(fmn^T3!=H<+jSgPVb;l
zOD{ENqC>dPUj1=NZ#x^)JIdK3ABp<tASX7lQ==5uLh2gB&QE8}Sgy~FbW%szwEm79
zg%`t}HkwHmo^uiUQ%5M+?`Dt8H?CFvW7yr19Ub9$z5+vNd=sQ!qg|b~^&uITQ>5M1
z^ZXYmmD?;#r%^EPPXq?>A4TZ70lzd$Ue_crpcz);*<tWaE(aedwk$Qt4tX<IYBvM*
z2aj##D2m;rlsi74znVb|50jTGVBL43QohemnJ7LepTx6TjJ+?7(;1s<suK9n>19@v
z{JOi=wD@wRG|h4u{KxA152uqK_!aoiVK<HxmQ489Guf~g%ohh%BA;E)D{@XFL0$2<
zxzwoDxfi#^qSMEwX&Y3eC(lElUBB3xJ00;0zXfFu%Xcmn`Sl~g)Gz)Szkm$CVakZw
zgf;i*#0JsCrg5b4k@_f6X1olL)dD@ggOsxU<J&(UHaZN5E{0JoHJoBL^Q&qV9`nU!
z*VP(~N}PR_5~i`A5?6j{DXeA=9+W*m1;2K!Em?||)+`+sBdWG4^xznc1fd0yR*g=S
zSRO?|UBHqvh;W>wp)mKnRbT=S!|`S`=b#cq53=7OwZhkpA10+WYI>3cDqcLMzP}iG
zO1Lf^Ks+;dU5_j!%Gc}jUC(vc(u<fnTEYRd_0A(FXkVFuHeEIGto|(~D7m!L^;UL?
z0urc5sQP|a6xjI<TmE~2>hozL_`W**p8a{E>Mjob)I3hJkr8h_miA8`w6Sr4+LxMy
zy?bLYfhmS;7zISDL<^~-*wgSv(wr{<%(*vo3WmTZC`F3uI8hxGl0q-t@}iCAjer6X
zq+Aa;OJJ|ROo}qzDFN;wnqg0a7|r4qCu<rd?&}@Efe1immN}xVH=<DR6N<Ijpu)Cy
zHLF0}jbguF_m_W&$htw&17jYK)n{FM2Hm5wC3LgC36#b8^owIZ@42OA%@xj*v#N6(
zCe^8ayvr{dqDix9Odng{t-(P*yE#Wz>QFL7JEdt=)5vdLY&wC$$`s$dzEQlT#|pBH
z*RlD5d+zP==~oV=;qFXkk)(=7LV~L8I5L@rEfmQnhzB^J#iLwaMvqR|C?D%Ptrghj
z>t=+|`D~F!Dfykf=PD8DoRwZ8GHxTN<W`4o=JO$+*`}Ym3^2-gY1AyXQvh)Z@fR^w
zwnu)Kf)sjBjj-R7`d^y3jNP9ZqWOp>P75xT0AWqd_Fz|k8v`KOkk`4GJZNc0bS=Lp
ztx`Es{lrv_Yq=Cy>#q}`sj&z`{d@7pl~+;yy>APW8fA)o1OH<Bua-ExmOm=uGepx^
zXGtnavQf&3opRQ=7Oftj3;<0Df_;!Yhon}$*=TTW0<1}X`E61Or(wqbRpu^X(qCFw
zv>(`h7;;Dl&Ay6$`)y%)6m&)Oa)$dvX>7HEk3n}X2y9@Al?d_w+7VK3Ho_uJySHT?
zJzeYL8bz#p1Tpx=1QuZE#!ocitTk)ABTzAW#y?m~xPa<;CyZN#>vO?#6|{Tf2{ynA
zGc96c-WD!d<W?tn;s=xpyVfMCReo?hT7}|r{e)TsjH`~@#9sjg_B_<%@Aq7lETLWf
zDI~<0A!QBy>DcAl00Pvds>5C#N?w<ppJ;|Q&tzrsN{|A6Ni}heiSR`kRNlXPJJgag
z&ip)|aNE|Ms?^W1y{_z&_m%pQGP}xJql6t%P~Thu9$fT3__XGO!b7rLc)QmV5wsdh
znEYptHoJv+azn`B=tZ7HRs?|Z@#dE6S$jm-x)1jgBDC<wM`!whPMkc>3<EXlBJKf}
zN5-z*%dLmCj1~ZJH?oXp%cL>+auF}4-HyLZ!Z?ja)d_pcjtgPzE!B6`&oAbf^wy0R
zdW(2c(Lc0m7I}k9>aSSx+KgaRv2>osjC+vkR)|Ir`lk9fE#86`eNaqO5aYAq|HYdP
z%vB{{XrCl`n$b2I`AnS~Qo&98qA9kl6KA=Mxa$(+Oc-+~Qu&|9!pExQjcAIe>b;(^
zo2!zlys_Zvsa5=)fFn$oe_Ccx$>oElr<2ZCYa*5ymLI-0`zb4dg!76eh>+Flu{y9E
zySX}D4MqAN3?@rk=BI&S0ouHcrbpq<wh9haRd@WU-CdpXZURYAlFWPw>6p+A3f%r3
zo$ZJ<AHjB}e`JU=Fz_2<mSaq(BR&QqAh#qKdA+QrTkDr%=lBvEcpEn?gGSj00cyqc
z_Nx}xB-Q623?yZ-d|z=v!#^KWl&!g|cS~*Q#qw>wLO4;PW^ez_&2yPDnVnZ)ezCBI
z$bX4-GV`_VD=_!6SQ_c}QR&3pm^?COg|XS4noE~16VPpcjhx&&HBqRR<EyoK9{nUr
z{C-X8#0u_uV7=}^_{_MNDj)xy({Zgia7g8|+i~Adq3>3J6;j#;6A7)Q)6@M;6^6FA
z^_}-N8$PT`UrNFl$t$wY=B<yO6&1i*{rVAiqy_&<#^l#^<!2HAQBBQQs;4_VSGbc|
zVEs^T>HS58RLEaNaL?skCnt}lIabgG8WpzR>)5EOE519<hqVrj#du|A9$>OOR27*X
z={QNU4r}Iv)&Mfn;ZeWqGcF5-BOb!&%D=1$F@{p&%3Q{r&drOvmiO(yyZk<dx*5}1
zR&f4MC3i(1tXt~h0(%lfQl#eZo?d{hcjqj>gjLTEBgP8@c`p*qy>=#cnBod?)#?3b
zB+Zo*kGqF6%iACzR33Jex?+$g7)MeY7+KR}-5W0ot{SY;aK}3@lbhl%P|9yNj1HY#
zICl=-oj=cBQ|r-raXYYCa8^~D-w{8+T)t`}O|s;trQR=v@2RELR@wGUh#n}$A1;rS
z+hSc*pNl+7KEzZT0!@NaCs&`*NyD}fUx`JR-pDX1c@IO?X<;~@p(bm*g}2}il={3y
zxA+%{oS#qScDrA&yABXYdY9aye=&7EFR$<R03*vtDTjy2U0lmgQt`a2hhEoy0fYfx
z(D&<uM}4Ns1Fgllm#SoO^{c7`31@-G`T(Qq<g*5S_Szh{C%eus1{052E6d5cr*}E`
zx{Tb6tk@F+Y8*^<(b2VL@%vjJ&o$VRyHa~IaYP+~jKHpetpY#ZkGK{$Txu8q4QtO{
z8@2`vMN}dsmyl@=(w&N!R8*UGq?UjpDo8OdHNo$I?L?wkkidTvL|y!*-=y+&S;7>;
z2OGWNRs6(N=#Tmz8jnw<AlI0ta;~@}U&n=Ys|BegE%)ATrGw?u1sA8{5;|kw;%_5L
zImfnhFKykfGoDovo8q?zaNHLO&fi2k;B(Cs@DCfWqulv8y2=8NP3FA!68P@U{(Q-~
zTt?rqvQlI?&{-{Ifi+6NVm{<C$zIKgSMlrHPuMiHdzi8&*!`=R8>S_oUVtFTnjXra
ze+bjqlJsJ~GbCt{8X1jT^xUxewRMxzl9}TSWSlg0ew3lm6d?&eXk^*n*z&*RZ9pB;
z1V0eL%3EL&`}9J^c4F%NGS>O|(*yL4gO8GrD;fV7$JY_QMB=CfFG3N<F19r|_*GTQ
zQ47Y2Vg?*;a@8;>nm~kDkNJniU-uQfNP<CX=neg&boSI5MOU=$ypfAceP&*E7>972
zP+a|H<H%j2G$i;fr3L}(GXHSv)u)obK>t7~$+n-mR7KHL%-xqEGNp&cP6yTLiidV%
z0?JrjR`vt$^$Z~A4b}gni2RA;z41i3(=r)5x2QK)ddqqz=igi{$|M?*@%`sWNrf_i
zu#aLp4;b#~TYfZ=mT3K9IYCu<LJ}sD29OYAF_Son`;$_H{#PSnV;kX29*~8(Vj5Bh
zc^}n!9AaJ?ZTDJ@*F4F>g+lHA=FC27S9CJ6oR1FsK#Q_g@=nqNwMWb9G)1wnr!6Jh
zmwQFL<`m_wg&65x{np*XA}I@=1NCbQ&e1c#Iw{zGP!9_(k9T*vwK?0Y`Hh>u$x;D8
zsG-9ws7h@HYu#}fM`@1=DDF!k2r0%`l{?-Zw-);;lpZ|6uNtWq$!?wM<2Q8UMvs9R
zVliMp6~AJ6^LVJi*jL_U@e=sG#eHUb6M3#qeJ#P(Xq{|IzkHIGPBv9q{bIjdtkhg6
zpNjA=rUQhek|}Q+3?m3tIYbv1Ego(?yZJk&{`L~?{lUe|gdFgiTQ&Q}oF*UR;<R4<
zs_<)kiT7EWSZf#woLXx&;$0eU7WA{N_Lt;<PO^2w=(VJ@rDBw-OIb~$YUjItjTajd
z$A@R5_ufn+Jxk0faC94J(aSxok>QI4LYlY3R%HIwe;)s0#nfnX&{8NL=Z%Za15Nqx
z2YBk-izz}c2dnxH{^}75x@KqoN!y=3F*6s_eTE5>*)SqIJ<>^?6Xz$}Xz)2loI0Qt
z$PT_tN!d3vYiTB<4*)9kYy<un<Q(C1630U@pJ%{x2MO9;RUj_xFRvQM^~kQTl5y0x
zPcLyVX09fBUgBula=UcYsxhER{iF8hDk-*t;_>EEBOLq>E>{3w)CKvV0M0RgAZgdN
zDp7kS9zByvyM2V7DBSp3zmKWEdb?-ma)eljeYE-3Ihu1qo}=>C;z2P-jgO?Q`@Z?x
z&U&1bxZIpXf>K1u#AH~Jn*pHY72N0$EQnPa>RmUJ{Nwv%vU5r87xlWy2GBZHHjlV{
zSW1De{&u-#Z~i4x+T(lhH6TkY{wBi11jq0dESn-WmvY0UctVjNDOOqIF@@#h)x?Us
zSkl$skIeDdO@}&HZ94S*&W;eX<5-sD&r<T3z&dSn@LP*VzLe9uwmPSU6T%mlX(f<e
z<E${citzO9ps#0tK*MX=hK%7@39X<=qKs3k_rvmY6+VeiUf;mBx;DbMJUZkYYBvnu
zNNmI&41;oj()!8kIm~K#a$>r%BWyAbk=Y;m*?rxLWn%eGjSvj=AVKmDUH#h#nVSr-
z^yY>?h(uzU*<v1qV!`O{W`iIZmK#T`IEQM%D4Ate&4Zc#qlnArMlrKlY>jEfhSlAz
zlFh^{3iF*c7r^~q@EvB#5jehsXwMqqCN2N+3@O3~VQ`OL+o^NedoChQrQ&VSnT8qm
z`*n?UZ>uLMc{N+clLEV|8J6k>B%3~GIX!mSEag*7lmBFYJWD^h$qmnm1@KV|3|BcY
z%^a$ngl(lQ^WPTSqEo-NG5zd+W#{TI3=IPEOJ~0+%gwxJzv!Z{m7KJ4;AZ$KOW}|!
z&!wg6m8e!BWi&QgmD1CsrwMHvI&X8xa3F=w#XG9xcQ1iI&99psdD1jaF-y9syj-Zh
z%?ik`q4pMiUM54!<$qS>!qwROfYYWDY%x$n9J4^(x({$y^M|%l9Xhm_UqTL+&4ig{
z=c9cNv0Qa*MoY5mT!+=Qj5YLwHn79E2x}&y(nCuX+XAObn}#g^VHw^SFEZ0HpYI&l
zp6D%docJ!3YW#;wjc+MI5e^iYj^=qIssh-zMLz#KvTpqmWs(M~YL&T*Kz3M9#`h%D
zj&rh|Bw~+P&-pLH1Hm%^ABhdVcN-hL`w)idMMG!fqeQ1=(@xt$^Ot$=wxgeT2xddh
zX+bpYxa34zrP?_zTvb)vQANM}#{76DO(5l&SupvQlAy)(_2a8H|41|%vGa!7PkzW;
zvMn##1UC_ujq>_V3NIaV<a&?>kzD^#(A>SqJLjHbW|%ye+MX=MmY7*%7fI(XDYN#p
zi$<9iOK&=zO_mi$>NN)7_*9oord1{9U5$n1@pa3ApF3!K8W(7flof*jS@+-a5<PFD
zfe))U+C-;52Nfi@V`=DUr(tWP2P|cbY~2a`-l=cFE+qm|k<vzZX4A0IvCVuY4zCyI
z*vMy(e^;}A7hZGQS-x7JQW(nqtZHRdD$X2o0xP9Psh`f}b(Qd){mmBgm{gsb8Kp|L
zG|u^I>6>FJbj3@IAxqP<dKj7buK17}pIy;jtf6$NVpW~!;4zp3%3wG18o`*CujJ8h
zJ&&I}_BlKH^|VWJ-I}-l`juB69lP=l4bmEQEb7=Hx&vP3HL7>Ut{t~PqdKWQkIGt$
zHN3B1Rkc_2GMD;~;{D?Z)A4&bYxH7YT6#HGofp`5A6oDMQ<C1wTWqmNva0DAXo5CJ
za=<S&P9(HX5e}no&#=|^RI`L|dy+VB{CVINrUEH_?%N`l+<yHCu%YrKtu@h(PwS1}
zeeGExGZ3`i-XOvdPuiGxG44h)R(<r%S?1cWxZf4YsCJ>vpx)WZ!D`^3iwN~U!d?e|
zKDI?kF2!9vV%!zcgZ;}-XpjYq{|Rq!?OJ$^gKC6s>EHpISs!u+XZP-7ZI&;hz3zMF
zfvXce#KWz=IN!X+NFYzlx1Y}<37h}Z^pq|Z?!mbbsdMeO%kW{2E8V{2+cUlbTmO9D
zlHko>Zsu#xeRln@@<DI(c93(7v(^zA7ueYEWA&fnUc^EFqkv&vVq7i-_wtFcQww7|
z>#qO^4O%;K#Flt<$|2+8Cw|~`ulV5eWJX7UmAR^=2gK9rlzSf5Qa=_lF;B$j01gq5
zdR=()3t+$V`OQWCG&#rnKB?b}9x*s7iIA;<aHK`~*Eb4U)2^C%P?5)u|Jt5)vN3|w
z;w`WKgZV8`dcrBzhRakj>(e){7+(3{vHsk$=Fi#^fiLP9eamp?X`D>pTRgXIN7S{!
z6pGVvYE0lOA^el<n}3iUka<0?+)N$_vo4|G-M8v1VfaS1&h}}Tl8*nU<*%h8lXqu$
zYIFMsp$p@2Cbt*MH-~%YGM0RNIS<;$$;wzq?bhQjAxAL=TG8HJ#gFQE3L91jj7XXJ
z0;-%Uyx%`N!$bXxJT!;Vyl&>&0ppWdnX$3h=Nzan1fRbj|Hn;7QtiX2_PP#HUXXS-
zVI@jQ-TB3JV{LWu==o;z;Mp2UTKV#Ly(xuen}YkF#WuGf3nz#jJtwK?F<KFTETT<A
zGEgVmXa=c_WACy{DCHK#sIMBR=3><1Y%XyWI$WF~M?od=W2q=A-X~+MH3?~7yAIJ$
zT1+06ZwX7g%Nc*LDik56aT&qh&Cg3vK7sixVO5KIi}{Qe2`QJGk12<7>ditAho*fa
zrS-2Q-g%E=)wT!+=oc^7<FL@P_4sF0XTwNA1!*jWqN(HqK4#s=Ms=a=X-jkpEe61d
z$or&*_1XbC{?rFG)FI`v0X0&J2z_Ev;v7J{Os_xu_H=$119mPYUvGa;WTm|QYk_dz
z&T#zU9{+?Mbs?DD616N@5F(q@w3cY5mt-m0|BwaTeRn+^smk>Opv{PcXu8}Flav%Z
zCke`=h}#xu++w5j6vFZj2)_G%0Elg`jrT=6Hky52w8}t*4pqja>fZWb;*v*QZ1cP1
zSwANpx>Q*Ct&Ch~=hQ~`nV%i{PfxEvu?g~WotSW)A?kx}d@zKA160hyKJlkrO5^k2
zo|BNLmxA9)53excuB26mKGh4<zY1C2QbdbE!R9|SAz%0k9Q$gcE>K<gB)c~TA?$iQ
z9r^Jv@!h?nG;(%sn;^H}8{`l!zG=&iSn1+Gjhd?xN&&lV8EeO!@upJ6QeJMrVRXTr
zEP_R#cn7k>P&1f$cFUqJlfCm|GjrOU3><dSmoS}K;37%(*;PY-viUs6DTJYj-`;Bj
z5RqEHhGKfmOiziHd@kGFcg8-Hak0y^+k9dYlR-Vbx#SJ<xz6Pv)BjZb;DM689kHq4
zNIVw)Sk*j{vdOG)XVpl$cM#z=Yq&K;=SkG*bxn*UHt;#j{=&$5_Q!W&%nI8B1Ahg3
z5BUX8^-E|{JZgM9*)gKt3}&77_UH{#452%+0fbT%3ITK1RF*PkG7!5r>OYpRRyw=o
zOq$iI%2&P{O;4KpIO066TRGLc#9a@}*iQ)-=x~YDOJ2l&97`bA&Sf|}t@(;y<dxl{
zWI5vnfur9hscLYsZx2+zHu4(^lsqbrDtb^y-*Kjx!WU=0NQlx&UaG;Z=f}LV2vybx
zr~fp`bcv*w&gLlX((rxibSP5C{LOK@WSq_?Opq|}*`Gnn^Dyz+=be8tK;eSvT$XSS
zq4_=mON*=09C|pUYe&J|z|=rQ8CdE18>X$id3k>K0E8>ZnM~r%@VVr^$WpqyhWwrk
zpSeJYh#cuI>-@MzE?Qg#I;{Uu*eORIC2E(<#F(*34UKf%2rev|Xz}K3g(i2+1@c{2
zReenE<aWxu`{uz^+H>aslab4t7h#tVpRjLv*u&rLhG5K!v{g<-vD`+N*L!z(PI^_m
zas#IW`tTm-3AosY2{EOP2ZBe9agpIHd{=3&R-FrDB>DA~XCV7c<p30S`nb`FzoOMG
zALjHIy9J;!5(VRkwbW;K1Fg2PyZs?EF4`}Y7YoW`XF?XFV0>Q+$#co<*d`VU)5ya4
z_}C&^>U9gbTL6omGf=bE?O63(HmG<qK1)f0{?^4|o5;HL!8Yxo4RYz7WXSCaEjP{}
zk`zSH;sU619IS8JFgP6|UB3JH#6Q3_z(2;`0IAK$T!61TT<`y#S)`+Y(hn;!*;MAd
zyc;OAqxbv5$yq@!*K*}ZuP@l0%kK7I&rFhXhI{72sJ;rJj;Gg`iOsG`;J|xBO+tQ}
z%df_r=WD}Wy<tlW_PY~J8x35Q4$C^4Bgrg{+tOdXW0;PvuIu0vh^oeu%}z7(L$PW&
zA0y0K$DmgNme(GPjguR(-CcnE1J~8X+dQWz9vFZ}RJ}&yy-O>c#PdJYhvBhI{&&_y
z`sBl*M8&9hZOHE7e7aFyPazkUCeKErZUQfO>tdp9JnKT-Gd(g{ei6sv9KK4fT(W)N
z*5y-FV@_t9!i2Rr`ty4RZ*37bKVzQfPd^y{J8sk*y&1#Ldq1lP(NQ@<UJ~b%j(3}I
z=o%cU*P}kxGunY>CnG><asq8No(~~PMd#q5qR*p-i3|2F7H!Do=q)OkiP%7-ROh>o
z|9CfxSFMg(Clz1*LR@CeY4yha>yoH7T!~P)Q6$e?3a@J1wuxxEavX#c^7KwLP6^Dr
zTc34kf(z{3;hJkaiV~iZ-gHj<GHbLuKRq^XWvQZ|C=40vj3*@h-B7L)*$A3bjvF^L
z#eCM<o`@s;`8~aEaW?Y5O~Og5m0`9h<eR(bICY~#2>kKM=fxq2<G2?SzsK=-uV!w-
z(0YvYtrFcOjW=<N{jW)V8>0u<j<JMUkMtn8@ZKA%9+^@nbE4cFPV_M1&LzaG+0Pv|
z)_NzF6O{yDlA&%a<%>*=7IB5JvK<FZ2Vk6L)6k(aPj41mN=H|RasnjwEWBI4SjHlf
z0#PZ+QeN0GRzCWNmWD(wut{-9X)u~U!=osrqC+%OO@&l%DNFAdW<I9+&1n#>Uur_m
zSDlx*LHJaN+B_`z!KiJgSPW|4?J1>XJ6_-`5O)#04AH!r)E}dOi3yLliVc51Tsrhu
ziO^!+1*yyg=M1}<OHYX-M_4QQ56T#=yE&kw2p-(TDl$#GUj5E~*^?nRCil+W(seaF
zb?<U@kaEWR4wK>A>RS@4EOrbFCCa2#bCZ$lPxT|Z&d9*iTykD=q7zuPfbOQu>(b6!
z38W-Tke&-o7gOOJ2tE~JKOarh%GyP1TbV3y!!BZeS*3leCZ`nSs1c{B&Y!tf73*Kj
z;O~xCG`U|>gfF^<%Z-{%t8kW60z2E&IIUDytbW9m@%XQC;5tP))}=2oZguEx+YJjI
z)BOT_#tHnucdrqr`H>F}H;X~lYl?!H4<=ak;nYJ!o>#IZ^m_#pK~q~<)7b&ESR0)5
z!P{-`HA4`*Jac;0%$G9+`S-4|x-bBl#hJi?k~}J*AE-Pg^Nr;81@<jNZ&`;#6vjT(
zEyN**As+223jxsNpOM4Mu7YFUoSm^%Gc9t2>ucJ1E>>Zc=N&-e;kb;TwD;klPy@##
z1-#Yn#(3&@9&w)QcX&M%HJ$hzlR$DGc|F|s?Zh(kbFNbPou?O@@SDav_wOez0TPEN
zV@>?`rCB{Jb+J2qeMw1n13_R8qJf%@f<*qIqL=Z`h(*v}ic)dY<1ZGzTjH6&8&F5>
zM$;Pg{Er%q;f0AmWMO-bUtLxZW7`_8-gQ5sN<}Q*g(O#;&o}${@*c+6MP-zHqfpw9
z3A{zbJIZQ#jmAjom~k`QWHukJuA?6t7hzn{*YyLz=v&7X5JYx4*@Qo<!4v&QF)D`u
zki}kTX>)%IS1MkYX!Suue9x$QdbZF@&jy+F6L<7znIGP6T5XGG=?H>>tO!pOHt5HU
zsNX)km{uTw3!eNz-&mZV-z|1H$yXx0YRlW*?2_df^F)8*%@Ae@-gM&BX++U+JR-G=
zOB|k{s^_0EkwuaAyseS65|sQw;(D=44F~Hw!0pSpeCW#X&+Ns3QJ;-<{eE|l=Q}Fg
zK?{8-T;4;`zvMw$-$8NlZ4^L#71W%|j$1<O$5dUb<cz_W*tWmG)|8hRZ7zlKG(Wyf
z;*x>%OM46ijEPD2k4I9NnQ9Sp^&uXy%8e5lb$L@Hd4xVymv0bR;Bh|b8NH6}YCBhl
zxHce3)~C`h@_HK?sL4_J?3U1)-_X2#jM$yW1A^}@g_(izS^XQTdP?9W{{5PBFUugT
zyV53J^8l1rtZR$=OU4Wwi8{DolP=&&4nk0eL~uI6q{Jai7@Ck#m!?kb#29aMANhKz
z5Hbmbh1$eU>1uh&{YD1WrDZ^IKNNS<6{>`D5Sg8lJB3^`#drtHGAEb(@!c}EbNWR#
zFR{L?x~QL<Qo1mzN&OppnNe=S{mZWNIkBpaBKGCr?R`nZji|six-3`)rZeI;$l9up
z^i**L_X{oWU))}Y>`7aULyjhH&Q)&y>>t{|^n690mBb`YV6;9J(}SFilXd9Ak75P?
zU2(p|X&6hB(c8(BHL$r<G3kG{kIO~RP5z@Ovl#bC#WZCF05TrS%M8BCV(_oh_}a?%
z+~9YO*JKytd=yuv>OC7m8dZL$`0UE;=kGB@g`uI;7E8;=a2}QGh=^)AVP;U8d%Y`5
zy>Vk`rHj<kfQQM(v_kkEl<S^NorlT$qmZZUGDZfshUlgLnJa-6&62##8TX`d#)9l}
z)igeaW0;(`M@1WjeJ{)xGJ{}TKW`&Gev}&W_y_xsq7`T!)pfRQ-JD!O@G~m*b(Xpp
z%UAOJPMh7Lhh-{^WA`(71}xT32D{>8J=XNz*c63SD*u~*vu)cd$}t}s<hkf;WhJ&7
z#D<`>wg%t)2g$45vi~E3(aQ{>JJ$*M8*-?#?buRp;i9>&RYN@2s9YWsdCLr(XI7Eb
z&v-ut{m5OvZJKu*((x4aR4MBcwTqRSQ8i`&@!-Z*5jim?@}A(1B6E|5)C0-lmX0B)
zT2UuPOzo>?61c0=03vy<Dn@IuVdUQywYOOg^g_q08u4o~qBZ?!AMN6I2hH%>BpX%#
z&M%G`BxInuJhRoR)%~%ChjP1bx5gw<GxRde>2Ug5FsEtLx*N5h=he$yCwJzHWGsD5
zi>^iFr&2Syk;A^;wIPQpC7$iZbR}as_B6JTy?z%y`jfYqmmY<t9&md%*IH7w3_;Z+
zI+}@@8=hF##Fxj}lcw)|c|SNwX~Ig!_EVH1lOC(ejh=6m%XAdW>*O@}oE_e|j55}}
z=o~(fd}wl_!Erhk-}RQxAaV{psXn?kqz0t0okeK<m9^=T6`@l57vTQ4WA>XJ*w*5%
zM$^&5z5!EXgy*!#ZD+)u)<rhuy%f7($g;t@;eQl<!$2SvY#*!RbP%4k<||**6anoO
zbR6fn^p?c6aeaVHW@+xR6CEJ~Yq8U9(gNu{<SC*@JIp)iW$L>@+sZoDXzF+yy8kFz
z)RY~Mb-x8aSoz6@H&K`Ln_n?b=l^p+OL1mGv}PaA)DsC<Ce{HmXgA%R!B5`Yw~Rgj
znHE*rnoQ|U%+nB_gVk9NX_#7gOiP_kOIE1Ea~L^XW_g?&jEbD)AHDJVrhcI1cFf=h
z1%J%3>uk+n0a?qzGWym;+tmwY0|<6=gj-RSf)5u&rPIiq_TB&e+iJd#wh>A=s6~3U
zN#d!oxTQo|lzSrz6l+$2l8XFS@VvTI@w3#G6Xk|Wh{HgC)GKbSzGar*p$P!J5&TOo
zE(&*hX4K{QQa&*wZ@;WK@aDEns}23W^mv1oFC!{<Nh__aq4VOC<QfW=iRtgz?Ao1<
zP_7)PoR`IJ;&AVPwu*)8XP)Jvb?cZN>bcHO!IUe#c`NMOu)4X<N|yPa&!jI3m$;5|
z^^3y79Z<uB+Q)1;Xw)T!BtWInx!fMOQ>M<X;Q&k!GA-6Ey>}E)!@NdW#~<eV;hmi4
zb%j`;qr~ntb^3j78jn%EC9tb=;v^2f#&Y7m;9crtR$cuRybRMd^i7l(cmV&_FD}v}
zj_kwZgYXo*N2v1=R~tm0V8p8BuJfwO1AJm7<jg60cta6Z@e6JE>bK+o@K4tK%gM%o
zxLE=f7j5L|neJk{Azr^1n1Cj9fod4W;x{LepMoI($3u0>kRndrLoX&0h<mW;JfKj@
zTB3{drkjj?Ol54pFTqSYSYFjJzMo|3yA=AMh|h>^vp_TZ0krjPQf*-fGl2ZizJBkh
z<U7qtz!=@6s#8z<4|15DW{hDW|BC2AF>B!&egk=ps|vVXpx~y5z>if_LB!_1syaYW
zr6#cM%TQJQbDSI^y6XNTbg_y427rtKG@CvM?xX(nakqjs+QnrV<@Q0)BGN9?G)^c?
zZagSQGw!O!Qr*2gv4q-{375Ie=#2|_Njcu)3y1^H+fK&2D82j0Cp6#=&l%n1?VfsY
z669eesA{PCX#`{EGLzL%RPfSC)+`k2%o(v!m|8QRrRg2JnE51ndE{FiPs<EfN0o&S
z?Q+~m(~oW5C>lb=Rr{_r#=xT4$#}Or_PXE^g<EruHI3IB7)kzfxm9fQo~vUXK1%7A
zqn6YSUfOcM%e*Wl3!9hC^Im5mB3#Vke_AMOZd)e!D1JesGy0HqA&8RxlB<`ZmbP$v
z6o+K%XrrDSI4KX6aO}t-Nw=mx-EJ8G7h@lI7pETr)3y5S;N##8PH>75!+HILF%0iI
znV+MUB_2&Z=@+p38bFX023c^gMc2%pzO5`339Fv_!#0YK+SL|Rt`@-pi$9gyz3S|h
zr41@<itY0G^5Xf(XL1mMGFz2y*n->n>#ylrBeg<CotcI+>dZsK0>wEUX^y#8$4M@K
zW{*JwFtUVBTJke+&p{avJE?px^Vc|6vETTBJoJUy(D)N{>1XGsBZ-P`hR4yIQl$IQ
z<IJUgyQ*iH{(~~*EX&N<T2Hl+9H)l%-UJnop33B}nHdUCY-?*g=IGO6V@{CH;;wHJ
z@yO2DH%6^qi{*Tqhmzd_Qa&9lv7OHkLo@sjH+12@nt%g$heh7hQl3S!^-1xQ$url*
z7FWdp!v|y&e&S<1DSR(T-gtO$u%-xM+t`0xO8wX}-4~sF6b!ufV9RgPiyuk8ac#T~
zlB#{A`lEFW$+?Z1k6g~H2M?oO)!B}XEIx}1pPYy|0f<XJqX%VGN)&H_fpn$ImudHP
zNJC`}=q-=xJ8ISOMa!&4eartSFkHHduy-MZOz}9)$T@(&Z`YRXSEylFk$hNe1sv%T
z^=o>!^!8{(W!In4IuA7%RiERh1VCE)SGo;RNgTxqMQ%y{=POhKh1Y5&6FtPm-TjKJ
zELo@SM56MKmcl-fv^4Q<=E%nv{@E{ctQuiGrG$P-+exU*WaJh3-2#0`?W#sy>QYSs
zg_N11AX<2-!|7i;iU+XdWM5a4m@xfgq|ow=sn`>y_tf}3MRRMUO3#pN0ibOSWi6;)
zaL6G+H;k{XQ1!ul<}fe*%A8G`e-a#HId7@wq!=F=_qr1+fb3;=7C4tDq|tC`F;M{v
zyOZmQ{EMhA%|)qay|D+>UWS#R;;$36#ZO=vV@8m$(ZeCCGLT^T8)USXu1%81Gj_I3
zK#u+A`D6r2^n!4E6-rdS6d6HMd0Ukeit`yC`cE3)OI&P)bdC2O9f4HskCl?_HiN@Z
zf4gHFE(G6Q?(jJ3UKF#uR<|1rL>Nmmj0ht{u!1$cH8iH);E}Y;a+xauJW0iD3uk{Y
z>tSlvV5qjs{E`wO>wV4-p?15g_U$jG#kOTj(#S|WEE%RA=}X~vN?3Pd2non4cor(s
zA2V}M5+CdgB)$W*D1SseKnHX8yDk7clK_p$n+!qoR%{HxJ*~BtZ>yjy&p(J&YIV(}
zV5>12G3V^cnqPmwUlT*cRR0HtM!(NV5*=^U?NmfvR|*6szRA>q%}=SWEfcXXuZxNl
z7rV^$+48m$aT^44rHGHAMT<AY{7_{alhKichB%f4HR_RyG#__ArBt@3oI_PG%+%Ej
z!SRSdUi9&vY0Yqzg+oJKYA8hCopMuG1~e8Zdu&k_XiU7RH;grlJIB?@{=I!*8=G``
zb6ac1`op&G%UJMZ;<9n-^V+6wfqkT3il6-b(hZC9eH8=QsUQ;>I`<ue2b3!v;<weF
zMf%TMOy96%n6i>ND;oBjF~Ny!@d|zv_FsMk@+6Lto*FOv?s%6j1LsHPjm=PaRQAWy
zyJeY+Noqbphztb<m!Vmgjpcb*I?q9llf({i(|LYS_O{~KtMKGn<np=!`CLSwzV)+p
z@8B_XomSrZk@R5_|B0PD@_kxmGB(OAO89!o!gi~v)W|3GRb&Cb3$HGKt<g76)y#*<
zw-I9B)Nr7VZUzu!mSKp*M@A<E*SEwv7E&M?XVlGW-f+!bV?*0Cb=2Kj5Ca+3>tXn7
z5Si60?_o4Kc@h&UdVu~paT+-yy2ND9ZV%vqAm4eV1egD+70@#g2Lv0FIyYF%CE2>~
zO9tKJD%XxT3H}#|W6B6m?}|6GUCK56O{{BK?heb3bUXANtl?1Z$$3UF_TbK&(qM>J
zy|v_~J8gcT$mDfsY}D_M%S5^?P1i4NS%>G^zcE;~Fo=qdWq6;Yo)w=}qm<|ZKU4Q>
zQn%A!IMRCjHRMyYXfkQVmt8r=k4vz)&#wM#zs&0bm**4$)BD2my9j3WTkDYa-_HbP
zh4C~$<=M@Gxykm}p1-l9)mmXpLm`wl`EQEP;@v8guHs%TjmV<8g~}WZTmJrG{8`-X
zd<X&tdB9Ti;$@qmif{@2bEjx*1^(nSnuoBKdHwk1>|g{!<Z$q-YfiO#sBc}}OoDP9
zxU)$-nD|3(Uwnm!fid6diod6236;Ovc+IkvIGPUJ*VCWbSsJLm?fC0L67zo?on=6i
z@7sn^2_+4Nq$nUAlG0$%9UC=38a5c+sr&#XrD2pb3`UO*k&e+D-6g_6y7Arr{r-G-
zK0VKUU*~Zimx%oe#<M$)jL}C>acRT5GOe98KmmLAT1#{v<#Wz>x6jwD7&_}mx_MIe
zjlIF0t$cXf-5+ie#y~G%KG+(WylJ3wG!v@iG;USJ%<tQIrSG?~+%htk(xc1yN5)J}
zOh~YSX}-{X=#xoiIP$0y?It{MWY^|-^=nJ0t;cm*RmboJrc-z<{A(p!_9@UBFTzx5
zGS4mQJFf<#vz~m2Rd_!PaD}J;X^oEzeJ?1eS(k0=D<k8Opi4Z6!^x9@tVV=010g77
z<Pn}EA*hcz+Dc?k5E5HU{q%#d5I8qmv|qgc<B@cyI{2-t0<ZbozmWunGgCu7N(}sa
zmO0Jv`}n|;TW0Fdj~5bPJ$wf!R;;*;2&_$5b~!b4j9Z+aDC?aMI@r}x6roE2)VDue
zE1r|#ewZFG@=ny5b>d3_cUQ|qdDo-G@A8P@KA3_@6n7I^7msZ{-FH^J<tZND&+?(`
zad*iDjVSV%Yc;Zkgk)mBs-IA>Yf1b?_by|b7nz4`JEi{-{QrfMChiT0=+VahHnzoy
z<}hRHj{dizkm7l!HnG#J$Td&x7EC8Z(iVaCVf&Bxm9<_RBvmcOoZ;G|M;m7~ej?Qk
zU!;6aQi^P0=yG-|FL2!fn<cdCR;DJ9{5?ZnvmQM)v<CXp@rlZ8|MEt=S|U_F@mQ?d
zIc_xtM?8hq(Mt5Ay_xOK(E+zUN5kj#EuFMTGJ%UJ3g@PCclbgA0p+kO(Asc)<K%R$
zDZV<LQ7Y)7M;QTdpWab#oca`X(G>_~IquOh&ZPrKTGfido%vkARK0l_+6rJgA{WWN
z*u10I*QQ;|fPM(`6Ur6BWPTO;BQd|^>2FpIzoA)}RtGv_q1_%|{_4X=1(r|cT@S#T
zrH7>N%Snd&F2a;hYF5hQPJ8~8h3n3It{txDqJw|TA5-0&GH3b9^hNT+Bq)2zBr5q~
zvX{5e6kyhg<a+qN_f=dRc)z<Xa5tmv(6c-2%TupAREP!Nc@jL`slMJX^gQ_LGzlT*
z`Tjo^UU*nf-|N8!)axcs$11CT1ouTQ)?OqF<(jK*w@9%h%s@0YD-<2>ERJ?!bDtCE
z9mg){xP8M~B4aQj-i&tt2$*pCUIFc=Y>;_Z4W9pyV`H(F9X{JTTE`T2`;Z)T=@SbW
zow>6+w%DjXb9<{C*C#PTAxW3o{-@0QNhMSNQJ;*YUO_Wy6J0sVsoD;n&g?C*(Eg>U
z%jTkBL5Knz>uMx8uo=^rh|?AATfAeu8n<Ak{YNk)6v%Zke-4pbbKZx0r(w=Weuf7k
zlnuGpMHb>zS)&EdPJd8Hagpwu6-;^a#r-WrT^cM!6@j*yMRf!@HC{3a0}Cwl)Qb@P
z%l@G~Dj(?TCwN2n9T>KXf+;!gYjP{k=&Kzw>P>{O8?1|q8XsN!BPhdPep8=0eSLBx
z!y`D;!GV0hfR=Dss*5_-AMcX~v&Jm|)26IM8C|9ZTMVcPR?&FT#K)vlDf`ETU#2uz
z`FJ=XF&Ul&Z$)3M*5WsRX4GjDMs`9Z#UjloLYOvlbkvJsBIqfC&9(5$RO(PTC&3Zg
zBw|xt|M5*9ad~%G$^GtNA-uu)=1&Nab1L7xA1Ot9cfiE;LzHSCXB0tzr5Mf;(~?u2
znM&)Toa=Iirwh`(26;)fAyjCmG|(y^b6!grihiimj}#kwFuA{#qtm<m_CsXQ9|HK2
zb1Iwu{I?ynK{)$z^ivsx{XEpd?|M?+ESfjv`WwO2ca@{Vb4WO@td5bi0pIXLuq%e)
zM@RUm^lQPQQ2mHhEGC0II#S;SUl20tXCt5lG;bpbX`cD}hEN7)J>@~TaDD$$u3{-8
ztjjO<-5gK?u#K+SVCqoKb!msVca{zIB}?aQ>+^@+YsIfG4;gk&uaYR@oY%6Es>u<!
zo*GPnXgG7l%KpI2T&l;S;4MJc5$OBzFU4rIj*yrmKzm<?q0j@XmUf<<JUUk%%o3M&
z4I0T#cl6DPfHXRnW(e~?`EQZ5;aZfu7#-)jYv7!u<nqo^8PRThf;0ZYB`j1g#4UZT
zf7FYZEke+M*O`q!f3rnz)=x;BBK>J$?CHAV*&wT5y#+ZHJOY|A*4&{fqE}YekMD-(
zG&5GyZj9C#3^!Y|gk;B>^v!dbAT}IKp~(2l)e`%w<h*!K3wKzwydE#FkQOZYk2&E#
z0voEwH!42Pr^*H69&*C$4$q$AxnyhR-N@V5huaZ1JeBIzCJr>mT@QG^G2J7K9E97i
z$FP0_Je5m+UIkcN7fet~$dHYD5TfCZHD`_g>mA}@9_~j@Zx&<!T~Q9J>f>vIiGhl)
zL<{YAiZr;Vh!a%H2y0GwfH3@H;wzq&IY}9Q%zCsXy~Do)<Xi>%6<4w`za?dhMO-Vk
z79%?+PnK)vOY|8O%p0s2{6T-Wz6m5aWT)r;L8HjTpf%M~(mk6Ol^kQ+2^s~AoE2-o
z{^qTGHmiSutg{P??kc*S%%2juK+(GB6W^WjjGit&o>$50;PdwKi){W>RdQ;g-x9%g
zA6iomP{`dJ&n(qSQA$*y1NF+$hiZGFfUrpaSj^ZTFFt3F=R>mQ=OHT~%|aMIui3MU
z>6}4-!GyL!uj5BaMAGNTukuH((eQN<XcuD~WBNidEz@0c&D?Y%^B~4$#yu=M;vGe!
zV*d_^@$*FEyc{Fz!fjd=Ri+z{s)qf8FWt<A9tvglcHFwUQwww4d`p;|+aF$E9!i{-
zcK;cCC<Cd87_$6%$+F;g4Yb#(q~$didU48uW_2`!Kx1r|-$e&xJ9P!lB$1*kt^w58
zdARG28kF+fDdzx4x1SqxNVm{e!6VC1wrFua(xY^)q&izWY5CtiHyOxj`P1g6+4WRy
z@z0}rYXX0~V_$2qf`?_gV$a<qg`3BoPfND7qCCj!Zrr@~b~zI7;i&{H4<D&mcwx^u
zBLy27d@$eSfTC*M=8vW6{&J|$q(l)HBR|Tji3_5r%INpxR(_ZRM1x6g<Rjiqq#57V
zs>ZHC)9$WqmB9WksJ~9P-IE^=q+@t1&$HSOZ=kr`D9C>-EzESfWOFtD2zs<4;oIFA
zTg@1lVToLG!-ubH%$lcL*u!reT^@Y|(Z?pw?yeaB5l9x=k=D&t%IYO$&4{0goLtB8
zYYD(7($&<dO3;>NU-$EFp-vfs`Bdf;2sR-qUfwjlMt?oeN910^ROlSzy6@qzcfhk<
zebk3$;xoi6MyI`usK_nU!?S~R(H8fV_y~8y)SCDXmkKUNugTw^xN(l|x(Fh3zl<T%
zHaCd$G=r4o$5X;`n&RLVI5PUQKu72JeLdGmQLG@GKzlxe`qR}EWX2Jmt}V0`0x(S}
zj>J=`5iaYob&B~*A$&YM{H|*CxkmRZ-atGoST<Zwe>{+#@(uAQ=Auy9r1~IF#=JBF
zQMK#^xX<OxOZ*^FI6%vJigGey$b*;gtvUSYNU=Cy@A1Wq(+C9rw-79uBf-C4@{e>H
z*6Z|x|7mFM-;_mqnD}pffth{_0@GJRQ#NkriW!(@I?M33bJCKpZPa09R$8_T#SE^u
zb6sIy@E;pG2A3D8Po8kca9afu2mC3xCGe@j6T+oN7;1aYF&m8dO{w5v9Ju$bg{;rY
zxVWN#N!#H1J-_$;S&52Jbv%~?c8Xb8vP^xs-sUmqxthLa7SRS#S+q}Z(rPX6J44zv
zlG3t-=BzRCWHgvo=l6Uk0YYY~)4=>{PO@2u;qp;~3mar$HXtVFd*fN<hvq|SvEA>d
ztFlZO${Y^#5e8@86H^<$J}tUxMISxtPeP*m@RaO;UJ;>q^XAHcq0DFXdGhz*mgd$<
z!^8iU-%w0<_^ltbD!9zYs#usA)GME{hB<rqWtSd4N?K8gTt5}~`qcChGrG&~nOUH#
ztGCCjRFA@K=1~c6pLY<H*~z_#)*w)$?L)7dYWdsH9KGhI-59?u8>x4Mr)Ukz1i`_L
zEAKM0qVPM?K_G)9h{^K}yYRxBmo-;Yt~<CEa+Qj75zGq|FNhPJY4Eo0ZH!1Wt=E;F
zV$(dls@^0xyx|p!ypgUbP3%esN9MX`bsKG0dNm=!;C=<lH^E8a<3l!cF+n9^D-fp=
z3F_576WX%9(5jy$8xb3$eg;hODVM1~Sa(%6ijxlv;KB_r^@HsgT;a7W^^-X*7n5mk
z9rc&q&7a3E<5yR&Ujdz}p1>gXizkJNeCLW>g{IB%7y2PEK5ViHWGh==0D>Qn6jrQA
zST6|9Wk=69NfF6+2w3Tse-56z8h;|>vo+V~2samI2#$JzqLBq}|2<H)eozTeVc`#m
zYKZCRQqd|hz~m6i=8!qj-qC;BDvYDP2bq&!jCoFKu=MpW^0Q`QHcH-Z!Bu}_u}{md
z#VvZVOj1$wr7*gn<lLX$uL?qtAlC*@5g*v+rtL3AzlnX(I!0ttS~E{1U)G#8orwMV
zoAx)h2*@q`e$`>QE0R#c#7WNcqeL<6hZUw;rE${R6%HPI6~jKKfnj(9ksd#gCzsZR
z722Z3LKr0qub2J8R^2rg<%$2Oa{eIUxu?>c_$e=h<1>DCTDUy(8MtzQ1L1EY1aN^M
zXzsX`3bFNC6`y`eCvwQ+NmpalgE=_6LLkV3t@Q;Dvo~GoK7&Vu#9A5b(p9<4FlI)S
zeX>EWP)uyjWnuE1q64S>00NXk@XzA3Z1Gs;qkM=My|ts*b}=@mb&T+DzfKM9T8b=#
z3`G2jaL-R~{6D+<A-snTBBxgPLgI&K#BtlZza=MeTSvEleoc!woz+1qySyRUqSw4b
zi%RE^7{s@<`>t?Q+f6Gec?{K?r((z1I9*@ZZUVw}5}+zwr`&-wj7j$;s=uDx`b3^<
zP{QVcc>c+0=NJ_Ei_-1qif!3o*~UKWb^$T<AO9&vKTPShKL!{r?8KP4mMHqd_4Eoj
z@0yd^rf+DxEY&+98H%l(SvBE+TlY^^+a66d3@V*N{WW^`u;k{uZ^QxBMb*b4dyqW&
zq2IHxv}!_T)pA2#vs2hV0>O5Mlrx0JF~6Y6qsO>WupW(L-?3+}t7ezS!71Rt{9drl
z{6B)PiSH0lcGI`-M%K!y(k0H;3*E5YivQv9%hkV$-GgIF&;OdQQZ^h)nT1nG_GC}6
zgBlw@z2{_7bj5b)zQkCMw+cEGrpt~xkwd=FUI{X}gQvq!n3y{yc_nS*>;PE>YIRD?
z3~d7kYTh1fX-9acSimCKMm1qCuB^*us^kx5*49*^wKV|{+%g7oKKtDCsX9Rgr^1vu
zm)cJ<WMXU{u256b$9fUtTCg#ywHUtcvB;*uO~aRChQaR8yBXCR$YG<T#MfhdRzL-U
zUoYkS=jy{4gtI*l@<bGVU5Y;d=<u^H#%kD^FVaqvLMUWx8xucYPJgWPVu#~{xf7%3
zpZe&Ko9$1orseYI>bws3X+`DICIm@+0T{xRVt^8w6*h0^EX*2qUJ>4>|BQWFPSo{r
z5nc}to$nrR-^;|tCn^HDeyR5E7do=3pN{H)V@}gz6-jbWD)N?2YH&ds`cu3K`h!Q1
zjS6?g!?B|nf2Bml7HEr?0Ac=C+;ve+g{Ti!Y$G6_y-n;4ZAKHTKxEvS+Gj@GiV9y-
zW15gRAbvdp&Aw@7`kPwkz$|_9{$lfO*o_srmy5GPa8|?VHlu+;;oa1n)7GdY*N^X>
zOu@*2@xw%X)E4;*>InjcxHZ1qmN)>b`)&3o@jtSf%!WqIix1@JFtXVYj<%KuJB(7O
zmvFSu{3AHvLN?-=+DZcX`=e_-hVM@uD2alseQLXdXJXp7f|p*z+vRMVsg+-ZuNyZy
z#(h7No-Pe5=58La_n`{|%QCt3EGg}^=UkilP?=4|rmo_2`Jj3>i!}r;p$qU6W-2^_
z0uk|c>&Os39sPUDbGUq_GnJuoQSBJ^Rk%30^QZH#4kKx@l1uE^(E|di7pK{;uO(Zv
z9(EQQP~jxAdx}>Sg-B$ROCK&zRC`1mvzqxCe1I|!!5hmmfy_KpEzU2NP?dH)w!ex{
zeEQfScXav5)Xm^Of-k9m?giU3@l)@<2!k;n!kzzCCc%zHSc_$L7do6Pp9KGUcdg}9
zsB<%Oa#CLdG7I75-jgH#ekW1+D5gQNbc*zR^>0A{(mbY?*n_7ra7+fmAUe;Xe#?D3
z%)S!RwX0`hQsBBzizaNadghp5%1<n~Yx<F!qC{5_U*OA31hoA5Rd11YX;I_81h-Y6
zaA-H}DzyD$Kji%CY!u)o_VhzYu%uc^>#i6pV+D_;mTbmWBS${zKzoes=u5j3WYm-(
zSXJ<g6i6(WIi;ueC>zxyQ|_hr+gj*hv-OlvDXtHtvL)z9@0i`u1CDU((0z^vl3Loq
zYvlt+XC<$7HV<`$Y6GT5S4!`v6wXOE6O{2#Gyz0-O<lNKbNuIuBhCYZRS9Yv_OJD$
zXpQ51=HYH)cS!{0xenA>jU*&XMg2iZ;jdvGg#?0ZW$%XkfS%fQ7x-<|^*3n;s#W{o
zpt^pyMTp@Blf=9_(VPL2<xtBRo>8CnFyGlP%LQd3dvXK+q#ku1h_ID{)KnY$S}t;u
zfwN59@V}}pJP1_5?Q^GDN0-XF*dh&#dy}1sQ66w~$B&`u?w-og6KXNQz>$*;jzU!{
zl&7hr$?Ej=ugNCkV3-mn*&NSJeg&^)(;H?h)4}K7OQXiePzky!_b{2VqH0;N^89gZ
z6XGM2O=}sOi9zUrDP>~v!&$nD_njH`0W0u79=uZgexz95(7HNXvFmr1KDSWF_)X@(
z=PJ+B@|NfN-<10t?dB>JXNe-Dt|}ojDc20U-pMP(vZrBzTXlVGIHnfht-uyx6&sam
z?4LE}I>Ax1#j-77Xi@s<EMKPS@T0l|LAO~1uAtZZlS?4<hnEUl=XE=ZJlo_0UUsxq
zF<P=|N^XCGDw)<IZR9bboY{%A5;vRLAKFy3GRD*VBprTGquAaGRRdcR<2aW>S2>mp
zlZfklW@9Z5XVt3z!UkUI3iI$g(0$*tm;^lIv>Dx5D#CZ}J}z@_Uv#K8eqRh1N-=wg
z{y4Tz`(Ufsx$u~Z>_)cjPgs3VFt6i$EM73+wT7&oYg$VyZ~M)gyV<_rR1)n|mJvKB
zN~WmgYRq#YG3ayIRw{O+DFF~$l4LG?i?6X5_1F`6#iVYoQ}>u+Y^GW5H-p=j;MKUk
zHk!tMI@6L*G=;oCci_h8^o~OLbB3n2QY5Fj-br!8kT!SHA2~_uIHLwJci__cFc@xC
z6C0inB43)?Q#ry5^>ONVy!xS=m#Ik6B*Z@gl(}}?e*kWvF5ZU(dIdg}Kzw)&nwu@9
z-7l%ne6Tx_WYY{Jo#p{koyFJ;9bIFyx<HOm4F^wje?~+)dhGd+Awb`saU-ISf&6L0
zu9^A=FJpXnfkCGP`1MW*T&{r==T-=Nv7YOJT5|}1D>y=;sLV3f+jA|1aR}wAb5M|=
z8s++f1i!Z)zm2L$o1jCKT3|?lfJ4evO@VnDNkA=tRv<eXyT70B^jbpm4?!)gI1XDJ
z6TYCYhSTk%v+4gqR*(2#eQ4Yu{1qpQ)0vo3mbh$#7h+zUwh(C>Z-FOl(#}Mm!)5xg
zDXRYk{e9}M9}-OCKy2i$+9Aqafj#>{l028;uj@oT(Q~DDb~E#XNO7Ujd7=PBp{+Q3
zb~Sa_YkzMio@F~)zx2Ade|=X6M!|u;(l3j)X4EDeKV8-YuH#i^s|M8Z)xe*(kdOw#
zd9T<Hl&1w1F?3xIKEsDVAg5xfHU;CbZrA$fh_!SI!w&LgmeLi{Y>9R-ka}*>9R$4J
zmF=FXts?hS=)uvZi`Ue%ZQJ1R!o{YfO4#D>R0<fC`qj9LAU`$Fn;S%u(qM+mZi{Hp
zrI(l%OKClgW)eM*2QOHCn*69?7(G2vgo2cjSv$Q<G~jb#pr)J_!O8iw;@91pYg~_C
zVQng&7dp)5ED_uz@{#SEK(K6noXZPX`b1|Osxl#fLcix>AT#3M-x62O*HsJ7@jF?L
z=y(vIacFt;pwO_+*ex0zEcTvi*!_a^+G<tqryNxP$7J{S&+Z7JEt8!j8jksi#<Fz_
z4P(h#>@S;%@+pMFvjJH=^nPY+Ee7w4%2h+Zdl>IjofM;GbzrWBET?xWe>_k4g|m0+
zBkj+Uic=k07P*cu#FNoiwurg%>Bt}9YweIiBm8zVl-=xPF1^lX_}9pt)`5UJ4%s%u
zW_Z%H;5Oqmwl{9|>zvF{f^h=~?PTyUujq|4NMcIuqwa)<&)%f;R$T?7z>h0keCxoY
ze15>|FdNzouMy|)6}oxO;cdp<Hy^uzC45e@guINfXF`LE+>d0TPcM6IN&VnE0dH-Z
z-(?<yd*R^@N$Z7v=HZS85SO>0Cjk?Z96leHOa*R<l-Zsb{-Wb6aE!LYwq!`TjXS;s
z@|nn+l$>)(R4VXMz1FXbR%1%wSAgJ=O!2iEzP+Up)8fwtM_%h~cy)jcyVS_sEWi;=
z(a4BZpK#n&s6#Jc?qg{9^G)I)`L8QxMl|ssY+*oYRe581A@$If-ParA+l(8*LqUz6
zo|;+6(7Axq)C2sKnYqWDIA}tSG(h369Ss)FHreMU1kYo?_<VQ37<ziY!@veLsskUu
zw!U>JHyU#Gc4F#jE#^(&WZey2YHEv&-uDwZ#IRD?E=3&wBPek&MsDSDoPAA^iodJ2
zC47<26bRLw?sA=(Pq(|L&*e1iI*AHO4w_-EY3uk}=OeAOIxmhpm3kKSJu*1ZIogSU
z(hx5}h(*ot^Ajf)%q!~(Jyj{7b6{v!vM^UtbkT1rz-Md5qBfv$gC@03$>hbjSdXDB
zHk#~5De57vpQmnlPN!EYI-J+EsAj>r^Mcy1HFxo5ihl&R*>?aWqBGbnY&-E@<H$Xn
zxzo*?zqccyuGD94HTj-#jPw^bAc1*H8BCtsyUxu@vQUWgl3$L-Z)CNLmguyxvUKyB
zwA|adm*3`JT}SRB88&;NlleUP+hu)vZi}Iytkx(-4OXU-vZimiJ69vIU{QBwrgj#i
z8@>zCwb3Vw{f~`i)HX?e3JBk?uAHj=Suy9jm7o;$HK63i;Ua6T=&-o^fEGGG_pSue
z@T~Oo!z-SD1hF{guuolYqOC#t+}OVbF!eNAa{t36WcPe3V1KCkSiry~;o7mWPc(On
z)o@dfxSEZq;(4`9jDup|0A4rWCl^RZ(MSowMW;s)X9OTenYhU$#402wUUx6n+?ytt
ztVo-GZ}0<Pu964^!15~0I8FP5ECZox@mAk3&(Y0%v0BF4qWRbD;uSfd=q@pQ1N-MR
zD}{6Gkzh_rlU>uSR_-87s`fo&g-SsmIuaNM^*7*=A|8E(@85oa2jGr_Q7V0!UkV;R
z(Rdb67NTGrLj-J$_S)W&2e3vr7zh0UofQer%hSAaP9-jk*hC!Gk0*jGz|mv(iyXYi
zT6GWQDc7+DaR8G@ziND;yHP;Z0|KGVn1*M8NeL2NP76yePK${bXD|eLkL{1pGA7hP
zbVc`_&WSUVKEFRHG~>K)9HgOG?8)8FP>ml4y|OiJJwZNXFF&~gmS|#@m%fngu;8<N
z2J?E5x*Oh%#`8h52^#bt=!ZQ5)f#5<d9J?P`dH|mxD1q@?rwTG3#vHKnmX5J&$j3I
z`l({1-!!mr3ov?AL5ball)kKxfj0eJ95bw0mb=E3dkfp|*6p<RW0)lXWn|QY_`>Y3
zSR11TfkmXx#^6)Q>30-{zdcfU81}kAG-WQY8u57-AI%3;Cw@C=Ex6+whv1Tx#MSiZ
zRy>R=M=Nn;TD`qFe6#nD>{yFyHZ;Se2h<s$p96JjS~M5fn>`&biNUr|ym%XcUx8*@
z{<@jsMt|K6SPch9SK7z;@{@pXFbrA^?a)^hMyV2g*m!3dcddkLy5oEU+`UxtU~Yvp
z;ow>$_KP#{EvL!PjvlpS<TtlHDC`A_JORVpz`t;?mM%J`lbt{Jwdeb+yKB*9N2PJz
zw2l8{(D;Ed+dxww+iMps5scf+HWC=S8PI->XKfTx?K$LB>syG1Vz^^-<gQWAm|C~t
zMaS#8^Ew(XKzH5Bv5XYBX^!AEh9vDR=fP5`nu|%hEQ)j(&0#!dThq80cIyzWirX&K
zH#(~{vv=qUelDAI&mvQ3y%_r={=YzOmQDQJ6hT)3NR=x6*7B?_tgXFnmTmDIk)f3s
zF8-noA=bEvH~7#sd9lTlAV2PtXpQ72=ru@;PL%bde~u2?KNmcX_`4QA^FoPFxy1yV
z2wQHsnx0~QEPkl>4*oeqRD}O|3G1%LJ-Z9-VDCcSuKsI0RZ^?|yzLD*Y#058#uh|(
zzoy9CZ_JLovHX(3-Lio6I!1K_;I2GKlj~T!F0FkZ#<p?$N8!@m93DyF$^EyfXuojr
zXf1@Pak8~j{f==$n-KG2U28#XD|b<z!Vx9q^wyJIgFrx}P6yBYoaDu|9c9-Epyi3k
zj&<JO)Dh??P!s6KukW0e{8647z=rVl>2-Dt{o!s^r$?G(ji2DtHlm}{;1Bv01$-RF
zEh3*Ser<rj=;U+6{YSJD0z(Y`C%Nq~f4}jY4`Uvs1kR9P$4M@Uqu=)oHjCXGZV?*A
zFb80R_JdRkJd=_hb=D)FEQFs(=?q>^uE%ZmLPQ;PFEb>M<!N4sDGa>4<ud(pQnWv~
zGQk~-*DI9;>kX8-;v>3IXXDWZtH?2}R!TRjznDapVMk&iS9lh?z)`ID!7dQ^F|vH&
zp5ENtsE?}?dE(IM5IxP#3H21XUN|WUxUpy4aA`Y4nuY7?eYZ3B1LQ@h+`?%0s_Nl-
zN1+;d*U(T_hjBpLFWIvmvLPVv&&;)&n+_pO@hj|bk2e>7<Z7o3G?(*mon9)Kbo%<r
znXg8U{}e{MsFY?vc1l+~zvHIH(Ek;TNO|SEVvrlacR{t56PfzGuV6N*UT)X90IZm!
zcSl!VtSf@q7>V?-GX_@~NN`&DMn3%BqAU+IQyKknU=F%N<G&eMXKkKN6+U_6xIN%`
zm;LObz;zcB&r;B%yTNw)3zOnP^vM&*)yM;emam;NUDqKem6q*okt5WG;@n~}?K!H4
z(DlOCLljR=X7C%+_qNMB?k9PPnK4v{4~}47eR=(>Ab@nRxjEr5^1)BJLKCP%WxFk3
zWRqRNUMByN)O5<mX1+|1{U4jIT<HWmXW7$nhdFW7MrNFjLHkHmuZeV|m8q}TreQ3P
z;k%C`KyVDLweqL6+ypC@P2cj<CW>AYfQ>Mt6$#QPG>YN%FT@52xGwzcrF0~U#W{d7
z*?4V9J@@VzHbP9_0`t9#H4t)DCd|o%z=u&3VRZPU|EbAq2gu2@P2&_16E6qhh6hvA
z!Fg2;b1Ge=-IIHVlo&9mK<r>Sa^g+OmM%Oy`h-Guhwb+-6!3unZI4T{S{+}^>CDcg
zZ&YlK@d@56J-%j8JcO`eGSM=)OrrW#&Gvy=$25AJq7e{LV&rF8Z|jm0!4c?X)kx|e
z9T_0xzg64nz*ukW(DIe$&GFGWfH<_YDAyU*f8Ah}1IF?j&){ODy3(1RmN^ZST3PAV
zgy9>Y?|+^lqhY48^7u5joJFAzkb$T3F~8Zb#;w(!9xFwzvBXAR(q1+(s~dEgqd4{i
z9ehi`K0}>ww@!~fbnE0gt&!IG_j~7Ujezsv*Vd)Q)oLZb8l*p6f!qs<U+95lxeXza
zP`#tRRk1X<j|rD0=WdHjKsccXZDnZyiKSKelP|8O<~`1{Bg@)T8Qt{cW{P=Wy47eW
zBc%W;WAD#5aw2c?&<w1xxnVxlT_XSqqjS$%ZgB*ui&tRmvH}W7Iw%;q<G8vAth;_i
z&+qcqV`VCc|FAEjo$<FjWa-ru?qrBI_4O9aWQMX}x>_^Ck9C~~8wuOifDJh@4Dj%}
z=Iz^Ah2$@sRSq-Hub)*lT(nir`!N4Hmkkd0C{<5=uCxApn&7YBQPh+jiA<>+WOTkv
z?;;=Sz8XhDR~JViOuOB(^l>_E&viVo)|}QzWy|c-@lV=bk~V*R`)dI|lbqL)-Tnsl
zdyWfW+B2>l5Z}fEUU`xl$Z_!O?=9rEjAAg52XQWaw2l}M+VD77N^pR4^u5z0HTt&4
zjXZ`W=1hD}fmO3PuIKI;?q27#81t~}=WaWz<_ZB4Il>Bqe%jkd2TjsCXLutaf5>T8
z?Aok$%4u|FV$+dMdKV1ESfryRCf2+A@i*-!oucT)mr!2j{dM>s&)4iMTC3oF<il%;
zkGqCy?g)~{gEXqJgA4%otK-N9LOp(JUSB3?`h4z;UgPy_iFeAbZaAf@a#f0WkL`o;
z*ggCMy-n1cm)3cj`VW>uxL$C58#N9{`aPxnAUJ}qG&d2lmOIDwkHFQ<_ez@x#ZbB%
zUh-5mnLdP~E@|_!)aM7Gc}jLSp1Pe=Bp-?-ND<>XZz^T|j%S`rz(@iUw2ZY}*A>QD
zs37*;fxb{+hZyG(T9Iv`1E>G{V;04IA_j+Xq!dem0v<DqF;se0njhcnRRZvu8b3Uc
zKbxsZwq)d&Jl3D)S(e4XgpWA;SUVuzRq*N{y85BcHPmXOS<_b2%wdZ+WU|&~<k42$
ztE}FX0hW;EmeF-n#q3@32ZYC9{>*tXi-u`~q6Jute&&;1V<k7bS;N7J=<OWM)28A9
zmQt#?)I#^>XKYaoa)0?xCrwIb*qr+SW6>KPl8jDlF#Uj~wz7o7X2K@g!Ni4Wi`Ayb
z$~s^iEIIU$M=L;yd!^(zsC6$uqEyVE&KiF!C)u`0TK6u31lKV&qV+gn%MgwAsp)~4
zYqp^>4}g1pauIL(V(0<zZA@B`o!e@hh+*DxrjbkEVbNtD4APX?po7z}oge*srC*@p
zrXtjI-JuEvJKa428h;v%b=8Q~o*5|L8{)g{f{9I%M;Fswa<^!p+;s+C^|~9g*}09v
zht6a137<W~*OT(;#vvy<j*0Ybv7hdvem|N@QpLg9T(V<c>!KOku+4VdTO6}7FUz_q
z%?{kY*%sypq)FqK*%{);pos&9YmP3n9AY@o@%lFhnn6j+Z?48aZHW*@G|ivl^-3~w
zFHZ3=FnHU$dRK>c*zbS;u-p+a^h?{H7BCHrpvS<0OiWnEZdbFrNzBBzwRb=|*c-+A
zxW#t|*gy~M_u4L$daBjbI?VMGKemMrn>eEcGq_TAaXP#$B~{E)<+*_l-T_UBFOCzJ
z>oBS5NoHD;1sMnzEc%1QLLh}=IRm!LET9(F`Ge&3<Xb)6E<v6|qd~l4a${xze{7&h
zbkb}5gic;d-azd2l-!$8sn6f`vgd0-@)3|7jpL6U|2um8BUn2t+Alg*lP7D$^}MFm
ze&98s*P!Qb8^Y9_lCFs_d)>&M2n?yjtLD2~iC>Mq$%|KZhP>5&Ww4bO@3X5G&4_=L
z6u*Qsm+=dv8)AQN2xpuEDNlY6pqi+6Do<Obw#Iv{D`}m6ii_sF2DgPJrz=a4Or9d#
zO!`}s#joh)72;B1K1KUwWkrKy<?SYj>Yx3&9%U(Rc{IFIpT>kTxb)7ou=$<63(?qN
zS>}%9FATo-v$u-kI^mg<05w~yo8RoeyB4lpH5H`XvCV=jrdkf{ZyLof(@60>qlav}
zO*Zk5<K74V%1Rpn(z-}ZRp0Kpo}#^YR?<=oznk7K7n<4CZRjfcFM6?CA1?!O*Gi83
zy*BH$o;#XbEHK=}t%2_gQbjXZKi8>)xx8tUMHO1D2D0~ACn`>pOMdYw%9|M>RNQ#h
z5P~<f8s`iqqDd^~x?&}#%SB`1&3^YNTAx4Vy!5gb%2CahDF5%thch3`qmfOUXhF33
zD}JTKiX4|d9%pv<xE6zH<);67$@!ba89oDk2hXWaD2G~GR~T$ELHVzHp8a`_^~%-f
zuJ^79g-hrzWGn=S%hK!y@nS@7#d;@)Q#LWg?ja^cnvw>){|NRf|5A13khF5yIXgR3
zUqvZ3ukGs-6<AGIxSGSwA4z+c_%QTcNei3MpxW5@SNko-Le0yVNm8{BiBqAx<66#b
zZy$*<HT4GaU~@H}<ZUfx$UYNBMwX`aMzbtDZNht78@%^gR?WZW5Z%06^LOH();bId
zwR?S9kq>>Wl>QiIpBdNskc9B0vQNJ=tOoEU;q8k|gM+x}y79y3tNSghWo5-RrmDSb
zJr%6i;|IDw;$YAX4U)S@2P?NUBd^wEV7vNFUA{w*GD5d!msK~r=t^CYyritzv!*}^
zY@85CX)7qldM^2AjDAhx=g!eh)9FxrsL%Y>OqQtpgsc(S0C@h~VdK>atJfpuu%O%v
z=&E_wKZ0;CgtW|Rpsd7{tEcpP75BTt0pvG<&Gtn1_+K~zg8R25kiU?1$F|8`PMhkW
zlhMSWCwE=+W}k_&rlVl+yHx`Ak=e+PR4$tLEKcvYCrq5QG5@-P7(3Lv%+B|nvwpiD
zcEh83I9Nu=pyoXSrGbiNB)G|x^_GJ4yc4Ic==B{YRTPY9a&(2GSp{Xoa~z5v*$y$@
zI`Dc-GJcMI`vOz#Om%p#$BN;UUr%lI-lG`Tn!;FuK5k{%Q(LLP-GpBFrsWxON)u1X
zB*rBNN<|JeKVm=fb}G|VHErAhZiS`OdW-&08dhEh;TL&x2XfO|*u*jSfIhunc#$;d
z2C!IsfkVCdD#V?<0!;fyfWh^NV948(>m14$_hIW!xD*L-K${H=;ND}^+vf$<UD^T3
zy7BmQvT})ycNF&{GUIxYDy52TyBYpR)vG3C+F*#OUvLZ*O>MKcF*|4Y$D!c|+QU!b
zT5W0??(j?`k<v%>w<M#qZKk>K7`$31stLoE(L81F<cp9mfYC|!cC)i0acK6pdo$__
z_~Y+Ea>vFwXkQ(}KY|M-Y3l&V!C#_!(&S1@>g&gqesF+ICty>KT7>MY(>hjUb2-rN
zn7{z>ECq~x8g{Bw7*_hi?vdljfn7TNcS*f~>K*N!w*%hA@7Nrf-450#Djm;|iOmuP
z<$EBXvy)=QdCADEBvD}}9ij|hQcuw`DE~exf`PupnM<<>F6_WSL5e<4mD7XwOts39
z7NkBYqUB#hNXV(;>!V9EqJ{TBf9|aVig}IV>at|?WQi{duf6f8*%Ejsh<f5^;MLcq
z<qf8vD>yBy-N=jWwZ870CpYrV3CCazM|*sRTUk!LY}A;+<UfMH+8v?aFXYzPQ`e9}
zRP0^d_~xBM3^9OTyF&Sd7ny@-SU>FeR`7`KLuTaJ&ikBN-AGw46r`r)a<+SM*gB@k
zhA;X!TMBEI(>-5n?ENG;N*ZKSXNCTCXe6Xlh$zuc+wGH6MHRoX87OEEqhB;U1(<6k
zK6R4eGV3wQ5Ulrh@hzX<k@Jn53=v!S%fo(sCo8D&HLxzH6?`Xh{q+VjfXERlQZlY(
zaMskYdpfsA?+$2saA1dRrT~T-=f2#YYuXK>GjxsIA-%*xv>+H^I^%PRHs$6{wABJ9
zRHzGag7%=Vdu`F@*O^X=r^P_J3s?rmfGXbviQiD!L)E^1mP%BoNfAkI-X+Msw#O~Y
zRbplahSs7Ei`~aCVh?E1d`-0TOFycEqeQ8P4PgGjdecz$l7%Y1NCM@*xAwsK)02Xo
ziA-e6jsF;-{>f!3asJ}5Wp*@)8Rxw}Go|Ymz|^hMd<a3d<nyf(kVZR7#N!^jQlS8K
zk5iXOON3d(QV4xneWt{ZfFbzvNCG^%U9@uACxXBi*FLix>LRY_k>RJMtO;q)E|W#v
z@Juc+2m%*lz<1oH`SdFH^;|ae1>@hVWK*fi8%0PANI<F?`)G0c@iD6D1?n+`7L$k|
z6bzH<A|ZWN-oG(Ffi$icOY9<Tw|$5Gy+`v={rHccfew`AYI_F^O7T{1i5uoAn4C>f
z-`+%8EhIKg*c??SMo&9Izwt$Xo^}zRhaU&Z9*o2&(pH^RbZE512h?l7={l_lW1*MC
z5v44h3bwpB_3op|TUU9y8KRRwqB=5h@W{hfUZTeddQ67qdv~e00AS+g>4Rq|DN=$_
ziPFYO^S!<&B{ze$zK*Zb!X+$r@QTM|@N}=JZQzC}rqh;B^H|o=PL$eXF;8^m2E*F)
zNu{BXM|R)+2;+D?`L=$+_B)L?(=7VqUZ)C4C^p~0Ym(j+jZ^JvV%2DJRZnnE1w68B
zEP@H~2OVIuzBRKi<JG<X*Hg}y@}umC%?3M><u}pWajKTi)i9N}{M2jdTNpX<iErNg
z4tWe54&F{@1;>;NK+o?rFq~TB)x>?T!0MG7Bu?=lRg`oKhT{0U;ucuK299Yv*Q1Mr
z1uvD{bg7`VBVba$L7)Q-OB(gJL+|Uco!~%4&U+^VvbNiWZ(UEsPkV%vPQ?oz%r@`~
zYfGo?C~l1|S{y{i+?_2vb@;pr@`FfAyxb}U&!uhc98K$yTZ;P!LTSzv_@V<)C&fL*
zZqZ@~9E)_NwiDL&z5}AA#UI+-L*(!yLEH!aM~JEsK;5h!A_u8B)_isYt$_pPqYtmJ
zP?OS2ML}KSr=4tz#IkiA78!Ii(85|2XoO%LX=d_x{FM6Vs&&}<^{>cX{&JSYk{d*%
zgWy0WWl~6I=lW6-|9*Dki_^)jf)7U<b^v6FeY0_w)0uY99PjW$Iw~Aaosw!uBm;`P
z8wee^E*8OzJHe%`wti6R7<iHvHvSHz@Yv0pFK-hs1Nlj-b;g+A=3_%7m}~p}UWuS1
ziX=`dKbnd?Ons%Gm!t8;;um)gtS)5Li}?=483vanhBiihsJ-rXzXB#t;;g_A!`_Go
zR}Iw$r+8gWMHMW(>5pwGL@7qcl*P}Z3~Ya`aFOh2{&cItpLtg9SD&7WN?s^lHVS^3
z?Hhz5>f^0}1VT@KLp4VfmN(sqWshH+OqeRVP_5j)Gr2F&Cu-U%=l9j{=L{2ctcxh!
z%JyY2l}O^^ZB~K2_jZMYp%O*Wh4Ha2>k~(k78f3OwZD7=;nt1O))YTNN1(ak-fk6?
z@^_4u75Iw!oHZlU?1(QyXP-ps?PuPbs&DLlOk^h|23=zL&Xiuad!*PL{p<IcA?1>j
ze9ly9-Q;CK$zsns-dE1>;a2*?-jzLGP=b_kv+Og)sPOaae{waNFp(Z=menfMYmnF6
zysTJn%3_@F*?v;j(J)b|KCYO{UMv6T=US5m@7JH12WIWb!PhGLbn4M2_UpT<XZ(Hd
z$Pf9FbOJNQdPiUC^7TI@UbuhmsblnH@!3NQ{&xkD2R~jq$;fbL>}kYYY?Bw<FW?+}
z>@cxR#V=xDk4G+^V)&SWCJaff**(rrMJ+a)eoE|U?+F?5RPy_K15~?3Ef-P`TyWcO
ze97#Fd-2YXR0_7&QBVCJU~7icX+L!$hU5FkZVH_4&i&HK^bM;TV{BvkM<z!nDpX1y
zZhQ!{VUjoA#@sY|f4cPTgG8zoMbNo;dln9}2A(RoJhY^aYhDyK8ugjb*o^pfjde_2
zvQ~Mqpxj0zHel>MBu2s5TVFk6X4=m>N2<CeNcYmu&$hIy)IwkR4VC8#{&Dr50>w_b
zrLDXL6Mrcpi?QG<M8+P!zIj_KgXWw2*$nMG&()g~uia=1bjt-BWKU|9y-F+t|D@!w
z;#rahwKOovYke_#<`45G^QJzDs|&iCFTGlHq&zvprp0fB&Ff_!T+38MIS@W9eOf$-
zx2M>4JPx8Y>01-{DoOa9!%>C1%xR-?Skhpxp-Y0UoaHU~o!P>J3(IDpgeJv_;<GGn
zA%n%}Lr~tF8Pfm0BBIGwSMpQzbQWYTdRq4e_t|)f>yUan`&D~Gme;_uy*Wi$?Kc$e
z4GFw)1C{ALgJ$DJfhok90P(u7fDil+(SrGSZFNqe)(b&6P`fM^we7ZB`bUDzt_J4#
zM3>CN@WQyJDqQ`%sGi#C;C)O3rI#4zR80$4|4AwQ9sPQtn97%u2W>iVrT8ky2@7uD
zUdhGCrGWHX!EyVkL6}kCpA)a$z77b{@K-0?P=!zNz@rh@V_>0n^@e`gzVHcB!icfQ
z9;xQlqgvlrReHYuqcB#uK`NF<xv!Mc;EC1T%z-+afN|Ee)dq%PKunCpc=F?GukHkG
zWb`CxHQhyP6X@3CD?Kq+J)fvlR}Ds{t_rYvu9YJj6aV5nPQxUQ_e^E83K;$)*IvPb
zr~T;VJfHmb${#k9G${K0x7@3V6~0-ACK>vKmweY&MLD%kntFIk?|rQJSWCoaS7>PT
z0mQ1QyVEHR>@VaMhVP1J8XE`*6PM%OFY@PGJ2!U`>fhNYHNL1DFn08fNj}Mba`<fL
zIk^?;`+lAKjoM-`=>gIi3nY<1Ti(e34KU-JFaoqjmBghJ4q@M(UeT{fQ>=64wq&Sd
z;u1?u65GZ0Ne-zXJ+cP2eh@fP3oiF`;dDH7E{?m@aQRWH#IGxOa}DgkB!4u)%Jlt7
zk#Csf<c7trcJ<<10;R6KGwHxmL*4wB#oaU=RILeIbvxGEN<43WtrRfezJ))za&DwV
ztla&_J_jR4J&DM6d70{c%$!y^Oa<UNrac!uOTD3%T6=d$(jCLoMKkfuNrcDp7!e;=
zH>jw8jVQT62bio$JWzI@YySdw&5Mh#IMB(onF=kU<dGZ;I`)QOnxniYu_bdo^);4J
zBKxw?Hzt`^Kog%#i!a%bWs}#0CuP_(i}LYK_B<sOD=i%nJ8#G@jI2p{ERVn1rB0vU
z0ETb2#<jmp=o-Z38#YabRv}#z&c+axJuK?eF{orcS|?J%R*BIuqh(1REH#(iYaXJl
z08O8xe?x00fyT@ByC(VBd8pW61C61{t#W?)B?TEJ<3FeVfrxEpZ>0M)O)l^!ZU2zI
zic~V$%yG+aYOuBb^GM4m68w$=PaAvZDoT<zCAJLY-!2Yp6J`6?A)!alX>b0};=05P
z;*gSB$yJh5QIo8Zp*juRK-b9rM8Sb$##<g+C5{U2ux>II8^1}qr#Dm8rStGlii(}3
ztj?*_{|PB9;7gtwh3%&T=_6eB1hM7(G-Z7tZQm^X(Bi)9`vesGoY)uv66LCRAdF-U
zsEg_$FjTm7)xg0m2kJACO|J6Kc!bF(jfKkaAo~pX1^35o<Quw{nTfpQe*`iPK82%N
zu%w>~uOSNfJ?Gr%K@hyGO298gY;k6a=UD%HwXwS9E!j7^`OL^<&AE~q09e1O-VZLn
z7$u0w#dD&9+t>=;&w&mLp^!nR4o+wMGPDVYbA~EnFZjAs(3!jGi`2I%&%ZpR>qJ^U
zRe~1w&%00F&h3f#N0$(;KVWGzW#5Gcv_|4i=$;gOts?ST!6Db1lIo?L#Fpq-)cf5V
z@RB0b&!;<k5^_uB0I#|j)uxv(x=>CIM|8Q|%CDOR{17>00OmDz-;rv&`op8*U872v
z>Iy$}LR5fQv^$22R|PkhXHGx^uHU-l*?_z&wGV9Nc}?wGIcq8zJ#m{j+<TAiv{LOK
z!GGk4gh9=vWL<OOONeIgi%@{<uxmH#h(;4xCVfo)T+`_4QVUNG<%?WHfkG>lBBI3C
z%5tisDf1crdPI4kqT5V=_+;bctRGj&v_@jg=eRX4!*&mw0R}(MU7a_)OPI3P>EMm{
zI7r1<7@}5qixXF3co{)lS;w2>>xxs49U9h^zg@w7sNvpZ*N94?&1Erb1J$ih+BJd|
z2t)7gYuKAy2pBd;2WF@>2mB+L6xQ?jknqR-Kg!V02g<C3b|JE;`S}~RL%!_?YViiQ
zh&lWv%&zj8DS-I(XQ~F4!kWNx|7>ekSu>KQEBZLs?F-l2CFp{qoXxEnPJ2Lh!E>Gt
zZ_(Q+m$P~7x;@I<-bfT&XdIm}6Y;eelOs}-e5sl<DbgdAdp;tgwZXH;M|Gw+Sqx|M
z+3?&~k*NG3%ZyKw0OPtGcfAcMj@JE$`%)b*Fc>W=$qdG1!RjyLkRT5qfz{xzt^R?4
z_T(b<-6ezfWtZ`0xA}LSK`$}9Hv^)Et2x(2Oustbolu4Q!IBU!Ya)4rJqap)P0N1)
ztWq=ER4Qf@Ke<?(-PG>|<#gp<YOD#&n5@x6&hJfh0i&>Qt%LicgHT<IL1<a2YD@l;
zQl$G-7_Zzr^BP_IvHp%fe*;H?3++Z%GaQ~n6{%K%ycApkLe@%jUn3W8f<n(cuwYAV
zD}3GRd+5!s<>;A1R=XG65}y#?YPViM3m4sAC-E<}8v?@I|3Xz0lS8fi4cMxLv=-yX
zs%xN2aycy#^G=<=?obhT_3sW0oc9~}F*`tB?eu3XZNg~fxk<}6y)wHhv8+F7o&BP;
zfQCtx=(q!2h%n$9^~ugms4iXaX!6y6zL`fdh!<355Ol}jLg3t&sQT}h{te{LrCThb
z51L9=zn73O)Ssdanltu~CrT2yXa$zGci!*)ivPI}5wqB`KB_xSD|4O``$ypQ{aN<|
z{oz6@s}R}>A#Tw}(a}>ap?y8tkA0Wl{wOI(wA|2D&5wS7e)rN2MnWB~${^&43=~x}
zzKoz75nQTqa+_qqk)y{*Cu5@hLsm+zi<^pGO={E&fA>KszHiO0t5rUy?5Zw8q>5YQ
z6sq0daJ}23^nycetU2Ta3RT#`oMK5h_Aw4NWG~rn@{2Ek&sad;Z)VnIxC^_&ugiZG
z+S>FDw|XEZI16;VMmPHcFh7QosFAtRaCt-_l`4uArrt0dxbvA+y@sdP?cr~XzRK1<
z2OFMOqDnFH<(mXNRsJhBK5+8G9$TGWq|GKVx~gjirgi!@vc<fFNXdXNYErJw=qETO
zg(ZGo$`O8|E;R6l?erQWrCMF3FU~hD3i?RK0oF&pJeuadhj*zIkeRvnFAMY|&++!1
z?{;2C9jeyC0GkiJcPR31r=LYM)fzPuZ~4%Y>^6Z-3xTEW7K`wAFJZoVINc$B6_<7s
zf)*u5{qI!{&q(%9n95{mQ?>oP%3N_VK_3!#)YTjLv#7Z^h7L5L$R#5qv6S7|WRNXT
zKI1dVMEVy0ieR({T52W(+<30V1}W5BzX4jxyh~ZOa-%XHu=SijJ}ErbN|iU>m{4u>
zoX5C5mj)Wl9dfZ9P+zivALL|K$RH}#JD3m;)fzYBXWegt+Wy$Xq?=gyAIcnR^M?5z
zeCW5f<|q__Ku~kpR_3Yr487Ng%?`x8f2g0b`qB60=S;ha;%oU&8Rm3!5-cSN%EA5u
z2P6QqMWwSTiZMTUHF?Bv*JEzV;D#UK{Uc(L-%+DQ!B!?FrdI2o3FZWUyI_+qW$3yn
zsW#0k=qb8B*pc?is((H@k*Msc3`w2b34V-ZuTSgpnp9%c-M$_X&VfPCth(Bt)g@}y
z`j?e8Bm*^18?BiO{q=eDV)%_<KH|eNM%c4&*Yy&kQK<uGLux(DJK*kag103HW&zFL
z;%gIiF4nHjoaxLAwb&KPaNrkj6y1UA5m2XIEmS~|6{l*;#w>H5jmWkZ8Krk=2VmT$
z7h*2$>$B}?^AkC0#tmBP_yryP?WemiX7s*-n_1Z3TAyYQecpJ5NlSXWxK&U+M7zzg
zquswO<`0aO+1as59_W#JGOT2)wh=#ijO3gOD3*IuV$r}AO_{|#?1GI430o2ob<CDN
z`3Q}lh}VuZU1om+)#V2r#}#pRyS+`-R`2m6O{UtfjlAbnQY*IA3yJ9)o)wjza5VFk
z#ei&VchGiR9wL!{l(qyk-8I2ZKMD1nVOD|N*4Mx^Oh-#(jI{^7pDgy1vh;LdUj0fw
z+dVKpJV%H?&@?6WrD88sxibwkg>!rb0=}f=Q58R{AhLGweif_eh3$4^DD<8*e!&NW
zW-2xb8$z=c9~wP(yT-<}UPilPQ`d;T`Eyq0y`PwqH54!{M%p?UsEOz#JN=C5+A*jU
zjj6<O6~!#}=G5WP!VdHB9EaJ+nax*ovKCYHN5RqAl+6h}s72OuzvOH%5%aFTr=D3=
zpOcb1$()79r#D@+IJ6z7I}eixq*+f_`u_s^HU!CY;V;D)JXK(+F6&#d_u`O$_3PGc
z=JS*?9m;!(^8Gi!38HwC@nMI{iP$=-j>q1;FHh257~?8gN9kPAl&@xbc%@0po)2>T
zaPC~5l>Y!IdRA_=;>{z%wq|`>Qns*&e-zFK>DTnHCGnT-nee;7>zn<0NuQmu+S@#g
z*!196Y-!F)_B=_}rybets|ap`jPu4Z^{+VnpucE6C-#5vd(O=r*NCpslEY6VZf89i
z`N&__+P*{hlm7q(Fw`|=n%BUZAkyIjZ}x_u!dY?t$^t<A&3vV)c-vX=?zujvu4)$=
zmF%aP?j$bQ^cnl3>0Q+D_G!rTF&M=t@<7({Cb@g!J#yz%)b1Yc_VQ+)X-VbXyPxIs
zu3yCs9@HZ_J!@j*;uUorE1&T+B-yL_SEF}p9s@F35jCDyr@c|OK-^=R($mkGW*51p
z2$nY8j%zbWkbRu_JvgqX!#2xxVq-srbIbz5Mq7&LydK47+T1U_V$PzJmhEG8EMRo1
zI)q`ja(S++#4v@qGAQ|vTIMg*tlaQD>dd;_=4Ed#;~gta%yU`r%uQ3iwPtPVDJ<nF
z`@Gf<b7iSMmfmCW$JV>u2F%SLOq${J*pJ%+uORwYt9WkQLuwr0G4-wKnAF&U^TWSr
zh>#bawaIH54aL6V(GaC^yR~}OnHt=Taf;Q{ygv=?o?UW3x&<~U%E<Wd#Wp@n#hp(b
z*0H8j9k(OpuXORh!x`?SX&{dzjP1`A<9;g8?sT%y%Nw9mgT-;nbE-1tM=dgzOyp*W
zq}?4hX7cca;g3p#aaQA{WNu3&gf<Tq8k4sQe>!xZzD)ErR5LGAK+Dmh%D*>SXOLrD
zoEn49Bc(b!qRoTSff2N2CBdmi8FPwBK7L196Yk=G7`#=FF-J<THytU9^Dbxt{_#fY
zs}$b3H87MPS^%4e85GrB)Uo~CQpU6Z)m_vp1@B6~Ei-7K3AYCO)a7Ee4(fjHKcxUn
z7&Sj4s;d>Jh9-b43>tF$t%_n})}6cM_o^rtD^FalHk>|g)Dlc24NirM_NZfwRLaGD
zDoow_mRBo|M>RBA<8kX%cKoV+Dp=(CRQga<O*>gO*FmQK`F0l|=kI%0GX#RQwEKds
zn$j;_LPv9BnAa+6weZUYL~QZzUT<f3<-Mz~@cr@Ji23|0bo!$c)C7z6d;^uOuL_~K
zyhy*FTE?|lO_<J05t`{d10UMFMGSGNF2?|3u73!j?#UjjW8t{vvx%H@jGij`cwsi<
zWqQYlei~14r^4?1vU*oRZ}72K?_}4BDBaxlBU?kkTf<S617|(>tPM-U#g(v`VdyLB
zc>WmMFW!xh;ZFYmg=c5S`ZAwhDLCjaY|-&Pt>L-%;y5HD*=wI~1!`AKl+&hv_g23C
z{{V$$X04F|pQ_aH{5iBvw(5S1iowyormGPy$LHGo9q|qJHdr#}`^OclXZt+(lHxo%
zEKGXwUwuR1TX=Tj1>^mSww@BVPvBGk00F7`zji;ZO!!*M_FmMkqy}r2+usBK0IgcK
zf3h?#xt@7t{4rl{Cxv2<q(Y}(4#us*oP8+vC{Gb7bv`)LzhvzuK3*o3zuG7KE2+}|
z0A_6#)k2FH!S35KYw9J@B3_C=Dwq2%HR!b67<5AR((Ziyr~C-gtiI)pXX=&E*?c6k
zwN(PzGDrJ3uV!667pSQg%@?t&jX3r*l}7rWU2pK#-G1ySKj0N|-{I^K?#xQtzuB)x
z7QuQN3|o8FiV2g?+~8vHL=i3oc&l;vP7%?6O6+#~3XK>JxT^Md7qlp7z8s3*X#<h%
zUACKOv8sg_BC+jmBVL47y}j~|j4RNm9#v#>N?Mw+#;eepom*{Dj_EyV{{U%UTGNz^
zqR-_+C~B^=hVM(~4_ZnrE{R$&-Rd^C4c>*9bvw>ldWF&a!LziDkJhy;Q|3d(XU4mm
zf%4YHoFP?DVcxMt8ga5i2^{2BWz@y`)tRK(8)v;`yq;%bzglEP1b;74pQT!Z?t|;v
zs>LDQgI3}BV?*A9MqQ1L*5nNLHJ>&+^X@9?m@Ig8`F*O39m5DdUwTYPd{LP?ky67C
z+9XJjjm^-Wl^)W#!5sFbSr;LOOAeXhs0kM;xTiDz?}Je&EOC)ehTwTpM!`Tm8@&KF
zNk4SHqN#?<0l4#0$-B!OXEe)rmvR%$O9I`i7C+r#(2lifravHm&%IKOSzItYj+FVN
z^A$<$R3M6CvT?`z)ppb~NI2b|n5^C2c31)hUbVYxRzjmYdJ0woNq5Jv3}JdzKi$gi
zt(>2jeQIfLyu*eSpC;vQxN2i2SqGiYNTrfRQWLSL;$Pk5@_i|iGP<raxKs+(EI^xb
z%gT<`JkDKkrvuc~7Iz5DJmqE`_yaX7Dy)0+{{R810v*OJpz;MOI-ivN@9$K14N*rY
z%kN2y*0N@fGax7B?M%9mvW|H)%X@;T6!1IKW>e-9xNYrBMPnh8A-?eRsbiCJ;eJon
zsqQ{t{c3h$AO8Sak^=0Aff80co|vkv+q=_z;8$tR2TF&`rcgi0)YD^A$G<~ORouBe
z)yZe`6@Eafy)o914UI?cjz;6(wKcFM-^%|0yGay#urt8vNX#2OvFkz*^C?}t4k_D9
zL%o=O6$FZXMGf+e+5Z55_M_CINg2IrFFE#7q{}DY2Aj9##WW$~A1-;vr8S#tW|!{$
z-f676szfs;g<{(%+fpip+wD|}ELqHL?qHNE$g#FeZ_&QBdsMnh>+6e%p@z;mq~G>?
zb;2*Oti@oXJ*h{_(9kOvE%vm;zIgPhE(vW;(p4J|nS)J+77T>&r>Ql1;?<$G(<gP>
z9&$e_ksh1yp2}bNOyPKxnPO#S9^r+4y8Jx8)NgbfDKDhNf)xcp<2Cs;@IK*}$6`I*
zVk9y31Rv>NroRgQDrnal97}OJLnLT7sLNv?O7*ZYQ*w^%`2HfRT9jzYF3{Hgy$-9!
zo)5EW8&!oi+~oO(B<@^bXWqGsJ1x>>Kq9(d6?{U{ZfqmBvA9{Lo2r0E<IQ4eQh6|f
zNF>*Xn^TNxt0U|<TD5AJHtIX8=vn^H37ikT(xYoIdJ|ecc$01iN?76ZI&t-{39iT4
zruvy}Y&PVN%C5r_V><%<Dt5Fi-LX%K0*iyi4n2#LMQJTVecTF?IR0;Hwy_qbs%>ID
zDcM|tHHnw7q-FVt996~3R1FzY8?yARQX!G;rz~N+iYSgV>sMhWhWToGih9NseX3SX
zxZ|xJOKrzWP9e>qNv%V5so}L%?N(ORo6@9?KWb{4key95D}97FVOFiQ7LyzM)w@`?
zHVEdrTSXr)9V*op*uA?)HEp68*-lOcbT&F~mmWzO_N$Kys^kiEEEVydD|pgT6stKJ
z8h&A2tJ1e1lZ@6SrEFXBxUC|ran-bubgE4mM_tNB2tUL<4SB!V<nZ^19@T*sjyFKw
zzwG^M(h{<KO{0xpPdj*?Cf^f!H~3XG)aRuxSy}AJZKM()@0zkXtW6&8%(qn`go<!f
z^%b<Hn%HxXD<fpxQYvrgDgIqqf)PxE&Hgd?>cin5gl>FUsMy@z*jfOxC8DWexZzJM
zIM2O&f8sy*C^o&~d&6nrohA(q%#a1kpEhHkuL8e7d`;l}JI4MO)ie!K1-R3-8(A%9
zj~Mw}VB_)R9@Y58u6R24;lGZxuMS@BH+o&%wygO709ml$H_-4$u@&2frx$kB+2!La
zRB0saj{gA0elFEKL3o;$seLu|%l+n>LVx<-rEs4Z^v~@L4lA_{pO>)XJ#a;CUSj7=
zP<d|HtsDI<b&mncjCV3d-F-!R4^((jlTx|!HlHV!q$t5(;vIh~=zKqL-e+7^O~s+L
znqauXoEn!&pUHF2PfErt>OCSjeCu8XU~BNmwnui-1$15?ypHD7fOlfEyg?XfSwP^{
zrWR)xq|AQKe@dP*S##7?D^zG(VIs7&_45}Pp@PwsrTJBdt#sZM!*dw(4Rbd{#&F7O
zbHTR;-ufumg^#%v&h%$ek=Ea7rgq8cT&BBg_I4$(aoE?XNotR591Q2BasD#V$&bz3
zoOQ)gWE>kt8a_p3Tv~ZjZX62NX4->16WW^g{{U)-u{D`V*q6iBrt)QF!VzAP1Z@?I
z=eJB(G2uNg$&3~ppGxgo9hNVZHPuc&>4iz%*%A0pR8w)a*!9LM(oss=t~mr}>0S`G
z-`V<;I4ywL;=Q-Rx|Djho}V^Yr!^rGOFJkoZk1gD+<Tsaxu1xi1+9ERXDrH_q;byQ
zTI&i6Yvk%P*1H>4^PvF!(@^4#<0lq=eR%WXj-%n5hgOZ6H0axMMS0ATvawbhNcFGM
z9~*cE*Tgn61XT>(PBUK^d{g*WbK!gGT_gR__pV7tl<JJGwLG>q8*x#zELaXXrvS?3
z3a}t^{uOOw-5GDnvotfL{oh~Gn+svHQ(R%XRe0zE*KE+dQY&s;QqTh><Yt*4-NjDO
zvnNqdecrSIG;z_5y=l>~sWyh`Kn+<j*NT=heQFkr{{R>1QLC`)Ko>UB?G+G{bxNmo
z0~J?yx7M4to!`Y$D(>q^wt)<%B|mBY>7-RExu9Y$7N*=b=Bwx0rpL*O21tw~>BrWU
zS(nnI+ydlOnV>_Jiqx4k3|{o*V$=%PED2csJt}qiR~1!w=e<g;xpFzC7O{2)HL+zN
zTyS};Yj2viEm?kFY7w38box)sz#jG1>1*YLZicxX9oNiWdRI|nFnJM4;<_WTndsVW
z)3vJe(EgR}U$b76n#Yci$h(zP9(rWgglTgD2tsg8eP8<<_*d;(wZ@=9Asj9~lvR>Z
z9Bp)ctE1^LMQ0iTj4lOUTLfCUZwHqOjC83yy7P|Pls?2Mzh#f#@~s~(lgS>GfY19h
ze*XZK9!RVHSoEdxMMzxE`#-%m?Eb>E{E<hJJt(k@-?RJDeVlrl((=cpH_4@8nf^fZ
zq-d3UR)>};jSlKXx7eU#3rJN~t+ZRJw2;yk`;oC!>S_sRjeF8f3pG)=v6I%Qk=Htk
zNBmWTsCbq*bB0lhlJiWN`d2@9;hW2i*>1knbGVIK`h?y-j^C@|nGeMH_h8H`oxkuc
z^xgjeAb-NS+mDBOvbI%U6|`&F5hyN)tl9ihjq&@@&1+lyNVbz0nNDlL8^PWnw$8}p
z_NOhbw{N$6Eco;ouDIf3ZBd)1`=3Tb<EwY+ApI)bek`|>q{Hzw^K5#CrEmA(n0^&%
z9}!%`_x!>5S43-5kq$DBr_!+N(Z5MY=T>9Y1fO-s=U+M7_`2bL<*}`6UmaX0Nxxd#
z_j1{hsrW^29_7tz+p?@XPfDj}3bNv-W&v<G`q!L|g6Y0oX0t`dm(*_h)tjjl^d_pZ
zgPe0ol!+txn>BVe&eA?*6-p2Fnf@BkRVFd=75Y$yHhDY$01a1-o7$wCmB8ksjkj+9
z0B5ZrCmCX-bri?rr5<w0)D`~#cA}AuP?2S1jcxZ6``_-=GD)7GQ#YQwNBiH>m{_+;
zS1Q&OqyjbC=ZbVJLaFqk+=g5g8L1&%&FDSpkWQ=@y+<Q_xTV@RW15YaH2^v`Df1t?
zI#c}PyQtk!{{S94R7{~;IiO_HH_e)YM>9#DhxDg5S#$b#rpC*+-hq-U6>N}cn;X)n
z-R+u*hSno-=dDPrOK#p_0P&8MERy`Yuqp`t_E(x~e98T3`Uwin(>>^ob2b3T^rkZg
zRc}gsGkJ3>4yJ%7R?pu3DU%Z2Ge+!n7^tR^ru*MF=|~9w0Og)@%?&57KT2=Qw;bk{
zD9gUmMFX(rZM5Cq)a74VQn=$8Uup(AJZ|a5D>l@}dX_lpvH|H-B0a@dBc(HT?c0GF
zr#H;Frfx0wu}ET;E;1>tmP~i_rC*hQtsVD&6$}PC?deWjwKbKPXEh?O+G1nA@7<)N
z-AhPA3u1-K98+GNzbur(<HsqtwJNJI#%NU|sKq)zy_y(Yv2Q66DcsCQW74-T+sx8*
z=*#C4F&nF&hdc3tMt{k!gHY3D{{Vz*PrnM@NR^bgrB1~~H?iFOC!6fPF|}_n<%18G
z>5BaU{hRbVoonqf+v)cY^JC9QPS*VE;7@`66V*Ij;_F!Sm<-0*T6x^?sT!#ypUS^h
zKV!W~Kj9IU#e(_K%*3c{Hq+4ct?>AVhpOcc##qb^Ja!Y3rpc6iZ=s7FV!}BsyqlI@
z0YJYftt%S=_H-kW?Oo@_%|0usG<jDBb&;dmKe~PEo6_z4t93#$1$nso(!^BeX!i1~
zJTRDLP8|XpK_5z;W7^miabeq0vh8fwoTGEs<{9$(Q;cE0_vua%=9g)}KDAt;atPxT
zo=Dusj%n}5N`rZ=IZ>U)NB6N(G<%m6$o}uWSXC`lqA5z{Pb#FhfUn->rP=dhr&#vY
za*=Wtj58?otAAz^brjWz$k|bmT9I4sT;ml{h@~Q`+HG9oogJ9;6{F;hxTzK$wW>x+
zRw+kmwX!Qx2#bB*KD3d=?N$+$I#VUco6FxH6#~JwPHI@&_<Pi>yH4)ZT&;^%(Ut33
zw%;vh!lxOlaA!EpA|9rSwkg}X9{uVy`G#qCRh-8wtwH^fswKh<LvVS=XRQl4gq`Zg
zi7d_ji!uT8#yr(N<E3HfmmX{&h6J2o{c3h+L|;5E`W+Zu)n+a#oK3i9rjAY3Mh3OB
z1+U9L+mhZJ5889Y4;Iz?MZKi&kNN0g3;B?We;WNfxYVH1>~1eE<y(7Oc-m=Ykl|5O
zWRh#~@8egGo8oW9?-F>EC&|9MK2IGXLAE@9z(D+KzYBd?<YM~}>Fy*5zr*icUXdS{
zV{r*m(Yld=?_4#Ej3H71$gZP7)HiHakn`5N6Fg(U{7qu^I`z~WcBbstIcFhT$yA&H
zUZ3$|+(WIjssIi%#d)!XeBj5cR(d0*I~|XMZ_7N0#bd4QaxwN^IcoEbA5?og_|867
z{`GnkbBpU=GdVu>Nr>UEqgeiN!&UtfOvA{@sdYU!Y=L^_tJq@FZJDQZLx}xJ1IvSh
z^s6>Dth;f>D>G2I@~${F)c9Re_~EN57IwM@rYE;powDP#Zr^D?Yq67Z1IB)}#a&sx
zk2F^iw;jN*PVkNNTE?27kj$w>M$ZKCMv`pqA85fHJu4?n3|&Cz-o2~Ve0kxvS((DF
z59wSTmX-Eh?CHmRQnLu^j*CN&Zqf+aD{kWGt($@|Svov;Z}w{Rt{yj5KQEx`U6Il!
zBlwOJPz=oEax1F%f8yh3brd183W538KYeP}?Jy%7zU^WOzQcAU9FRyIzLn0I(PwQB
zsBV#O;aiA)Vm&KoLA)(!p}zR9IQVzsf3|7j86T6<yNekFcRVn~U56Ka8PrXNjDdg~
z(zxFl{6DqvHlH#}6l&+b%-3NCCXC}J(vgnh6tVZL9of*gRz42+rSL}g!?%w-5V3`h
z<Fxad@~4fh)W;%~e@|NdJJdWGVd6wE-9(HZughOJ{C54CE$rf(Qqh;naJ!6qhs|7)
zv>vBiZzOz!GM}5;p;g=5R)(WxJ(Ti8G-V_Nf)BM&W!kyHG=@nXSQ}caD{ZGMZSO#f
z8;6%sG+U{$<zxAdYELjw1#QJT)X_=1zA0pkRdTy%0`d^uY9?L(0CZ#PP)9etPaKp1
z8ZjL~6!CJ&)}8*dtudkv*w8Y}GX`Atr@%?-YL4O;092l79Wy|S;fsF}p?+k=SB_!p
zQ%2%}i?`*?KW#>=88P0aQ@Vf{-zs9AlS4*ZxT9?_6>_yCbjo~mgg9i>>V8~(-}h=2
z;-$yE09lE>IISyL4(w*KE%WPFt=LB=IX_Aaj<-o|m;~jGb~<9KDz<tI*AJ$*%%pRQ
z>MU-Gw_Y>T+P90)lhSl}{?)XFVg-JI{>6G1li=mK+IP(<B%f;h`|w|xv@S^m{{WL;
zsQ&=6?}yHl@b5_8r5WQ{58^XfR(d2le;Ms+s;i1Tx|e9A+g#RXNfyyb#ZAYhCznNs
zat*ZKEY+Vbo5e0ska=dL;-uS5+f0$0X}^U@OSDoQ$Ts^?x7w%LDB3AlWq&MsQySbi
zYPYmev{7IvwU}<y-?5?UwV||Ae7~g@13UXZHtM3IZ7mzMYHg!!O@PS$$x+coGy4uE
z>Wlf-)(^ECYOH|gTS1Asr9$2tfxqRF{#DcM`c!*;g)0Hb&EYF%>jF2_RT;bqXFuMF
ze+91Tls9TmBzjOjW6h@cYhwQZ^6nS?+M{vsrp``^$@H&PXjQwKYg?;#p<QfCcijGL
z;5&K`!kW$Ib*6cZ8)bU>_Ndvm*I_mjYM!Q<_i?wkN`D@;8n*5_4|+^wwyIyJ6?tST
zq-68fsavoeD#r1Q(gG%L?w`(?8wAdK^`W+^=ApxS&;^D0Mrj-7T+@|HCrXQ*xbHv@
z^A63uv%B%8r6%FGH7em~0!^WMP^=r;qusaennp+3N{*FaT$Kz<$prgVW%<3m=|sf+
z?3x*SD-E?QNNeQ>Jhev5T9=a-KYE$Ekghm3<?Hi)lq5d%)ljuGZJGv1tTB<m_ohB~
z^ds7z9>!?H0uXhnFO9NEw%o{5)xgj5pa>vVPKKo8Z<pGVI!(iVYJm|q%HGtGsM>vd
zRE)~WyBt$hcF;b9o?_<*k?JX)V55bH3V%Md!ObQBIj3^GgUvA-q>J8=09J1F_*s|b
z$II(N@kp{2<FzydyS8=an9?&-BVyfYxI1%AfG%0F(wnsU)19!{6y;sZnm{M;?d?tx
zvmtG~3Tph_>0@@_ZaD8i0<Q0qBXuJ<T<~e?p;Nwpz21?OGK1;Tn3$VstCs0b-RsBJ
zn;*;8nInGx08T!%_`lY3Q9Etv)7GVr?`JdtI<c3j0+D}q*gu^tarvAeIrOHp25vc8
z0I>{$IXM3Sckr)q@VACz@dv>DOsL;1JLN?^whY&WMYW_IvtL$x7qk0+!JTCY2Xt2|
z7og7HTF!eNcx8FD<aU3tmx#PS;r{>w_=er|_icWE+?%Dx`nEte=RaEd%l1uM6!A8!
zvdDr~fdp~36dzo3^skCPWxF`rL)D8e33%{70y_Q`^vA)QC};SAAMc$VVj*@h1&(k9
zRVe#P$_CP2%LyqTY<T%}{{Z+zZ;XOAb&qmmmQ(FsWv0xtLA^*T$Ln3x7KoZ;hUAYa
zSCc8@AB}mYjdU(N!x##NL$!NjwO%t0skG6SUx@aJqkWh!y)*$QL65CE8*?DdS@L%D
zuP2oE6qSlt+t!g=cGd1G!^g&IHBGIWRv5<Sqh;Gud1jT9bu&I@LmYK#IP+05o%=^x
zvlkU{C9%Fw%hQ^ZZ?#UUAE2j0Zg*hS6L5uneqZ-$(#m|k1w|Ar)p8hwi-G#n6J%<4
zHZxY2=BU;Xt(tVo(bb|rPFFeYRfW+f88r>WrJ!QG6H(c#NhM_|a-DIdrP;X|B{S63
zr*Y+GondZWa78WExbh{(;Z%&Y=G47z;KmQFO9Y?1$9kn=+bu&#LWLPQs5Cp%eQ_+A
z%6Xi9D!hs1z1li>+vUYAH5ZK2s*W;1yOEawh8X6&`$4vz!g+1w81m&IdCo<89Ci7p
z{{RIU{iJ>%d@J~&4~4uf;u$n;Hf=Hs**?mn&N4ibgY%J`e80-7H2LMYy|SYhb-DT{
z2`ZR9X_wm7hJ#}*=B0TJg{)tGYljGf@iq8ir~cW086dYd$?*qRF%R#`tbaUK2DR~f
z#lAjPJ|XcIp{s-ZH&-s|eyUC@tx<~VaL*Ll_fhwU?TPzvSa{mj<G`K<yYqC1J2fp;
za-!vP%xV6?^*BD&@)WnP=0`U6V%X;xz#Xw(U1z9yQ{cNCoPth&I_|8l%+{fc0tZUo
z=_@14o0E5u-RSoobU5CB8n<bD*7woJyA{i5_l0BN)2{VTF8vFVr1hqy+~;vRUyB-O
zxYFRdOcE25UNd#MRTyy9@AE_bz2TQYH{oyr>s~9aY42>$8zBt8rA_W>9gcQc!zgZy
zI#)~KZF{0iCKT5pd%8I4D?6sPD9eI9=(x{Xxbm)59Z%<2H<si!<v91FL)E<KQ*m7S
zRhy|3ZoqOqYNEo|F||00jldec;Z0er<mIvTs3%>~=PXa9WXP=(WF53fe5bwX5Zm~Q
z#YAqd<F!$=xV!LHs1<`OC9pn~=Uxi&!{|52r*Ky3j+O2Hb&rfR$m9--FdTO|s*iGd
z>TSymJ|xrp=)jP-Q(kYc=x=8sU9E;V1P&{6MbvJ*J$dCzmw@#htFY9*!>d@ALan%R
zJE``oTJA%-Jl4k4q+G7k?^v3}#FkPtfZ!h0-Y%Mv$JC##avmbF7YIP)j-XXa>LxMx
zj60i#xP-_!$EP*T-$V;=4sp}IYj)m28Z<5PgX>XR>8*DrzyUs|t#e+-Rf}E~*B?!I
z!te)5_5T14_>nG@JFw=wHula<E=1lj^sTK2T$<ltKb35@Gk)v2^wx`g`<=t5y=wW3
zyE{kHxL*YLg)gndZp+Pj4ZX0FECFNf-jTd^F||!rT^$xw>IQhPHt~nXt82*S)26ly
za0V7!{{Z!?(R@v%-rcBJZ6=Tv$I6DhZ(R5@;_Z6z-qXXkvk}pjFc0cQc-Xl;k?7EI
zvbo?sI`F(&ypqonA2LQHkU8eOmr~JYv5<nS6nC$;v_IKD;%>acGoYI!KgV+_kL8-)
zll}?E<9`xHKGiOpYCGgJe?eK&tEQ#R!*tKjhFz<Y4nBsU=DwKt-TMuENBBz#z5dSr
z+0ztd?rftcXVq)x2`%J?MoA-NaBx+I59QnOuDHoaz=Ltyih8hWV=FE`Q-j|In;v~d
zP|HcrT5prTp%lepwB?g+0wQ}>10s`gPPKH%9zH4=BNPD~GH&mRnPn|i{JeCcHZ%cX
zWxXi!&fManRx4Ba-!3Qu6<xITl&`Hw6f)+c#(E6U1c@-I<C3YYdwUv@Nm>Srp|?|&
zgH=_NZ+f0OcPN)0^yT?#5ZkI%`KT3#j%mupy{VC~KHpkaZ%Raxs~_RwuiCcd=dDs$
z%C2hlv#QW8RCGEVfm<2;>!*iq;Ocob#aaBVl{l`aPq^C}bv2@PB1rb{gEp<KYZexb
zw=%NCp!dnee!hG+(c{uQGi3sFiZB=O7_ZBJ*x$pS@RR&X((;?*w*C3^;BjBA8cn#j
z)8bGug~ewkq{mFV<(fRRPokjMk0a8Oo#mQs02G@=EffIVw7W$u6k6a%ww=WrXxmGb
z0L`>*r|qL{OI?6w+BVavcG9x~Iw{^+q}xT94ckWAcG@=4W&?K8Q})vBqRa+usJydN
zQt?HY5pAYboDW)$ifMNOFDUy_<sWL3bsMTc{$?bYbnQs3#!W)ax`xWGGhG5E-QPUY
zFU|LjGjit`H2CBH0A$cIRbl0E)~53Z-Q=3B8x6Rpd4X5u>+e7hZ<%pS%|A5{m^$F*
zffFljkCrr>shB%)OGrpD@?xdkQLoKQxNXe{E<b3ehT5zFZfTpT18(eUu2AjA8Ka7B
z@0X#?F6uxg?~<dX-STuaigz4Th>=%yIxZ+xD^kV>rBNeS`>oU3o8*w$&*4wo#(KRs
zp1;LF!E(I!qKYZpXm9ZK`Dr<&WmQ~fnoZOUk8jJ@=BF+PdR?@hU%eZE_~WHj+fIxe
z1JgJIk=~kHYhch#$N6TNwx!?fYCquJ+L=Jnp49F(;+xN?{OUF(XaHp`NEoNeLCrpt
zHss6o_Mir;g{Q|E^%R?ERx3b--SXm@&hPeVXr>_JILJR*i*fDG(v811Y1tlyev}9o
z2ac;wjYTnkchFRe^7H&u0WKSEcQX&~bfjP1^r_jlt~&d6sF{_U7@!B*?DG!rPo^vD
z{{Y$1R&Ry6kj7UMqWwzsuaKd_wrlJU*>#p*0p8mY@~rQJ>{p7Jd?!{7{`XS9?5}h#
zG<{as5X`1F1a-$h<6lhtC-7@)o+=k6d7<(^!$jo{NzZEeZ}xPy8ZX36e??}@arv<f
z@%LW>zQXud;%GcS;$al+vKxj~Q}3KtP8QL`rz1RbsPHn4J^uhh=(}GE#dW8>wDR0s
zBd*30N>3HZL7<-w>XO*0W{FC&vvt9*ThRhdZXIyIjPYDXt#q1>wJf0s^4l24JRDbM
zK1xxs;^h&XCwH}n+T(69Q??qC);gF+y1sP#G|c;ue7-6l;m#{d$jab~i|qdZH?>bG
zr#W9|%bHKluUo;raB5Go`%~q)9L8#{?m*_Jg*WscomyL6rv++Ddvcvts=>^H3wHFY
zF~m1#6*MTz=~Cfi>sMkHx*>}Q>?&oqRq0QKH+1XiQ|+llU6nkOw>0S#70DH8qub9_
z_N9(BWah5S(KWf0_Goy=6+WG&;y?)G)YlV1r$T1CksyE###s*41Jsk#@vjN^f%|!U
zCHQr8))4Cd0Pv9Ct8Ht2CO<I~>n1|}Myfb@OH_hbn88`I-tM&%qOqEFVQvmSPzkK9
zN5#;&3l*zMJMQE9*PZxJ<JW?|D#n+(e~7JOPzEh7q!BU`>KNj?8+mk#t*Z{9X!18j
zNz|X`QS9X%EHzS;l9w@lYwPV<1-c^J+mrK1obg>2ho-HKx~hK&{c9gjjr9Cu_mLlP
zMQ>T%$0DdwLmtFbrLD`UQt{=E#;U&6#YnCisrlb2VsHmq!|@l6^bZPIwbrq%*lBjj
z_oTYGL{a#YS}3=&INOTqb?~rkVEc6Z>+sw5{P>$6#BbV@#n%?oEcO-+9+_<Dav5M4
zVm-XV*!DH{kL^kS00kV-w0{ih9vt{*b^WcaT_Zx;!h#{UR_e)~OMVskU#whxw|vqw
zJd8=)-EvPPpHtNP*IXmX8@<matzT&_Rk4z+(6bdh8Y9#m2H^2gNZ$2QNdot%xg5RD
zrrzLOg<wuacb*^APm&e(;MbFBw}ux3B!gZ4lXd%)VYCXaVjOIa^2*_3d~%|x>bFc~
zR|h!$^+jtj4WGO{>W-y1&5ES^nDU;-;LUYgNdbumzolUK%SaO>*Ae{Ny6vui!+MV0
zH|2aEPW7#(c#rKcJTH<nkD7?fLhab|TdP9fDPNkRZ7E@hmLE)4W2sr9%o#u=`c&5X
zKieTAZb<AZu1ktX6tT@{i!@o-25VzY)c(~8Qdc9WJ!`VF((TsOt)`7e<-*sVYnn9v
zAHI1b8Ht8TT;h|u*ikmi9r;`jtyR=5mDe?$Ype@}A)5ep?^ofNPMrQ`tC<&LAIsXn
zX9RRL(0F&o7djm4vS!#SjP`2HxjK5)QEoc8P`|BFHa)XR@#Uw8BiSJaPi%jUYgybK
zJ%ADy2e38aIzqjj*O>|0zgp~I@ifuMft&HIBP*jBM?>2zEsmvXHVw>w&MS!VHkW_l
z>yI+(23u)A<*qWdtE2d+*y-tT9|Ifky>nX9d{ol4ofcb5qZ<vFXygErYf0>38yH>^
z_;CI`kPSE|noa`b^ZHXfN$}5D@Ghrpx3;em$Z#96H5bF5iZ<U4yhyXFXHAi^S3IxM
zy|dyU#tje0ULQn_k>m}%9OEi~I_KrN(50cux4?GsSBF-3<UVj2=DaJ#o)^9FPNZ$_
zqjrz@ByIt(r2GT%ZapsDA(fN>HkBr!@xF&+@iNv`fpXu%RAjHxwApA&w2u?;Z-{)B
z8)O^Tu;#v}_$%Vu4MO4wF9J!<Fxan_HE)M{?|^1$F0AG@*vXMs8-BQ{v=1Cfq~1p?
zGZ>^-<O)hxD_9?_UK{ZCk9}&AM{RDy*J!SqKMC0;(y=Cc@n0DH0{x?w;%KaIXHmFu
zl7A}uGeg&-)U4Kc<lGo!8s?ooSE12Lv}d8IZ=u>lw`@u=+XAjFgeu&oN6>$TG8xuP
zVS&%RSytWK6_+W}X6im2vee;dr?-w&Wx))!@<;6f{{RH%@DIgrv}zW1r$p7I<XqUU
z(YxJyiu)yGR_{$RhTygbwI-Ce5%^>AhyDr2@i*bbw>Mg4w0<SiBk3i!R#p0`{{R~K
zgIe&$sp0#hYpChhRu=%{%eaY<{-VF8JF7curC{m^J&rm0@m_i2?|_~R@vMy|tD)Fk
z$Xh$5fl2)FS=w>g+?zCiIw4CO07>+uZ<J(^KBNl$#PP5E6Pw{jiW+$agAI(+e;v3f
zx9Bro1>-;XBz}puNpwFD-`eBu#V`^t=ZX}i)eoupaI$SVImR=L8kje)KT7(O$Di;{
zKZ(8(yqPSl<JRC)@|cKHN8&|z&x*bP{6X+Lk*fGw{@DlowDR{a=~{CoxN{?d^8S?;
z(BF58l4X+GPcBq=-#FUJL8hSKbmyldns>OyZKmz156ed3=|BpsRdG&PJh6~PGNaq`
zdr$>{<-IA}PBT!dm8m@Aw;kvj9#kzoWZO`*p!(DIP&P=cR;P`sw(4@RS_N`+V&8hF
z!2a;6@$}6|vKIakK#OX{qp9E~s>Z`-TDfk@&58ilvdf$b>2$xms^4^eb<Jt{R>@Oc
z?}2<hd*bbS%TCj+-sa}vwyG#UpVFQ5yCz58{{Zk#Z-?_q{h_bT5snuk-X<fc9Z2@C
z(AZOIs`eG(zq4<GW%zgS{{TtTUJ!;TM(hr!y?Z+9O1+w}jIXOIG37>%J?Sc<>Jhhk
zl`i9xqhsq;bXhes%QUMhH}MKw(*eF&rqr~!T3|=znmn^kW()62NCtUkk1W%-QM!N|
z<(h3j%oQH$0Cq>E8)>_#HyQw4)SGE3IiLliiYNg$QM#K&E1`uZ>RN3L*g=y;6jE=n
z5N<SXH7?p;FX>Ze2lF9JRwJ)JN^+glyf59YwE;hiH7c_&@fwbocVF|GU_DzDD{iI+
z%?iO+GzpS<l(`<27V2H6)`6A-2B!IfqE_49k1%=~1U+QhaZBb6nC7DY0G61g+dx>g
zpO-X<$8KsLGv&o9G2QJz$$Z0|-CB*qZfZ4Q{6DQK{L#<?d53N>PgN?}q+B&RL*=$U
z_tJq5SsA-i)<#;1<SWvh<_ZZPxUb!&WYT$q%f~yZH&bN~T(u@mD*fyZew3K`b4W;W
zP~#xecAccr9^JU+mvHjOw@OH%X?FbB$fTx!7HYuhJRNVKNquy6*=}QjZJ5ROcL;pJ
zj-=<%)(p(AJezn!(=^49tR+8);FFB=-jRZ;`_nV!v$*J{{`EDoqZ#6wxE<;3^L>3P
zQD|W1pSU+bX{x10Y3qfih9_*$zcS*F;h}z70ChH%TXiKdFmINMMs3sv%><30F*Lhq
zRUxrJ0`6g+Dbz2@xTM|2Xadr=twpr%&svj@T4T4E*gr4(v;ib+E9!6AUgA&eolSQ(
z#hN{<+mVX-V%|29f~w@2`g8V9+a<oM2J?~SwmzBX^`Wf|u)oDc?*v^fkL?Amplzl}
zq{qHc9DZiLiPCjQw4V~_*B1Z_WV?%TJdAGS*US3Koj3N7kE|&3!!cpN`H%uXO8SRL
zv4d39ZPkbF)yC!Q268H)9%_*~E7bNlYSXp(Q_%bO!&-)#beN_~h(v83K@Ka-(r3EX
z@0_WPT4pQmxy>(wG+2JaZGgWj^e4B{x@}WPTX^A<X4VcAanyIOOBIJvi`MM;T&Fm#
zimeE^S}cmvKbJK+!d0=}ooIogQV!)kz%=QkQ6a&`J!|LMvHDCS+;5O6zh?SW<MO4!
z#!mz4Doe3hR4+kR6b7kQ5Z$QqL}UUO6YOcC8o3JE?ri3t_HBoj&$z2m$GBv3^rexK
z-WHN~Sm05UBQVG3T18xkexgLG)n8E4P2_l43OBV-z43R4bhuhut$$3liGK6Tc^ZNF
zn&b74+H>JQhV9;ML*f>J8ZZ3wyjc(AxizDwF2t&li@ncC(<~a!9kMZJ^(L=*CmF|o
zPM-DgXT;z5DtE)53C8yR7Sr#%TXp<0pY&^Q#gEPZ0FKwopAx^|sb3U4R}7Zg9C~ks
zZQJGFWdV8d{{USapXF5=4%+B*%Mlysu70BYM)<wpzk~K`b>i<3*=W(X+>Uqs>-`XX
z$MUa>zBPZrQnXz+&awD=p&d$PU4v8AGaMlLta;0RIIqp06nN`h@#e8EtFP*pn!ceg
z-?_Y)$sznt^u=Yy<+4UeBc=(#_Q|c{;ZIbq&N<boC2f)FKM_A@PaphNzPY*aABgX-
z{PUA<=dxR%{_4b}E<Ux)=#n|OURdPh4cH%$9M?S$nIixc^IcAdX#0~ZwENQbjlIrh
zoveCRnIUUfq&&N4C!C*5(>2X!P_q$B4cyT>cS#w4tw*Ahv;+4?K~eaU7n<KV&(gH#
zeaxIu)c9NWhxmixMn=>8SE@wo({;l#{{Y`4n(y?_+w0>MikRP9wz?gjYe>oe0MKja
zR99h^Q&O$FgSd<e&)Q#eMJX+r>E95)Y(I-$D2d{m##gq`w^;1RfPdqFSB~Cz!&LDU
z#!X95xxJMA=D3}lx7RtP)-1f=x(#ycy58fQflJy;xbJ<-mr`an000L;j-O6z21yv5
zq#E2=7D!n2%|~Z#9mJS*sB*X*(?G>{cB(T#?AY4X>b9N~j&%&FxSrjq{{Zl>nnv6r
zY<(*SQ6qI1Bg-IGjg*nny8RyD`DzIu3g4UI{F`zi#b-~Yn-$td(@jlFV3Zo%=`@?;
zJUapAn|R;6`Ktp-xO;PeYh@VSKg<nUmWMlux#8=8w)+bHHBNm&?DYoPPQp6-)vG&q
zjz5%V=|tK~%{K6a`*x|haYdVp;?b)P-nFHp>k>sMbsu~0X0yJ@wY;|kEn0;IsTm)8
zt}8_@#7ZXZ^EJXOtMZXoBa=<m^j5dH+4hLn12xXesYt802Nlm;O*Ar@BFN8QTEA$F
z)NQGa;td>Ww~S#wDfH=0()9+Ad2L!IM${K%-F<5o>rXawZjFv=O+8JiC6?ckFg-hc
zDZ|USTxPP9Tni7CIoxx*KJ{HLj8C~6D^#W&w>n5{-WSMYk51Ki$s;N&DEWF~u(Z8+
zfr$?F_pSR|=%ieZYgHpB9_LLCr?*Qn;OA}tt`2D~ZEqwYn1<+0bui{g!C#o4N^@u_
zeHhwu=7~s3DB<pJB$)i`mhRoE?vdgPiy0Mp5sulezCVYO+$k}rVm;_zg*Venffhd+
z!`cooeG4{zDvISm-L^tL_q}~5`#$S7o@#xe)DzcnubplF5NgV)u^6}3yI%_UGsIdP
zs?e;24|7!-py~EL;qf+)p?Hfzk|`sNw<oaOUl;sS_#1tr-Az80H^LhtH9UTG=~@P{
z;&|Z!$z=<G8I5#PUtC?<l(|JD_4#vKDaJ&&$@D%+(sk`m!#AnsiJ+P@xHfy&)gKH0
z0Bg&wD%Klo!oVuvOSqHyR|DfO4B6^}Mi%S1^=x3*i|X1f=9hX+v@Qo_-bnsby`)pz
zN`reVGxVFnzqMACsO>XMk>lV00Iyz^rTlHuH4nLU8-Kv7@?%Za^=LU%bU#Y(d@th9
z5kWB8(yE$%+L^3rR$gc7L_ZQOB*xj4=cgj2Umn{x{InmZ_}9gr3Gqg+BvJjKzTwlF
z?^nc=tU2Cu^{D5AEkb=yra$32w*LU0i~3Rj015TC{PZ8Ce8hZHI{yF}{OY8>Ex7FL
zKN=LN^&5=Ot4EKl+!AJ9K9sNHdkHrKB7lAC<S9I3c8b6=^I7*EIlo?#H|I~9pK@(e
zq4pfUA=0jg%+av_0BZ)YHOns#YZmPs=>)+401JRce2I7QCs&1Vj>qw=slF`hQGCKi
z&(@34>cf_-x*nVHTlQM`b>n4rxY6yr!Z>)qR?pN|#eWumVjEo=Lq3Ie%Mts~xENoh
zdzH_|?P?XlcQwWMgW~q9c^j<J`MbJ}zx{e#)H*VKole$1Bk?!GJ!8Um6D7n2&~z$E
zu28EhV5<uHW5*hm#<-<~1sL@N99N5K9t*#jw=4_=bIMR#Gp#~))f|7#Mn!M8gY^Ym
z`AmM5S-dfRxc>k~bDv6{X!jo^yBWOWG!AP)v>WtX8)y8fi=$qSw$cx%r_B*^zQrn+
zBh*yC*>_mal*8~e#?x+9vE)hlP^6gb!!*8NrE43d{{WUR^}(fUNtJS9f6ATFHe{<J
zs^W!_maCX#Ua`3RDnT2ZaCxcbUBb+_8vgFJEn{809JVu5EuflwDnla^=(Xt|4gMHw
zUNh8ft*z~5iCy;Z*lBx6ts4s!yer}R&k$ZiX=entR}&m0aV|-()34Z{_E3|+_DSL$
ze%${6YECySF2CxR+m5x<{>q=S&xU>-={EP~$!66gAH1Gr$<MGoE9;93NUb9a2moUo
z)T_@|ur%+e=xov$l>it629xZmR%zQsRglc))5l{=_5!^PXl)cx%7$h&w7rcOxc>RA
zeZNYNX!NDZ24~u?eJNwqZddOjn$q&q`?TLFeJNRiPy11q?<G6?Sv^Rp+SwbmL-q*U
z?<o8z_l7xj7@PR&Rk%j}G5OV2(_`$l6KQYTB9}CfXlVyl=xE#c);Y7CMq_-BttQ);
z*KmK#(aqfs#=`vFsH}gsUB~<-zug9&+Pvx32K<FSb<{?ye6(h>S6Uy$hw`VcsRMr=
zKDAuP5q*AYDhThOJtPLC6DxEyz$tdq{!`HAlBUQt8A-OC_|U^|dUm*mQBo>~?@Q&M
RN;yyw?WLvLM%rM1|JlZyBgy~(

literal 0
HcmV?d00001

diff --git a/docs/search_index.json b/docs/search_index.json
index 693448a..1763269 100644
--- a/docs/search_index.json
+++ b/docs/search_index.json
@@ -1 +1 @@
-[["index.html", "MI² MI².AI", " MI² MI².AI MI².AI is a group of mathematicians and computer scientists who love to play with predictive models. We are spread between Warsaw University of Technology and University of Warsaw. Here we have workshops and seminars, here we are forging new ideas, creating tools, solving problems, doing consulting and sharing our positive attitude. Feel free to jump in. Mission Machine learning is like atomic energy. We develop leaders, skills, methods, tools and good practices so that predictive models can be deployed responsibly and sustainably. Vision MI² is a group of experts supporting global initiatives aimed at responsible and sustainable machine learning. We support the development of future leaders of responsible machine learning through internships, PhDs, postdoctoral fellowships and so on. We seek for research grants and business projects to conduct both scientific and applied research. We develop and maintain the software and infrastructure necessary to build responsible and sustainable ML. We develop cooperation with international teams working on similar topics. We support companies to implement best practices related to responsible modelling in their operation. We conduct workshops and training on responsible predictive modelling. "],["the-team.html", "The Team", " The Team Members Przemysław Biecek, PhD, DSc (Team Leader) Hubert Baniecki, PhD student Mustafa Cavus, PhD Maciej Chrabąszcz, MSc student Mateusz Grzyb, MSc student Stanisław Giziński, MSc student Weronika Hryniewska-Guzik, PhD student Piotr Komorowski, MSc student Mateusz Krzyziński, MSc student Tymoteusz Kwieciński, BSc student Stanisław Łaniewski, PhD student Piotr Piątyszek, BSc student Hubert Ruczyński, MSc student Barbara Rychalska, PhD student Nuno Sepúlveda, PhD Bartek Sobieski, MSc student Mikołaj Spytek, MSc student Tomasz Stanisławek, PhD Hoang Thien Ly, BSc student Paulina Tomaszewska, PhD student Piotr Wilczyński, BSc student Emilia Wiśnios, MSc student Paweł Wojciechowski, BSc Katarzyna Woźnica, PhD student Vladimir Zaigrajew, PhD student Artur Żółkowski, BSc student Collaborators Mariusz Adamek, Prof, MD Przemysław Bombiński, PhD, MD André Fonseca, PhD student Katarzyna Kobylińska, PhD student Anna Kozak, MSc Marcin Luckner, PhD João Malato, PhD student Bartek Pieliński, PhD, DSc Hanna Piotrowska, MA Elżbieta Sienkiewicz, PhD Julian Sienkiewicz, PhD Adrian Stańdo, MSc student Patryk Szatkowski, PhD student, MD Jakub Wiśniewski, MSc student Alumni Piotr Czarnecki, MSc Alicja Gosiewska, MSc Adrianna Grudzień, BSc Maria Kałuska, BSc Marcin Kosiński, MSc Adam Kozłowski, MSc Wojciech Kretowicz, BSc Michał Kuźba, MSc Szymon Maksymiuk, BSc Tomasz Mikołajczyk, PhD Katarzyna Pękala, MSc Adam Rydelek, BSc Bartosz Sawicki, BSc Patryk Słowakiewicz, BSc Michał Sokólski, MSc Mateusz Stączek, BSc Szymon Szmajdziński, BSc Zuzanna Trafas, BSc Kinga Ułasik, BSc Anna Wróblewska, PhD Hanna Zdulska, BSc Przemysław Biecek My personal mission is to enhance human capabilities by supporting them through access to data-driven and knowledge-based predictions. I execute it by developing methods and tools for responsible machine learning, trustworthy artificial intelligence and reliable software engineering. I work as an associate professor at Warsaw University of Technology and the University of Warsaw. I graduated in software engineering and mathematical statistics and now work on model visualisation, explanatory model analysis, predictive modelling and data science for healthcare. In 2016, I formed the research group MI² which develops methods and tools for predictive model analysis. Google Scholar: Af0O75cAAAAJ GitHub: pbiecek LinkedIn: pbiecek Mariusz Adamek I work at two Medical Universities (Silesia and Gdańsk) holding a Professorship in Medicine and Health Sciences. My interests are focused on lung cancer prevention and screening, the latter by means of low-dose computed tomography (LDCT) with special emphasis put on molecular biology methods, prediction models and image analysis aimed to enhance the performance of lung screening outcomes. Website: www.mariuszadamek.io Hubert Baniecki I’m a PhD student in Computer Science at the University of Warsaw. Prior, I did my MSc (2022) and BSc (2021) in Data Science at Warsaw University of Technology. My main research interest is explainable machine learning, with particular emphasis on adversarial attacks &amp; explanation evaluation. Website: hbaniecki.com Mustafa Cavus I work as an assistant professor at Warsaw University of Technology and the Eskisehir Technical University. I joined the MI² DataLab as a post-doc researcher in 2021. I work on explainable artificial intelligence and AutoML. Google Scholar: I63d1WIAAAAJ&amp;hl GitHub: mcavus LinkedIn: mcavus Twitter: mcavus Julian Sienkiewicz I work as an assistant professor at Faculty of Physics, WUT. My main research area links with sociophysics, complex networks and agent-based models. In the scope of MI² DataLab I follow my other interest - scientometrics. Google Scholar: mIwu11QAAAAJ LinkedIn: julek-sienkiewicz-873829 Maciej Chrabąszcz Master’s student in mathematical statisctics at Warsaw University of Technology. Interested in deep learning on text and images, explainable and responsible AI. GitHub: maciejchrabaszcz LinkedIn: maciej-chrabaszcz Stanisław Giziński A Research Software Engineer and student of Machine Learning at Faculty of Mathematics Informatics and Mechanic, University of Warsaw. His work in the lab focuses on using natural language processing and network analysis to better understand the spread of AI public policies. Interested also in applying machine learning in bioinformatics. Google Scholar: Stanisław Giziński GitHub: Gizzio LinkedIn: stanislaw-gizinski Mateusz Grzyb MSc student in Data Science at Warsaw University of Technology. Interested in artificial intelligence and scientific computing, but above all simply enjoys programming. GitHub: mgrzyb99 Weronika Hryniewska-Guzik PhD candidate in computer science at Warsaw University of Technology. Interested in deep learning modelling on medical images in the context of explainability and responsible AI. Google Scholar: aJeg3IQAAAAJ GitHub: Hryniewska LinkedIn: weronikahryniewska Piotr Komorowski Master’s student in Machine Learning at the University of Warsaw. Mainly interested in image processing and XAI applied to medical images. GitHub: piotr-komorowski LinkedIn: Piotr-Komorowski Anna Kozak Graduated in mathematical statistics at Warsaw University of Technology. Interested in explainable artificial intelligence and data visualization. Organizes projects related to education. Google Scholar: JIrqf9kAAAAJ GitHub: kozaka93 LinkedIn: kozakanna Mateusz Krzyziński MSc student in Data Science at Warsaw University of Technology. Interested in explainable artificial intelligence, with particular emphasis on XAI methods for survival analysis models and XAI applications in the medical field. Also an enthusiast of data visualization. Google Scholar: i_r7EUgAAAAJ GitHub: krzyzinskim LinkedIn: krzyzinskim Tymoteusz Kwieciński BSc student in Data Science at Warsaw University of Technology. Particularly interested in explainable artificial intelligence, computer vision and NLP. GitHub: Fersoil LinkedIn: Tymoteusz-Kwieciński Stanisław Łaniewski PhD student in Quantitative Psychology and Economics at University of Warsaw, Machine Learning Researcher at MI2 Data Lab, Msc in Actuarial Science and Mathematical Finance at University of Amsterdam, former Quantitative Researcher at Flow Traders His research focuses on enhancing classical methods used in discrete choice and finance with machine learning and how to apply them to explain behavioral phenomena and heuristics. He is also keen on finding balance between best predictive models and their explainability. Avid gamer who applies statistical techniques to deepen the understanding of best strategies LinkedIn: Stanisław-Łaniewski Piotr Piątyszek Undergraduate Data Science student at Warsaw University of Technology. Works as a research software engineer on enhancing accessibility and completeness of explainable AI. During pandemic contributes to a system of monitoring covid variants. Github: piotrpiatyszek Bartosz Pieliński I am an Assistant Professor at the Faculty of Political Science and International Studies at Warsaw University. I am interested in applying quantitative methods to study public policies. I am a founding member of the Institutional Grammar Research Initiative, which is focused on developing a new way of analysing social rules. I have participated in several research projects covering social policy, non-profit organizations, social enterprises, and international organizations. Website: https://pielinski.info/ Google Scholar: hnWiaVEAAAAJ LinkedIn: Bartosz Pieliński Hanna Piotrowska Information designer, focusing mainly on data visualization, branding and book design, with a strong interest in Data Science and perception studies. Winner of numerous awards, including The Kantar Information Is Beautiful Awards, HOW International Design Awards, Polish Graphic Design Awards and KTR. LinkedIn: hanna-piotrowska Twitter: hannapio Behance: hannapio. Hubert Ruczyński I am working towards Masters’s degree in Data Science at Warsaw University of Technology. I am also teaching students about data exploration and visualisation. My major interests are: AutoML | Natural Language Processing | Data Visualization | Fairness. GitHub: HubertR21 LinkedIn: Hubert Ruczyński Barbara Rychalska PhD candidate in computer science at Warsaw University of Technology. Mainly interested in deep learning for natural language processing (NLP), recommender systems and graph-based learning. Google Scholar: Wp0wHJoAAAAJ LinkedIn: Barbara-Rychalska Bartek Sobieski MSc student in Data Science at Warsaw University of Technology. Interested in deep learning and hyperparameter optimization. GitHub: sobieskibj LinkedIn: Bartłomiej-Sobieski Mikołaj Spytek MSc student in Data Science at Warsaw University of Technology. Interested in explainable artificial intelligence, data vizualization and survival analysis. Google Scholar: 1u49AqYAAAAJ GitHub: mikolajsp LinkedIn: Mikołaj-Spytek Tomasz Stanisławek PhD candidate in computer science at Warsaw University of Technology. Mainly interested in deep learning for natural language processing (NLP). Google Scholar: gq8NY_UAAAAJ GitHub: tstanislawek LinkedIn: Tomasz-Stanisławek Paulina Tomaszewska PhD candidate in Computer Science at Warsaw University of Technology. Gained experience in AI at leading universities during: Deep Learning Summer School at Tsinghua University (China), one-semester exchange at Nanyang Technological University (Singapore) and research internships at Gwangju Institute of Science and Technology (South Korea) and Institute of Science and Technology (Austria). Mainly interested in Deep Learning, Computer Vision and Transfer Learning. Recently, focused on digital pathology. Google Scholar: eO245iMAAAAJ LinkedIn: paulina-tomaszewska Hoang Thien Ly Bachelor student in Maths and Data Analysis at Warsaw University of Technology. Interested in working with data, and learning explainable artificial intelligence methods. Google Scholar: JkysewYAAAAJ GitHub: lhthien09 LinkedIn: hthienly Piotr Wilczyński BSc student in Data Science at Warsaw University of Technology. Interested in Large Language Models, AI Deception and Natural Language Processing. Currently working on my thesis, which applies Computer Vision to medicine. GitHub: wi1lku LinkedIn: Piotr-Wilczyński Jakub Wiśniewski Research Software Engineer and third year Data Science student at Warsaw University of Technology. Developer of tools for bias detection and fairness. Currently researching responsible applications of deep learning. President of Data Science Science Club at WUT. Google Scholar: _6eQsXMAAAAJ GitHub: jakwisn LinkedIn: jakwisn Emilia Wiśnios Research Software Engineer and student of Machine Learning at Faculty of Mathematics, Informatics and Mechanics, University of Warsaw. Interested in natural language processing and reinforcement learning. GitHub: emiliawisnios LinkedIn: emilia-wisnios Paweł Wojciechowski Graduated with a bachelor’s degree in Data Science from Warsaw University of Technology. Interested in explainable artificial intelligence, computer vision, and active learning. GitHub: p-wojciechowski LinkedIn: wojciechowski-p Katarzyna Woźnica PhD candidate in computer science at Warsaw University of Technology. Graduated in mathematical statistics. Interested in automated machine learning especially in hyperparameter tuning for tabular data. Carrying statistical analysis and predictive modelling for healthcare. Google Scholar: tAQS1gQAAAAJ GitHub: woznicak LinkedIn: woznicak Vladimir Zaigrajew PhD candidate in computer science at Warsaw University of Technology. Interested in deep learning, primarily on images, with a focus on representation learning. GitHub: WolodjaZ LinkedIn: vladimir-zaigrajew Artur Żółkowski BSc student in Data Science at Warsaw University of Technology. Interested in explainable artificial intelligence, computer vision and NLP. GitHub: arturzolkowski LinkedIn: Artur-Żółkowski "],["open-positions.html", "Open Positions", " Open Positions We have an open call for HOMER and xLungs projects for following positions. If you are interested in any of them please send your CV and Motivation Letter to przemyslaw.biecek at pw.edu.pl. We reserve the right to contact only selected candidates. MI2 policy on benefits and obligations We are seeking highly motivated and talented researchers at different career levels to join MI2.AI research team in the exciting field of explainability and red-teaming of AI models. There is a common recruitment policy for all positions. We are looking for people interested in research who want to create methods and tools that have a real impact (not yet another paper). proactive people with initiative. people with proficiency in programming languages commonly used in AI research, at least one out of Python, R or Julia. solid understanding of AI/ML models, including their architecture and training processes. good communication skills and proficiency in academic writing; excellent problem-solving skills. We offer: the opportunity to learn from experienced and motivated researchers at different career levels as well as the opportunity to work with a rich network of international and interdisciplinary collaborators (for example Harvard Medical University, Vienna University, LMU, NTU, KDD Lab, Bari University and many others). a possibility for a tenure-track position after the completion of the postdoc. possibility to present research results at international conferences. access to our supercomputers, among them Eden with 32x A100 cards and 1.5TB of storage, a perfect place for training large DNN models. Salary Salary depend on experience and type of project (applied projects tend to have higher salaries than typical research projects). Average rates range from 50-60 PLN/hour (equivalent to 8-9.5k FTE) for a student with experience, through to 80-85 PLN/hour (equivalent to 12-13k FTE) for an experienced PhD student. The base salary for a post-doc (no obligatory teaching duties) is 120-140k gross per year (this is a base salary, extras depend on activity). Low commitment projects (1/5 FTE) in 2023 Ex 1: Any object can be embedded with an appropriately trained neural network. But can objects be embedded using tree-based models? This project (an ideal candidate for MSc thesis) is about exploring different embedding construction techniques based on random forest or boosting methods. Ex 2: Analysis of champion-challenger model pairs on subsets of observations obtained using data segmentation. The project consists of an extension of the EPP method (Elo for predictive models) based on segmented data (e.g. by k-means or by random splits). In addition to the performance analysis, a PD profile comparison analysis is also performed on the splits. Mid commitment projects (2-2.5/5 FTE) in 2023 Ex 1: Development of techniques for explaining segmentation models for 1d (feature-producing models, e.g. survival models), 2d (image-producing models) or 3d data. (CT, hyperspectral images) Experience and knowledge of classical explanatory techniques (SHAP, LRP, CRP) are desirable. Ex 2: Development of red-teaming techniques for LLaMA 2-type language models. Development of test suites testing the vulnerability of the model to a set of prompting-based attacks. Full-time positions for post-docs in 2023 In addition to the research tasks listed above, responsibilities include mentoring junior staff (students, PhD students) and leading small research groups. Additional information about costs of living in Warsaw From FENS: Warsaw is the capital of Poland, so in principle it is one of the most expensive cities in the country but living in Warsaw is still much cheaper than in the majority of European cities. In general, a single person in Warsaw is able to cover all necessary expenses (flat, media, transport, food) with ca. 5000-5500 PLN per month. A typical cost of renting a 2-room flat (30-40 square meters) varies between 2000-3000 PLN, depending on the part of the city. Electricity and media (TV, Internet) are usually not included in the rent and have to be paid separately. The cost of the lunch in the vicinity of the main campus of WUT is ca. 25-40 PLN, while a monthly ticket for public transport is 110 or 180 PLN depending on the area of validity. Public health insurance is included in the post (this includes also admission to hospitals), however the medicines are paid separately. Research Software Engineer In addition to research positions, we also offer positions for skilled programmers. Required: Background in Computer Science, Mathematics, Statistics or similar. Experience in Computer Vision or NLP models Scope of work: Training of machine learning models for tabular and image data Interpretable solutions for tabular and image data Responsible ML solutions for the healthcare domain "],["contact.html", "Contact", " Contact Our rooms: 44 (DataLab - separate entrance in front of the main entrance) 316 (xLungs) 317 (HOMER) Faculty of Mathematics and Information Science, Warsaw University of Technology, Koszykowa 75, 00-662 Warszawa VAT: PL 5250005834 "],["mi2redteam.html", "MI²RedTeam", " MI²RedTeam MI²RedTeam analyses machine and deep learning predictive models through the lens of AI explainability, fairness, security and human trust. We develop methods and tools for explanatory model analysis and apply them in practice. MI²RedTeam is a group of researchers experienced in XAI who perform a rigorous evaluation of AI solutions in order to improve their transparency and security. We apply state-of-the-art methods and introduce new ones to tailor our analysis to the specific predictive task. We openly collaborate on various topics related to explainable and interpretable machine learning. Feel free to reach out to us with research ideas and development opportunities. We help organizations to better understand the vulnerabilities of their AI systems, and take steps to mitigate them. Our current core research topics of interest include: [ARES] Attack-Resistant Explanations towards Secure AI, i.e. a critical evaluation of the state-of-the-art analysis techniques [xSurvival] Explanatory analysis of machine learning survival models [Large Model Analysis] Explanatory analysis of large models, e.g. transformers Methods and methodologies introduced by our team: Evaluating explanations of vision transformers InteractiveEMA towards human-model interaction in explainable machine learning for tabular data SurvSHAP(t) for time-dependent analysis of machine learning survival models LIMEcraft for human-guided visual explanations of deep neural networks Fooling PD &amp; Manipulating SHAP for stress-testing widely-applied explanation methods Checklist towards responsible deep learning on medical images SAFE for lifting interpretability-performance trade-off via automated feature engineering WildNLP for stress-testing deep learning models in NLP Explanatory Model Analysis towards comprehensive examination of predictive models Tools developed by our team: DALEX, breakDown, auditor &amp; modelStudio for explainable machine learning in R dalex for explainable and fair machine learning in Python survex dedicated to explaining machine learning survival models fairmodels for fairness analysis of machine learning classification models Applications supported by our team: In medicine, we analyzed hundreds of models predicting among others: survival in uveal melanoma eye cancer, survival in sepsis, type of lung cancer, lung cancer risk in screening, lung cancer mortality, COVID-19 mortality, hospital length of stay, progression of Alzheimer’s disease. In credit scoring, we analyzed the transparency, auditability, and explainability of machine learning models. In football analytics, we analyzed expected goal models for performance analysis. … This initiative is generously supported by the following institutions. "],["mi²research.html", "MI²Research", " MI²Research On a mission to responsibly build machine learning predictive models. Responsible and sustainable predictive modelling is still a new and developing area. We are conducting a number of studies in this domain that examine predictive models applied to tabular data, computer vision or natural language processing models. We investigate the stability and robustness of various methods, work on explainability and transparency for simple and complex models. As part of our this effort, we develop open source software packages (usually in R and Python) for model explanatory analysis, publish scientific articles describing new methods or investigating properties of already known methods, and create educational materials, recommendations and examples of application in specific domains. If you want to find out more about what we are working on, check out our seminar, which is always open to those interested in responsible and sustainable data science. "],["papers.html", "Papers", " Papers Consolidated learning: a domain-specific model-free optimization strategy with validation on metaMIMIC benchmarks Katarzyna Woźnica, Mateusz Grzyb, Zuzanna Trafas, Przemysław Biecek Machine Learning (2023) This paper proposes a new formulation of the tuning problem, called consolidated learning, more suited to practical challenges faced by model developers, in which a large number of predictive models are created on similar datasets. We show that a carefully selected static portfolio of hyperparameter configurations yields good results for anytime optimization, while maintaining the ease of use and implementation. We demonstrate the effectiveness of this approach through an empirical study for the XGBoost algorithm and the newly created metaMIMIC benchmarks of predictive tasks extracted from the MIMIC-IV medical database. Towards Evaluating Explanations of Vision Transformers for Medical Imaging Piotr Komorowski, Hubert Baniecki, Przemysław Biecek CVPR Workshop on Explainable AI for Computer Vision (2023) This paper investigates the performance of various interpretation methods on a Vision Transformer (ViT) applied to classify chest X-ray images. We introduce the notion of evaluating faithfulness, sensitivity, and complexity of ViT explanations. The obtained results indicate that Layerwise relevance propagation for transformers outperforms Local interpretable model-agnostic explanations and Attention visualization, providing a more accurate and reliable representation of what a ViT has actually learned. Hospital Length of Stay Prediction Based on Multi-modal Data towards Trustworthy Human-AI Collaboration in Radiomics Hubert Baniecki, Bartlomiej Sobieski, Przemysław Bombiński, Patryk Szatkowski, Przemysław Biecek International Conference on Artificial Intelligence in Medicine (2023) To what extent can the patient’s length of stay in a hospital be predicted using only an X-ray image? We answer this question by comparing the performance of machine learning survival models on a novel multi-modal dataset created from 1235 images with textual radiology reports annotated by humans. We introduce time-dependent model explanations into the human-AI decision making process. For reproducibility, we open-source code and the TLOS dataset at this URL. SurvSHAP(t): Time-dependent explanations of machine learning survival models Mateusz Krzyziński, Mikołaj Spytek, Hubert Baniecki, Przemysław Biecek Knowledge-Based Systems (2023) In this paper, we introduce SurvSHAP(t), the first time-dependent explanation that allows for interpreting survival black-box models. The proposed methods aim to enhance precision diagnostics and support domain experts in making decisions. SurvSHAP(t) is model-agnostic and can be applied to all models with functional output. We provide an accessible implementation of time-dependent explanations in Python at this URL. The grammar of interactive explanatory model analysis Hubert Baniecki, Dariusz Parzych, Przemyslaw Biecek Data Mining and Knowledge Discovery (2023) This paper proposes how different Explanatory Model Analysis (EMA) methods complement each other and discusses why it is essential to juxtapose them. The introduced process of Interactive EMA (IEMA) derives from the algorithmic side of explainable machine learning and aims to embrace ideas developed in cognitive sciences. We formalize the grammar of IEMA to describe human-model interaction. We conduct a user study to evaluate the usefulness of IEMA, which indicates that an interactive sequential analysis of a model may increase the accuracy and confidence of human decision making. Climate Policy Tracker: Pipeline for automated analysis of public climate policies Artur Żółkowski, Mateusz Krzyziński, Piotr Wilczyński, Stanisław Giziński, Emilia Wiśnios, Bartosz Pieliński, Julian Sienkiewicz, Przemysław Biecek NeurIPS Workshop on Tackling Climate Change with Machine Learning (2022) In this work, we use a Latent Dirichlet Allocation-based pipeline for the automatic summarization and analysis of 10-years of national energy and climate plans (NECPs) for the period from 2021 to 2030, established by 27 Member States of the European Union. We focus on analyzing policy framing, the language used to describe specific issues, to detect essential nuances in the way governments frame their climate policies and achieve climate goals. Explainable expected goal models for performance analysis in football analytics Mustafa Cavus, Przemyslaw Biecek International Conference on Data Science and Advanced Analytics (2022) The expected goal provides a more representative measure of the team and player performance which also suit the low-scoring nature of football instead of the score in modern football. This paper proposes an accurate expected goal model trained on 315,430 shots from seven seasons between 2014-15 and 2020-21 of the top-five European football leagues. Moreover, we demonstrate a practical application of aggregated profiles to explain a group of observations on an accurate expected goal model for monitoring the team and player performance. Multi-omics disease module detection with an explainable Greedy Decision Forest Bastian Pfeifer, Hubert Baniecki, Anna Saranti, Przemyslaw Biecek, Andreas Holzinger Scientific Reports (2022) In this work, we demonstrate subnetwork detection based on multi-modal node features using a novel Greedy Decision Forest (GDF) with inherent interpretability. The latter will be a crucial factor to retain experts and gain their trust in such algorithms. To demonstrate a concrete application example, we focus on bioinformatics, systems biology and particularly biomedicine, but the presented methodology is applicable in many other domains as well. Our proposed explainable approach can help to uncover disease-causing network modules from multi-omics data to better understand complex diseases such as cancer. Interpretable meta-score for model performance Alicja Gosiewska, Katarzyna Woźnica, Przemysław Biecek Nature Machine Intelligence (2022) Elo-based predictive power (EPP) meta-score that is built on other performance measures and allows for interpretable comparisons of models. Differences between this score have a probabilistic interpretation and can be compared directly between data sets. Furthermore, this meta-score allows for an assessment of ranking fitness. We prove the properties of the Elo-based predictive power meta-score and support them with empirical results on a large-scale benchmark of 30 classification data sets. Additionally, we propose a unified benchmark ontology that provides a uniform description of benchmarks. fairmodels: a Flexible Tool for Bias Detection, Visualization, and Mitigation in Binary Classification Models Jakub Wiśniewski, Przemyslaw Biecek The R Journal (2022) This article introduces an R package fairmodels that helps to validate fairness and eliminate bias in binary classification models quickly and flexibly. It offers a model-agnostic approach to bias detection, visualization, and mitigation. The implemented functions and fairness metrics enable model fairness validation from different perspectives. In addition, the package includes a series of methods for bias mitigation that aim to diminish the discrimination in the model. The package is designed to examine a single model and facilitate comparisons between multiple models. A robust framework to investigate the reliability and stability of explainable artificial intelligence markers of Mild Cognitive Impairment and Alzheimer’s Disease Angela Lombardi, Domenico Diacono, Nicola Amoroso, Przemysław Biecek, Alfonso Monaco, Loredana Bellantuono, Ester Pantaleo, Giancarlo Logroscino, Roberto De Blasi, Sabina Tangaro, Roberto Bellotti Brain Informatics (2022) In this work, we present a robust framework to (i) perform a threefold classification between healthy control subjects, individuals with cognitive impairment, and subjects with dementia using different cognitive indexes and (ii) analyze the variability of the explainability SHAP values associated with the decisions taken by the predictive models. We demonstrate that the SHAP values can accurately characterize how each index affects a patient’s cognitive status. Furthermore, we show that a longitudinal analysis of SHAP values can provide effective information on Alzheimer’s disease progression. LIMEcraft: handcrafted superpixel selection and inspection for Visual eXplanations Weronika Hryniewska, Adrianna Grudzień, Przemysław Biecek Machine Learning (2022) LIMEcraft enhances the process of explanation by allowing a user to interactively select semantically consistent areas and thoroughly examine the prediction for the image instance in case of many image features. Experiments on several models show that our tool improves model safety by inspecting model fairness for image pieces that may indicate model bias. The code is available at: this URL. Fooling Partial Dependence via Data Poisoning Hubert Baniecki, Wojciech Kretowicz, Przemyslaw Biecek ECML PKDD (2022) We showcase that PD can be manipulated in an adversarial manner, which is alarming, especially in financial or medical applications where auditability became a must-have trait supporting black-box machine learning. The fooling is performed via poisoning the data to bend and shift explanations in the desired direction using genetic and gradient algorithms. Manipulating SHAP via Adversarial Data Perturbations (Student Abstract) Hubert Baniecki, Przemyslaw Biecek AAAI Conference on Artificial Intelligence (2022) We introduce a model-agnostic algorithm for manipulating SHapley Additive exPlanations (SHAP) with perturbation of tabular data. It is evaluated on predictive tasks from healthcare and financial domains to illustrate how crucial is the context of data distribution in interpreting machine learning models. Our method supports checking the stability of the explanations used by various stakeholders apparent in the domain of responsible AI; moreover, the result highlights the explanations’ vulnerability that can be exploited by an adversary. A Signature of 14 Long Non-Coding RNAs (lncRNAs) as a Step towards Precision Diagnosis for NSCLC Anetta Sulewska, Jacek Niklinski, Radoslaw Charkiewicz, Piotr Karabowicz, Przemyslaw Biecek, Hubert Baniecki, Oksana Kowalczuk, Miroslaw Kozlowski, Patrycja Modzelewska, Piotr Majewski et al. Cancers (2022) The aim of the study was the appraisal of the diagnostic value of 14 differentially expressed long non-coding RNAs (lncRNAs) in the early stages of non-small-cell lung cancer (NSCLC). We established two classifiers. The first recognized cancerous from noncancerous tissues, the second successfully discriminated NSCLC subtypes (LUAD vs. LUSC). Our results indicate that the panel of 14 lncRNAs can be a promising tool to support a routine histopathological diagnosis of NSCLC. dalex: Responsible Machine Learning with Interactive Explainability and Fairness in Python Hubert Baniecki, Wojciech Kretowicz, Piotr Piątyszek, Jakub Wiśniewski, Przemyslaw Biecek Journal of Machine Learning Research (2021) We introduce dalex, a Python package which implements a model-agnostic interface for interactive explainability and fairness. It adopts the design crafted through the development of various tools for explainable machine learning; thus, it aims at the unification of existing solutions. This library’s source code and documentation are available under open license at this URL. Checklist for responsible deep learning modeling of medical images based on COVID-19 detection studies Weronika Hryniewska, Przemysław Bombiński, Patryk Szatkowski, Paulina Tomaszewska, Artur Przelaskowski, Przemysław Biecek Pattern Recognition (2021) Our analysis revealed numerous mistakes made at different stages of data acquisition, model development, and explanation construction. In this work, we overview the approaches proposed in the surveyed Machine Learning articles and indicate typical errors emerging from the lack of deep understanding of the radiography domain. The final result is a proposed checklist with the minimum conditions to be met by a reliable COVID-19 diagnostic model. Towards explainable meta-learning Katarzyna Woźnica, Przemyslaw Biecek ECML PKDD Workshop on eXplainable Knowledge Discovery in Data Mining (2021) To build a new generation of meta-models we need a deeper understanding of the importance and effect of meta-features on the model tunability. In this paper, we propose techniques developed for eXplainable Artificial Intelligence (XAI) to examine and extract knowledge from black-box surrogate models. To our knowledge, this is the first paper that shows how post-hoc explainability can be used to improve the meta-learning. Prevention is better than cure: a case study of the abnormalities detection in the chest Weronika Hryniewska, Piotr Czarnecki, Jakub Wiśniewski, Przemysław Bombiński, Przemysław Biecek CVPR Workshop on “Beyond Fairness: Towards a Just, Equitable, and Accountable Computer Vision” (2021) In this paper, we analyze in detail a single use case - a Kaggle competition related to the detection of abnormalities in X-ray lung images. We demonstrate how a series of simple tests for data imbalance exposes faults in the data acquisition and annotation process. Complex models are able to learn such artifacts and it is difficult to remove this bias during or after the training. Simpler is better: Lifting interpretability-performance trade-off via automated feature engineering Alicja Gosiewska, Anna Kozak, Przemysław Biecek Decision Support Systems (2021) We propose a framework that uses elastic black boxes as supervisor models to create simpler, less opaque, yet still accurate and interpretable glass box models. The new models were created using newly engineered features extracted with the help of a supervisor model. We supply the analysis using a large-scale benchmark on several tabular data sets from the OpenML database. The first SARS-CoV-2 genetic variants of concern (VOC) in Poland: The concept of a comprehensive approach to monitoring and surveillance of emerging variants Radosław Charkiewicz, Jacek Nikliński, Przemysław Biecek, Joanna Kiśluk, Sławomir Pancewicz, Anna Moniuszko-Malinowska, Robert Flisiak, Adam Krętowski, Janusz Dzięcioł, Marcin Moniuszko, Rafał Gierczyński, Grzegorz Juszczyk, Joanna Reszeć Advances in Medical Sciences (2021) This study shows the first confirmed case of SARS-CoV-2 in Poland with the lineage B.1.351 (known as 501Y.V2 South African variant), as well as another 18 cases with epidemiologically relevant lineage B.1.1.7, known as British variant. Responsible Prediction Making of COVID-19 Mortality (Student Abstract) Hubert Baniecki, Przemyslaw Biecek AAAI Conference on Artificial Intelligence (2021) During the literature review of COVID-19 related prognosis and diagnosis, we found out that most of the predictive models are not faithful to the RAI principles, which can lead to biassed results and wrong reasoning. To solve this problem, we show how novel XAI techniques boost transparency, reproducibility and quality of models. Models in the Wild: On Corruption Robustness of Neural NLP Systems Barbara Rychalska, Dominika Basaj, Alicja Gosiewska, Przemyslaw Biecek International Conference on Neural Information Processing (2019) In this paper we introduce WildNLP - a framework for testing model stability in a natural setting where text corruptions such as keyboard errors or misspelling occur. We compare robustness of deep learning models from 4 popular NLP tasks: Q&amp;A, NLI, NER and Sentiment Analysis by testing their performance on aspects introduced in the framework. In particular, we focus on a comparison between recent state-of-the-art text representations and non-contextualized word embeddings. In order to improve robustness, we perform adversarial training on selected aspects and check its transferability to the improvement of models with various corruption types. We find that the high performance of models does not ensure sufficient robustness, although modern embedding techniques help to improve it. auditor: an R Package for Model-Agnostic Visual Validation and Diagnostics Alicja Gosiewska, Przemyslaw Biecek The R Journal (2019) This paper describes methodology and tools for model-agnostic auditing. It provides functinos for assessing and comparing the goodness of fit and performance of models. In addition, the package may be used for analysis of the similarity of residuals and for identification of outliers and influential observations. The examination is carried out by diagnostic scores and visual verification. The code presented in this paper are implemented in the auditor package. Its flexible and consistent grammar facilitates the validation models of a large class of models. Explanations of Model Predictions with live and breakDown Packages Mateusz Staniak, Przemyslaw Biecek The R Journal (2018) Complex models are commonly used in predictive modeling. In this paper we present R packages that can be used for explaining predictions from complex black box models and attributing parts of these predictions to input features. We introduce two new approaches and corresponding packages for such attribution, namely live and breakDown. We also compare their results with existing implementations of state-of-the-art solutions, namely, lime (Pedersen and Benesty, 2018) which implements Locally Interpretable Model-agnostic Explanations and iml (Molnar et al., 2018) which implements Shapley values. DALEX: Explainers for Complex Predictive Models in R Przemyslaw Biecek Journal of Machine Learning Research (2018) This paper describes a consistent collection of explainers for predictive models, a.k.a. black boxes. Each explainer is a technique for exploration of a black box model. Presented approaches are model-agnostic, what means that they extract useful information from any predictive method irrespective of its internal structure. Each explainer is linked with a specific aspect of a model. Every explainer presented here works for a single model or for a collection of models. In the latter case, models can be compared against each other. Presented explainers are implemented in the DALEX package for R. They are based on a uniform standardized grammar of model exploration which may be easily extended. archivist: An R Package for Managing, Recording and Restoring Data Analysis Results Przemyslaw Biecek, Marcin Kosiński Journal of Statistical Software (2017) Everything that exists in R is an object (Chambers 2016). This article examines what would be possible if we kept copies of all R objects that have ever been created. Not only objects but also their properties, meta-data, relations with other objects and information about context in which they were created. We introduce archivist, an R package designed to improve the management of results of data analysis. "],["software.html", "Software", " Software DALEX XAI with DALEX for R and Python survex Explainable machine learning in survival analysis Arena Interactive tool for the exploration and comparison of models’ explanations fairmodels Fairness with fairmodels COVID-19 COVID-19 Risk Score archivist Model governance with R (MLOps) "],["books.html", "Books", " Books Explanatory Model Analysis Explore, Explain, and Examine Predictive Models. With examples in R and Python Przemysław Biecek, Tomasz Burzykowski Chapman and Hall/CRC, New York (2021) Odkrywać! Ujawniać! Objaśniać! Zbiór esejów o sztuce prezentowania danych Essays on the art of data visualisation Przemysław Biecek University of Warsaw Press (2016) Analiza danych z programem R. Modele liniowe z efektami stałymi, losowymi i mieszanymi Data analysis in R. Linear models with fixed, random and mixed effects Przemysław Biecek Polish Scientific Publishers PWN (2013) "],["seminars.html", "Seminars", " Seminars We meet every Monday, at 10 am online or in MI2DataLab (room 044, Faculty of Mathematics and Information Science, Warsaw University of Technology). Join us at https://meet.google.com/nno-okiz-bxy (or http://meet.drwhy.ai/) List of topics and materials from past seminars: https://github.com/MI2DataLab/MI2DataLab_Seminarium "],["research-grants.html", "Research grants", " Research grants ARES 2022-2026 ARES: Attack-resistant Explanations toward Secure and trustworthy AI Machine learning explainability, fairness, robustness, and security are key elements of trustworthy Artificial Intelligence, an area of strategic importance. In this context, the main goals of the ARES project are: Develop adversarial attacks on state-of-the-art explanations to investigate vulnerabilities and limitations of the existing explainability and fairness approaches in machine learning. Introduce novel robust explanations that are stable against manipulation and intuitive to evaluate. Achieving the first goal primarily impacts various domains of research, which currently use (and explain) black-box models for knowledge discovery and decision-making, by highlighting vulnerabilities and limitations of their explanations. Achieving the second goal impacts more the broad machine learning domain as it aims at improving state-of-the-art by introducing robust explanations toward secure and trustworthy AI. Work on this project is financially supported by the Polish National Science Centre PRELUDIUM BIS grant 2021/43/O/ST6/00347. DARLING 2022-2024 DARLING: Deep Analysis of Regulations with Language Inference, Network analysis and institutional Grammar Aim of the project Developing the tools for automated analysis of content of legal documents leveraging Natural Language Processing, that will help understand the dynamic of change in public policies and variables influencing those changes. Those tools will be firstly used to analyse the case of development of policy subsystem regulating usage of AI in the European Union. Specific goals of the project Developing and evaluating multilingual models for issue classification for legal and public policy documents. Developing embedding-based topic modeling methods for legal and public policy documents suited for analysis of change of the topics between documents. Institutional grammar based analysis of changes in topics between different public policy documents, regulations and public consultation documents. Agent-based models predicting diffusion of issues in public policy documents. Methodology The core of the DARLING project is the issues and topic analysis in documents connected with regulations development using NLP tools. Issues analysis shall allow tracking how different options of AI operationalisation, ways the AI-connected threats are perceived as well as ideas regarding AI regulations are shared among three different types of texts: scientific, expert and legal ones. The extracted issues will then be subject to complex networks analysis and institutional grammar approach. The network analysis, backed by agent-based modeling, will be used to examine the flow of issues among the documents based on their vector-formed characteristics. On the other hand, the Institutional Grammar (IG) will be used to analyze the modality of issues, e.g., the tendency to regulate a specific aspect of AI in a given issue, its deontic character or its conditionality. In result the DARLING project will effect in the development of new methods to analyze legal documents connected to regulation based on deep text processing and links among the documents. An inter-institutional and interdisciplinary team of computer, political sciences and physics of complex systems scientists will elaborate new machine learning approaches to examine the regulation corpora, issues recognition, issues analysis by the means of IG as well as propose new methods of modeling the flow/changes of regulations based on complex networks tools. X-LUNGS 2021-2024 X-LUNGS: Responsible Artificial Intelligence for Lung Diseases The aim of the project is to support the process of identification of lesions visible on CT and lung x-rays. We intend to achieve this goal by building an information system based on artificial intelligence (AI) that will support the radiologist’s work by enriching the images with additional information. The unique feature of the proposed system is a trustworthy artificial intelligence module that: will reduce the image analysis time needed to detect lesions, will make the image evaluation process more transparent, will provide image and textual explanations indicating the rationale behind the proposed recommendation, will be verified for effective collaboration with the radiologist. Work on this project is financially supported from the INFOSTRATEG-I/0022/2021-00 grant funded by Polish National Centre for Research and Development (NCBiR). HOMER 2020-2025 HOMER: Human Oriented autoMated machinE leaRning One of the biggest challenges in the state-of-the-art machine learning is dealing with the complexity of predictive models. Recent techniques like deep neural networks, gradient boosting or random forests create models with thousands or even millions of parameters. This makes decisions generated by these black-box models completely opaque. Model obscurity undermines trust in model decisions, hampers model debugging, blocks model auditability, exposes models to problems with concept drift or data drift. Recently, there has been a huge progress in the area of model interpretability, which results in the first generation of model explainers, methods for better understanding of factors that drive model decisions. Despite this progress, we are still far from methods that provide deep explanations, confronted with domain knowledge that satisfies our ,,Right to explanation’’ as listed in the General Data Protection Regulation (GDPR). In this project I am going to significantly advance next generation of explainers for predictive models. This will be a disruptive change in the way how machine learning models are created, deployed, and maintained. Currently to much time is spend on handcrafted models produced in a tedious and laborious try-and-error process. The proposed Human-Oriented Machine Learning will focus on the true bottleneck in development of new algorithms, i.e. on model-human interfaces. The particular directions I consider are (1) developing an uniform grammar for visual model exploration, (2) establishing a methodology for contrastive explanations that describe similarities and differences among different models, (3) advancing a methodology for non-additive model explanations, (4) creating new human-model interfaces for effective communication between models and humans, (5) introducing new techniques for training of interpretable models based on elastic surrogate black-box models, (6) rising new methods for automated auditing of fairness, biases and performance of predictive models. Work on this project is financially supported from the SONATA BIS grant 2019/34/E/ST6/00052 funded by Polish National Science Centre (NCN). DeCoviD 2020-2022 DeCoviD: Detection of Covid-19 related markers of pulmonary changes using Deep Neural Networks models supported by eXplainable Artificial Intelligence and Cognitive Compressed Sensing Covid-19 is an infectious respiratory disease. A coronavirus infection leaves permanent ramifications in the respiratory system and beyond. In this situation, tools supporting diagnosis and assessment of lung damage after infection and during Covid-19 treatment are crucial. Preliminary results of analysis of CT images and lung xrays suggest that they can help to quickly assess even asymptomatic cases and facilitate prognosis of response to treatment. There are also reports of usefulness of ultrasound images. The aim of the DeCoviD project is to develop methods and tools to support radiologists in the assessment of lung imaging data for the occurrence of changes caused by Covid-19 disease. The developed solution will allow to automate the identification of pathological changes and will support the diagnosis of coexisting lung diseases as well as diseases of other organs visible on chest images. It will also allow to quantify the severity of lung damage caused by the disease Responsible decision support for radiologists requires models based on interpretable features. Such features will be stored in a hybrid knowledge base powered by two research teams from WUT, working on the basis of two, seemingly opposite, paradigms of image data analysis. The eXplainable Artificial Intelligence (XAI) team will use trained deep networks to automatically extract features that are essential for effective disease detection. Cognitive Compressed Sensing (CCS) will build a set of interpretable semantic features using sparse cognitive representations agreed with a group of cooperating radiologists. Combining these two approaches will achieve high effectiveness of the constructed models, combined with high transparency, clarity and stability of the solution. The DeCoviD project is a part of a broader strategy of competence development in the area of deep learning + XAI + medical applications at the Warsaw University of Technology. More information: https://github.com/MI2DataLab/DeCoviD. Work on this project is financially supported by the IDUB against COVID PW. DALEX 2018-2022 DALEX: Descriptive and model Agnostic Local EXplanations Research project objectives. Black boxes are complex machine learning models, for example deep neural network, an ensemble of trees of high-dimensional regression model. They are commonly used due to they high performance. But how to understand the structure of a black-box, a model in which decision rules are too cryptic for humans? The aim of the project is to create a methodology for such exploration. To address this issue we will develop methods, that: (1) identify key variables that mostly determine a model response, (2) explain a single model response in a compact visual way through local approximations, (3) enrich model diagnostic plots. Research project methodology. This project is divided into three subprojects - local approximations od complex models (called LIVE), explanations of particular model predictions (called EXPLAIN) and conditional explanations (called CONDA). Expected impact on the development of science. Explanations of black boxes have fundamental implications for the field of predictive and statistical modelling. The advent of big data forces imposes usage of black boxes that are easily able to overperform classical methods. But the high performance itself does not imply that the model is appropriate. Thus, especially in applications to personalized medicine or some regulated fields, one should scrutinize decision rules incorporated in the model. New methods and tools for exploration of black-box models are useful for quick identification of problems with the model structure and increase the interpretability of a black-box Work on this project is financially supported from the OPUS grant 2017/27/B/ST6/01307 funded by Polish National Science Centre (NCN). MLGenSig 2017-2021 MLGenSig: Machine Learning Methods for building of Integrated Genetic Signatures Research project objectives. The main scientific goal of this project is to develop a methodology for integrated genetic signatures based on data from divergent high-throughput techniques used in molecular biology. Integrated signatures base on ensembles of signatures for RNA-seq, DNA-seq, data as well for methylation profiles and protein expression microarrays. The advent of high throughput methods allows to measure dozens of thousands or even millions features on different levels like DNA / RNA / protein. And nowadays in many large scale studies scientists use data from mRNA seq to assess the state of transcriptome, protein microarrays to asses the state of proteome and DNA-seq / bisulfide methylation to assess genome / methylome. Research methodology. Genetic signatures are widely used in different applications, among others: for assessing genes that differentiate cells that are chemo resistant vs. cells that are not, assess the stage of cell pluripotency, define molecular cancer subtypes. For example, in database Molecular Signatures Database v5.0 one can find thousands of gene sets - genetic signatures for various conditions. There are signatures that characterize some cancer cells, pluripotent cells and other groups. But they usually contain relatively small number of genes (around 100), results with them are hard to replicate and they are collection of features that were found significant when independently tested. In most cases signatures are derived from measurements of the same type. Like signatures based of expression of transcripts based on data from microarrays or RNA-seq, or methylation profile or DNA variation. We are proposing a very different approach. First we are going to use machine-learning techniques to create large collections of signatures. Such signatures base on ensembles of small sub-signatures, are more robust and usually have higher precision. Then out of such signatures we are going to develop methodology for meta-signatures, that integrate information from different types of data (transcriptome, proteome, genome). Great examples of such studies are: Progenitor Cell Biology Consortium (PCBC) and The Cancer Genome Atlas (TCGA) studies. For thousands of patients in different cohorts (for PCBC cohorts based on stemness phenotype, for TCGA based on cancer type) measurements of both mRNA, miRNA, DNA and methylation profiles are available. New, large datasets require new methods that take into account high and dense structure of dependencies between features. The task that we are going to solve is to develop methodology that will create genetic signatures that integrate information from different levels of cell functioning. Then we are going to use data from TCGA and PBCB project to assess the quality of proposed methodology. As a baseline we are going to use following methodologies: DESeq, edgeR (for mRNA), casper (for lternative splicing), MethylKit (for RRBS data) and RPPanalyzer for protein arrays. Here is the skeleton for our approach: (1) Use ensembles in order to building a genetic signature. The first step would be to use random forests to train a new signature. Ensembles of sub-signtures are build on bootstrap subsamples and they votes if given sample fit given signature or not. (2) In order to improve signatures we are going to consider various normalization of raw counts. We start with log and rank transformation. (3) In order to improve the process of training an ensemble we are going to use pre-filtering of genes. (4) Another approach is to use Bayesian based methods, that may incorporate the expert knowledge, like belief-based gaussian modelling Research project impact. Genetic profiling is more and more important and has number of application starting from basic classification up to personalized medicine in which patients are profiled against different signatures. Existing tools for genetic signatures have many citations. This we assume that the methodology for integrated genetic profiling will be a very useful for many research groups. It is hard to overestimate the impact of better genetic profiling on medicine. Moreover we build a team of people with knowledge in cancer genetic profiling Work on this project is financially supported from the OPUS grant 2016/21/B/ST6/02176 funded by Polish National Science Centre (NCN). "],["mi²solutions.html", "MI²Solutions", " MI²Solutions Hire a team of experienced researchers. The blue team will help you develop good predictive models, create a responsible solution tailored to your needs. The red team will help you find and analyse any weaknesses in your predictive models. It will help you confront them with domain knowledge and make sure they are resilient to future changes in the data. If you need tailor-made solutions for your individual needs, we are happy to help you too. Contact us, we can develop software for you, deploy it, provide training, discuss your needs, verify the quality of your existing solutions. Below you will find a sample offer for trainings or deployments. Research as a service Our team has experience not only in groundbreaking research, but also in deploying these research into business. There are many ways we can help, for example help in delivery of champion-challenger evaluations in which we look for potential to increase the effectiveness of predictive models in your company. take care of the whole life cycle of the predictive models, from reproducibility of results to constant monitoring and continuous improvement of the model. audit models and analyse the sensitivity and vulnerability of the model to incorrect or unexpected behaviours. We would be happy to discuss how we could help with your organisation! Trainings Based on our experience in the area of Responsible Machine Learning, developed a unique two-day hands-on training. Jump into the topic of eXplainable Artificial Intelligence with our trainers. Responsible Machine Learning images/training_xai The training is conducted once a month in small groups online. Small groups encourage questions and the interactions within the group. Language Depending on the group’s preference, the hands-on part can be carried out in R (mlr3 + DALEX) or Python (scikit-learn + dalex). The methodology part does not depend on the language. Book your training To book a training please contact with trainings(at)solutions42.ai. "],["mi²education.html", "MI²Education", " MI²Education The demand for predictive modelling skills is growing at a furious rate. Part of our mission is to develop human capital so that predictive modelling is applied responsibly and safely. We take social responsibility seriously and as part of our activities we support the development of data analysis skills among pupils, students and senior professionals alike. "],["teaching.html", "Teaching", " Teaching Programming R 22/23 Summer Programming and data analysis advanced in R lectures, labs, projects - Anna Kozak Exploratory Data Analysis 22/23 Summer Introduction to exploratory data analysis for Mathematics and data analysis studies lectures, labs, projects - Anna Kozak labs - Hubert Ruczyński, Bartłomiej Sobieski Data Visualization 22/23 Winter Data Visualization Techniques for Data Science studies lectures, labs, projects - Anna Kozak labs, projects - Mateusz Krzyziński, Hubert Ruczyński, Mikołaj Spytek Exploratory Data Analysis 21/22 Summer Introduction to exploratory data analysis for Mathematics and data analysis studies lectures, labs, projects - Anna Kozak labs - Katatzyna Woźnica Interpretable Machine Learning 21/22 Summer Interpretable Machine Learning lectures, projects - Przemysław Biecek Case Studies 21/22 Summer Case Studies for Data Science studies lectures - Weronika Hryniewska ML-1 - labs, projects - Anna Kozak ML-2 - labs, projects - Bartłomiej Eljasiak XAI-tabular - labs, projects - Mustafa Cavus AutoML - labs, projects - Katarzyna Woźnica XIC - labs, projects - Hubert Baniecki TL - labs, projects - Paulina Tomaszewska Data - labs, projects - Weronika Hryniewska NLP - labs, projects - Stanisław Giziński Data Visualization 21/22 Winter Data Visualization Techniques for Data Science studies lectures, labs, projects - Anna Kozak labs, projects - Hubert Baniecki Exploratory Data Analysis 20/21 Summer Introduction to exploratory data analysis for Mathematics and data analysis studies lectures, labs, projects - Anna Kozak labs - Krzysztof Spaliński Case Studies 20/21 Summer Case Studies for Data Science studies lectures - Katarzyna Woźnica XAI1 - labs, projects - Anna Kozak XAI2 - labs, projects - Szymon Maksymiuk DL1 - labs, projects - Weronika Hryniewska DL2 - labs, projects - Paulina Tomaszewska ML - labs, projects - Hubert Baniecki RashomonML - labs, projects - Katarzyna Woźnica Interpretable Machine Learning 20/21 Summer Interpretable Machine Learning for Data Science studies XAI stories 2 lectures, projects - Przemysław Biecek Data Visualization 20/21 Winter Data Visualization Techniques for Data Science studies lectures, labs - Alicja Gosiewska projects - Hubert Baniecki Case Studies 19/20 Summer Case Studies for Data Science studies lectures - Alicja Gosiewska Imputation - labs, projects - Katarzyna Woźnica Reproducibility of scientific papers - labs, projects - Alicja Gosiewska Interpretability - labs, projects - Katarzyna Kobylińska Interpretable Machine Learning 19/20 Summer Interpretable Machine Learning for Data Science studies lectures, projects - Przemysław Biecek Data Visualization 19/20 Summer Data Visualization for Data Science studies lectures, labs, projects - Michał Burdukiewicz "],["beta-bit.html", "Beta Bit", " Beta Bit Chaos Game EN: Are you curious about fractals? The Chaos Game is the book for you. You will learn the mathematical basis behind these figures, find out what algorithm can be used to code them, write code in your favourite programming language (Python, R, Julia?) and also explore the bibliographies of three mathematicians associated with the development of mathematics around these shapes. This is the next book in the Beta Bit series for anyone interested in computational mathematics and data analysis. PL: Jesteś ciekawy czym są fraktale? Gra w Chaos to książka dla Ciebie. Poznasz matematyczne podstawy tych figur, dowiesz się, jaki algorytm można wykorzystać do ich zaprogramowania, napiszesz kod w swoim ulubionym języku programowania (Python, R, Julia?), a także poznasz bibliografie trzech matematyków związanych z rozwojem matematyki wokół tych kształtów. To kolejna książka z serii Beta Bit dla wszystkich zainteresowanych matematyką obliczeniową i analizą danych. Flipbook online [ENG] Flipbook online [POL] Wykresy od kuchni PL: Jak tworzyć dobre wykresy? Dobre, czyli takie, które z przyjemnością się ogląda, z których można wyciągnąć wiele informacji, które są zrozumiałe dla szerokiego odbiorcy, a jednocześnie docenią je smakosze. Na bazie doświadczeń z prowadzenia tych zajęć powstały Wykresy od kuchni. To zbiór krótkich wykładów omawiających różne wątki przydatne w lepszym zrozumieniu tego, jak działa komunikacja z użyciem wykresów statystycznych. Na kolejnych stronach pojawi się wiele analogii do przyrządzania posiłków, ponieważ zarówno w kuchni, jak i w przygotowaniu wykresów statystycznych potrzebna jest praktyka, znajomość pewnych fundamentalnych prawideł, garść sprawdzonych przepisów i dużo zapału do eksperymentowania. Będąc tak uzbrojonym, każdy adept sztuki kulinarnej jest skazany na sukces. Flipbook online [POL] The Hitchhiker’s Guide to Responsible Machine Learning EN: A one-of-a-kind 52-page story about responsible machine learning. Beta and Bit use decision trees, random forests, and AutoML tools to build a risk model after a covid infection, and then use explainable artificial intelligence tools to analyze the behavior of that model. The description of the data analysis process is intertwined with descriptions of ML tools and code snippets. All examples are fully reproducible! PL: Jedyna w swoim rodzaju 52-stronicowa opowieść o odpowiedzialnym uczeniu maszynowym. Beta i Bit używają drzew decyzyjnych, lasów losowych i narzędzi AutoML do budowy modelu ryzyka po zakażeniu covid, a następnie używają narzędzi wyjaśnialnej sztucznej inteligencji by przeanalizować działanie tego modelu. Opis procesu analizy danych przeplata się na opisem kolejnych narzędzi i przykładami kodu. Wszystkie wyniki są całkowicie odtwarzalne! Flipbook online Przemysław Biecek, Anna Kozak, Aleksander Zawada Fundacja Naukowa SmarterPoland.pl. 2022 W pogoni za nieskończonością. Szeregi EN: What does hiking in the mountains have to do with the convergence of series? Quite a lot! We start with the paradoxes related to infinity, but step by step we learn the techniques of geometric series. In this book, the conditions for convergence are explained, together with numerous examples. The comic ends with a collection of exercises with different levels of difficulty. PL: Co wspólnego ma chodzenie po górach ze zbieżnością szeregów? Otóż całkiem sporo! Zaczynamy od paradoksów związanych z nieskończonością, ale krok po kroku poznajemy techniki szeregów geometrycznych. W tej pozycji wyjaśnione są warunki zbieżności wraz z licznymi przykładami. Komiks kończy zbiór zadań o różnych poziomach trudności. Flipbook online Przemysław Biecek, Łukasz Maciejewski, Aleksander Zawada Fundacja Naukowa SmarterPoland.pl. 2022 Przewodnik po pakiecie R EN: The Guide to the R package was the first published Polish book focused on the R language. The current fourth edition consists of four parts: Basics of using R (+tidyverse, shiny, knitr and other goodies), Programming in R (object-oriented, package development, class system), Statistics with R (statistical tests, models, exploration techniques) and Visualization with R (graphics, lattice and ggplot2 packages). PL: Przewodnik po pakiecie R był pierwszą wydaną polskojęzyczną książką poświęconą językowi R. Aktualne czwarte wydanie składa się z czterech części: Podstaw posługiwania się językiem R (+tidyverse, shiny, knitr i inne smaczki), Programowanie w R (obiektowe, tworzenie pakietów, system klas), Statystyka z R (testy statystyczne, modele, techniki eksploracji) i Wizualizacja z R (pakiety graphics, lattice i ggplot2). Wersja online, Książka w księgarnii. Przemysław Biecek Wydawnictwo GiS. 2008-2021 Analiza danych z programem R EN: An academic textbook describing estimation and testing topics for linear models with fixed effects, random effects and mixed effects. The theoretical introduction is complemented by numerous examples for one-way and multivariate ANOVA, one and multiple random components. The examples focus on biological and medical applications and are based on real analyses of real data. PL: Podręcznik akademicki opisujący zagadnienia estymacji i testowania dla modeli liniowych z efektami stałymi, losowymi i mieszanymi. Wprowadzenie teoretyczne jest uzupełnione o liczne przykłady dla jednokierunkowej i wielokierunkowej ANOVA, jednym i wieloma komponentami losowymi. Przykłady dotyczą głównie zastosowań biologicznych i medycznych i bazują na prawdziwych analizach rzeczywistych danych. Książka w księgarnii. Przemysław Biecek Wydawnictwo Naukowe PWN 2013-2018 Eseje o sztuce wizualizacji danych EN: Discover! Reveal! Explain! These three roles can be fulfilled by good statistical graphics. Good means understandable, faithful to the data, aesthetic. How to create such graphics? A collection of essays on the art of displaying data systematises knowledge useful in designing and producing good data visualisations. It is not easy. On the one hand, we can fall into the trap of a colourful mush full of numbers, which is sometimes proudly called infographics. On the other hand, we can fall into the trap of graphics that perfectly reproduce the complexity of numbers, and thus completely incomprehensible. Somewhere in the middle is a graphic that explains, that informs, that is aesthetically pleasing and informative. PL: Odkrywać! Ujawniać! Objaśniać! Te trzy role może spełniać dobra grafika statystyczna. Dobra czyli zrozumiała, wierna danym, estetyczna. Jak tworzyć taką grafikę? Zbiór esejów o sztuce pokazywania danych systematyzuje wiedzę przydatną do projektowania i wykonania dobrej wizualizacji danych. Nie jest to proste. Z jednej strony możemy wpaść w pułapkę pstrokatej papki najeżonej liczbami, którą czasem dumnie nazywa się infografiką. Z drugiej strony wpaść można w pułapkę grafiki idealnie odwzorowującej złożoność liczb a przez to zupełnie niezrozumiałej. Gdzieś po środku jest grafika, która wyjaśnia, która informuje, która jest estetyczna i informatywna. Książka online, Książka w księgarnii. Przemysław Biecek Wydawnictwo SmarterPoland 2008-2021 Pogromcy Danych EN: Data Crunchers is the first MOOC (Massive Open Online Course) developed in Polish for data scientists. Two modules were developed in 2015: the first one is an introduction to R, with loading data, overview of syntax, basic data types, descriptive statistics and pipelined processing. The second module is dedicated to data visualisation and statistical modelling. More than 8,000 people have registered on the Data Crunchers platform. PL: Pogromcy Danych to pierwszy MOOC (Massive Open Online Course) opracowany w języku polskim do analizy danych. W roku 2015 powstały dwa moduły: pierwszy jest wprowadzeniem do programu R, przez wczytywanie danych, omówienie składni, podstawowych typów danych, statystyk opisowych oraz przetwarzania potokowego. Drugi moduł poświęcony jest wizualizacji danych oraz modelowaniu statystycznemu. W platformie Pogromców Danych zarejestrowało się ponad 8000 osób. Przetwarzanie danych w programie R, Wizualizacja i modelowanie, Strona WWW. Przemysław Biecek ICM UW. 2015 Wykresy unplugged EN: Can you create clear charts without any electricity? An illustrated collection of exercises showing eight of the most popular ways to visualise data, with do-it-yourself challenges. Grab your crayons and start creating fantastic charts. PL: Czy można tworzyć czytelne wykresy bez użycia prądu? Ilustrowany zbiór ćwiczeń przedstawiających osiem najpopularniejszych sposobów wizualizacji danych, wraz z zadaniami do samodzielnego wykonania. Weź kredki i zacznij tworzyć fantastyczne wykresy. Flipbook online, Komiks w księgarnii. Przemysław Biecek, Ewa Baranowska, Piotr Sobczyk Fundacja Naukowa SmarterPoland.pl. 2018 W pogoni za nieskończonością EN: Two mathematicians share stories about infinity. In the first Beta attends a lecture on the properties of prime numbers. In the second, Bit breaks into the Palace of Culture and Science. How should we talk about mathematics? PL: Dwójka matematyków wymienia się opowiadaniami o nieskończoności. W pierwszym Beta bierze udział w wykładzie o właściwościach liczb pierwszych. W drugim Bit włamuje się do Pałacu Kultury i Nauki. Jak opowiadać o matematyce? Flipbook online, Komiks w księgarnii. Przemysław Biecek, Łukasz Maciejewski, Tomasz Samojlik, Sebastian Szpakowski Fundacja Naukowa SmarterPoland.pl. 2018 Jak długo żyją Muffinki? EN: A collection of three stories for children showing statistical relationships in the world around us. Beautifully illustrated stories about the distribution of height according to age, the life span of dogs or measuring the weight of trees. PL: Zbiór trzech opowiadań dla dzieci pokazującym zależności statystyczne w świecie wokół nas. Pięknie ilustrowane opowiadania o rozkładzie wzrostu w zależności od wieku, czasie życia psów czy pomiarze wagi drzew. Online: Jak szybko urosnę, Jak długo żyją Muffinki. Przemysław Biecek Fundacja Naukowa SmarterPoland.pl.2016 Pieczara Pietraszki EN: How linear regression can help in getting home, and why it’s not worth hacking into a mad mathematician’s office. A short story describing the adventures of two teenagers Beta and Bit moving around historic Warsaw. PL: W jaki sposób regresja liniowa może pomóc w powrocie do domu, oraz dlaczego nie warto włamywać się do pokoju szalonego matematyka? Lekkie opowiadanie opisujące przygody dwójki nastolatków Bety i Bita w historycznej Warszawie. Online: W jezyku Polskim, In English, По-Русски. Magda Chudzian, Przemysław Biecek Fundacja Naukowa SmarterPoland.pl. 2015 How to weight a dog with a ruler? EN: Workshop materials for children aged 8-10. Kids measure different parameters of their body, such as arm span or height. Then they create a graph summarizing the collected data and look for relations between the measured features. It just so happens that parts of the human body are proportional to each other and you can use a ruler to find this relationship. Part of the StatTub project. PL: Materiały do warsztaty dla dzieci w wieku 8-10. Dzieci mierzą różne parametry swojego ciała, takie jak rozpiętość ramion lub wzrost. Następnie tworzą wykres podsumowujący zebrane dane i szukają zależności pomiędzy zmierzonymi cechami. Tak się składa, że części ciała ludzkiego są do siebie proporcjonalne i można z użyciem linijki znaleźć tę relację. Część projektu StatTuba. Online: English, Polish, Chinese, Simplified Chinese, Czech, German, Spanish, Spanish (Latin America), French, Dutch, Vietnamese. Przemysław Biecek, Klaudia Korniluk Fundacja Naukowa SmarterPoland.pl.2016-2021 "],["responsibleml-blog.html", "ResponsibleML Blog", " ResponsibleML Blog Read more about the research, solutions and education on our blog: Tools for Explainable, Fair and Responsible ML BASIC XAI with DALEX— Part 1: Introduction Anna Kozak In this post, we will take a closer look at some algorithms used in explainable artificial intelligence. You will find here an introduction to methods of global and local model evaluation. Each description will include a technical introduction, example analysis, and code in R and Python. R packages for eXplainable Artificial Intelligence Przemysław Biecek We have prepared an overview of the most popular R-packages, which can be used to build interpretable models or to explore complex ones. Examples of knitr notebooks for more than 30 packages are available at http://xai-tools.drwhy.ai/. Adversarial attacks on Explainable AI Hubert Baniecki There are various adversarial attacks on machine learning models; hence, ways of defending, e.g. by using Explainable AI methods. Nowadays, attacks on model explanations come to light, so does the defense to such adversary. Here, we introduce fundamental concepts related to the domain. A further reference list is available at https://github.com/hbaniecki/adversarial-explainable-ai. "],["404.html", "Page not found", " Page not found The page you requested cannot be found (perhaps it was moved or renamed). You may want to try searching to find the page's new location, or use the table of contents to find the page you are looking for. "]]
+[["index.html", "MI² MI².AI", " MI² MI².AI MI².AI is a group of mathematicians and computer scientists who love to play with predictive models. We are spread between Warsaw University of Technology and University of Warsaw. Here we have workshops and seminars, here we are forging new ideas, creating tools, solving problems, doing consulting and sharing our positive attitude. Feel free to jump in. Mission Machine learning is like atomic energy. We develop leaders, skills, methods, tools and good practices so that predictive models can be deployed responsibly and sustainably. Vision MI² is a group of experts supporting global initiatives aimed at responsible and sustainable machine learning. We support the development of future leaders of responsible machine learning through internships, PhDs, postdoctoral fellowships and so on. We seek for research grants and business projects to conduct both scientific and applied research. We develop and maintain the software and infrastructure necessary to build responsible and sustainable ML. We develop cooperation with international teams working on similar topics. We support companies to implement best practices related to responsible modelling in their operation. We conduct workshops and training on responsible predictive modelling. "],["the-team.html", "The Team", " The Team Members Przemysław Biecek, PhD, DSc (Team Leader) Hubert Baniecki, PhD student Mustafa Cavus, PhD Maciej Chrabąszcz, MSc student Mateusz Grzyb, MSc student Stanisław Giziński, MSc student Weronika Hryniewska-Guzik, PhD student Paulina Kaczyńska, MSc student Piotr Komorowski, MSc student Mateusz Krzyziński, MSc student Tymoteusz Kwieciński, BSc student Stanisław Łaniewski, PhD student Piotr Piątyszek, BSc student Hubert Ruczyński, MSc student Barbara Rychalska, PhD student Nuno Sepúlveda, PhD Bartek Sobieski, MSc student Mikołaj Spytek, MSc student Tomasz Stanisławek, PhD Hoang Thien Ly, BSc student Paulina Tomaszewska, PhD student Piotr Wilczyński, BSc student Emilia Wiśnios, MSc student Paweł Wojciechowski, BSc Katarzyna Woźnica, PhD student Vladimir Zaigrajew, PhD student Artur Żółkowski, BSc student Collaborators Mariusz Adamek, Prof, MD Przemysław Bombiński, PhD, MD André Fonseca, PhD student Katarzyna Kobylińska, PhD student Anna Kozak, MSc Marcin Luckner, PhD João Malato, PhD student Bartek Pieliński, PhD, DSc Hanna Piotrowska, MA Elżbieta Sienkiewicz, PhD Julian Sienkiewicz, PhD Adrian Stańdo, MSc student Patryk Szatkowski, PhD student, MD Jakub Wiśniewski, MSc student Alumni Piotr Czarnecki, MSc Alicja Gosiewska, MSc Adrianna Grudzień, BSc Maria Kałuska, BSc Marcin Kosiński, MSc Adam Kozłowski, MSc Wojciech Kretowicz, BSc Michał Kuźba, MSc Szymon Maksymiuk, BSc Tomasz Mikołajczyk, PhD Katarzyna Pękala, MSc Adam Rydelek, BSc Bartosz Sawicki, BSc Patryk Słowakiewicz, BSc Michał Sokólski, MSc Mateusz Stączek, BSc Szymon Szmajdziński, BSc Zuzanna Trafas, BSc Kinga Ułasik, BSc Anna Wróblewska, PhD Hanna Zdulska, BSc Przemysław Biecek My personal mission is to enhance human capabilities by supporting them through access to data-driven and knowledge-based predictions. I execute it by developing methods and tools for responsible machine learning, trustworthy artificial intelligence and reliable software engineering. I work as an associate professor at Warsaw University of Technology and the University of Warsaw. I graduated in software engineering and mathematical statistics and now work on model visualisation, explanatory model analysis, predictive modelling and data science for healthcare. In 2016, I formed the research group MI² which develops methods and tools for predictive model analysis. Google Scholar: Af0O75cAAAAJ GitHub: pbiecek LinkedIn: pbiecek Mariusz Adamek I work at two Medical Universities (Silesia and Gdańsk) holding a Professorship in Medicine and Health Sciences. My interests are focused on lung cancer prevention and screening, the latter by means of low-dose computed tomography (LDCT) with special emphasis put on molecular biology methods, prediction models and image analysis aimed to enhance the performance of lung screening outcomes. Website: www.mariuszadamek.io Hubert Baniecki I’m a PhD student in Computer Science at the University of Warsaw. Prior, I did my MSc (2022) and BSc (2021) in Data Science at Warsaw University of Technology. My main research interest is explainable machine learning, with particular emphasis on adversarial attacks &amp; explanation evaluation. Website: hbaniecki.com Mustafa Cavus I work as an assistant professor at the Eskisehir Technical University. I joined the MI² DataLab as a post-doc researcher in 2021. I work on glocal explanations and imbalanced learning. Google Scholar: I63d1WIAAAAJ&amp;hl GitHub: mcavus LinkedIn: mcavus Twitter: mcavus Julian Sienkiewicz I work as an assistant professor at Faculty of Physics, WUT. My main research area links with sociophysics, complex networks and agent-based models. In the scope of MI² DataLab I follow my other interest - scientometrics. Google Scholar: mIwu11QAAAAJ LinkedIn: julek-sienkiewicz-873829 Maciej Chrabąszcz Master’s student in mathematical statisctics at Warsaw University of Technology. Interested in deep learning on text and images, explainable and responsible AI. GitHub: maciejchrabaszcz LinkedIn: maciej-chrabaszcz Stanisław Giziński A Research Software Engineer and student of Machine Learning at Faculty of Mathematics Informatics and Mechanic, University of Warsaw. His work in the lab focuses on using natural language processing and network analysis to better understand the spread of AI public policies. Interested also in applying machine learning in bioinformatics. Google Scholar: Stanisław Giziński GitHub: Gizzio LinkedIn: stanislaw-gizinski Mateusz Grzyb MSc student in Data Science at Warsaw University of Technology. Interested in artificial intelligence and scientific computing, but above all simply enjoys programming. GitHub: mgrzyb99 Weronika Hryniewska-Guzik PhD candidate in computer science at Warsaw University of Technology. Interested in deep learning modelling on medical images in the context of explainability and responsible AI. Google Scholar: aJeg3IQAAAAJ GitHub: Hryniewska LinkedIn: weronikahryniewska Paulina Kaczyńska I am working towards a Master’s degree in Machine Learning at University of Warsaw. I am interested in Natural Language Processing and ML applications in social sciences GitHub: Kaczyniec Piotr Komorowski Master’s student in Machine Learning at the University of Warsaw. Mainly interested in image processing and XAI applied to medical images. GitHub: piotr-komorowski LinkedIn: Piotr-Komorowski Anna Kozak Graduated in mathematical statistics at Warsaw University of Technology. Interested in explainable artificial intelligence and data visualization. Organizes projects related to education. Google Scholar: JIrqf9kAAAAJ GitHub: kozaka93 LinkedIn: kozakanna Mateusz Krzyziński MSc student in Data Science at Warsaw University of Technology. Interested in explainable artificial intelligence, with particular emphasis on XAI methods for survival analysis models and XAI applications in the medical field. Also an enthusiast of data visualization. Google Scholar: i_r7EUgAAAAJ GitHub: krzyzinskim LinkedIn: krzyzinskim Tymoteusz Kwieciński BSc student in Data Science at Warsaw University of Technology. Particularly interested in explainable artificial intelligence, computer vision and NLP. GitHub: Fersoil LinkedIn: Tymoteusz-Kwieciński Stanisław Łaniewski PhD student in Quantitative Psychology and Economics at University of Warsaw, Machine Learning Researcher at MI2 Data Lab, Msc in Actuarial Science and Mathematical Finance at University of Amsterdam, former Quantitative Researcher at Flow Traders His research focuses on enhancing classical methods used in discrete choice and finance with machine learning and how to apply them to explain behavioral phenomena and heuristics. He is also keen on finding balance between best predictive models and their explainability. Avid gamer who applies statistical techniques to deepen the understanding of best strategies LinkedIn: Stanisław-Łaniewski Piotr Piątyszek Undergraduate Data Science student at Warsaw University of Technology. Works as a research software engineer on enhancing accessibility and completeness of explainable AI. During pandemic contributes to a system of monitoring covid variants. Github: piotrpiatyszek Bartosz Pieliński I am an Assistant Professor at the Faculty of Political Science and International Studies at Warsaw University. I am interested in applying quantitative methods to study public policies. I am a founding member of the Institutional Grammar Research Initiative, which is focused on developing a new way of analysing social rules. I have participated in several research projects covering social policy, non-profit organizations, social enterprises, and international organizations. Website: https://pielinski.info/ Google Scholar: hnWiaVEAAAAJ LinkedIn: Bartosz Pieliński Hanna Piotrowska Information designer, focusing mainly on data visualization, branding and book design, with a strong interest in Data Science and perception studies. Winner of numerous awards, including The Kantar Information Is Beautiful Awards, HOW International Design Awards, Polish Graphic Design Awards and KTR. LinkedIn: hanna-piotrowska Twitter: hannapio Behance: hannapio. Hubert Ruczyński I am working towards Masters’s degree in Data Science at Warsaw University of Technology. I am also teaching students about data exploration and visualisation. My major interests are: AutoML | Natural Language Processing | Data Visualization | Fairness. GitHub: HubertR21 LinkedIn: Hubert Ruczyński Barbara Rychalska PhD candidate in computer science at Warsaw University of Technology. Mainly interested in deep learning for natural language processing (NLP), recommender systems and graph-based learning. Google Scholar: Wp0wHJoAAAAJ LinkedIn: Barbara-Rychalska Bartek Sobieski MSc student in Data Science at Warsaw University of Technology. Interested in deep learning and hyperparameter optimization. GitHub: sobieskibj LinkedIn: Bartłomiej-Sobieski Mikołaj Spytek MSc student in Data Science at Warsaw University of Technology. Interested in explainable artificial intelligence, data vizualization and survival analysis. Google Scholar: 1u49AqYAAAAJ GitHub: mikolajsp LinkedIn: Mikołaj-Spytek Tomasz Stanisławek PhD candidate in computer science at Warsaw University of Technology. Mainly interested in deep learning for natural language processing (NLP). Google Scholar: gq8NY_UAAAAJ GitHub: tstanislawek LinkedIn: Tomasz-Stanisławek Paulina Tomaszewska PhD candidate in Computer Science at Warsaw University of Technology. Gained experience in AI at leading universities during: Deep Learning Summer School at Tsinghua University (China), one-semester exchange at Nanyang Technological University (Singapore) and research internships at Gwangju Institute of Science and Technology (South Korea) and Institute of Science and Technology (Austria). Mainly interested in Deep Learning, Computer Vision and Transfer Learning. Recently, focused on digital pathology. Google Scholar: eO245iMAAAAJ LinkedIn: paulina-tomaszewska Hoang Thien Ly Bachelor student in Maths and Data Analysis at Warsaw University of Technology. Interested in working with data, and learning explainable artificial intelligence methods. Google Scholar: JkysewYAAAAJ GitHub: lhthien09 LinkedIn: hthienly Piotr Wilczyński BSc student in Data Science at Warsaw University of Technology. Interested in Large Language Models, AI Deception and Natural Language Processing. Currently working on my thesis, which applies Computer Vision to medicine. GitHub: wi1lku LinkedIn: Piotr-Wilczyński Jakub Wiśniewski Research Software Engineer and third year Data Science student at Warsaw University of Technology. Developer of tools for bias detection and fairness. Currently researching responsible applications of deep learning. President of Data Science Science Club at WUT. Google Scholar: _6eQsXMAAAAJ GitHub: jakwisn LinkedIn: jakwisn Emilia Wiśnios Research Software Engineer and student of Machine Learning at Faculty of Mathematics, Informatics and Mechanics, University of Warsaw. Interested in natural language processing and reinforcement learning. GitHub: emiliawisnios LinkedIn: emilia-wisnios Paweł Wojciechowski Graduated with a bachelor’s degree in Data Science from Warsaw University of Technology. Interested in explainable artificial intelligence, computer vision, and active learning. GitHub: p-wojciechowski LinkedIn: wojciechowski-p Katarzyna Woźnica PhD candidate in computer science at Warsaw University of Technology. Graduated in mathematical statistics. Interested in automated machine learning especially in hyperparameter tuning for tabular data. Carrying statistical analysis and predictive modelling for healthcare. Google Scholar: tAQS1gQAAAAJ GitHub: woznicak LinkedIn: woznicak Vladimir Zaigrajew PhD candidate in computer science at Warsaw University of Technology. Interested in deep learning, primarily on images, with a focus on representation learning. GitHub: WolodjaZ LinkedIn: vladimir-zaigrajew Artur Żółkowski BSc student in Data Science at Warsaw University of Technology. Interested in explainable artificial intelligence, computer vision and NLP. GitHub: arturzolkowski LinkedIn: Artur-Żółkowski "],["open-positions.html", "Open Positions", " Open Positions We have an open call for HOMER and xLungs projects for following positions. If you are interested in any of them please send your CV and Motivation Letter to przemyslaw.biecek at pw.edu.pl. We reserve the right to contact only selected candidates. MI2 policy on benefits and obligations We are seeking highly motivated and talented researchers at different career levels to join MI2.AI research team in the exciting field of explainability and red-teaming of AI models. There is a common recruitment policy for all positions. We are looking for people interested in research who want to create methods and tools that have a real impact (not yet another paper). proactive people with initiative. people with proficiency in programming languages commonly used in AI research, at least one out of Python, R or Julia. solid understanding of AI/ML models, including their architecture and training processes. good communication skills and proficiency in academic writing; excellent problem-solving skills. We offer: the opportunity to learn from experienced and motivated researchers at different career levels as well as the opportunity to work with a rich network of international and interdisciplinary collaborators (for example Harvard Medical University, Vienna University, LMU, NTU, KDD Lab, Bari University and many others). a possibility for a tenure-track position after the completion of the postdoc. possibility to present research results at international conferences. access to our supercomputers, among them Eden with 32x A100 cards and 1.5TB of storage, a perfect place for training large DNN models. Salary Salary depend on experience and type of project (applied projects tend to have higher salaries than typical research projects). Average rates range from 50-60 PLN/hour (equivalent to 8-9.5k FTE) for a student with experience, through to 80-85 PLN/hour (equivalent to 12-13k FTE) for an experienced PhD student. The base salary for a post-doc (no obligatory teaching duties) is 120-140k gross per year (this is a base salary, extras depend on activity). Low commitment projects (1/5 FTE) in 2023 Ex 1: Any object can be embedded with an appropriately trained neural network. But can objects be embedded using tree-based models? This project (an ideal candidate for MSc thesis) is about exploring different embedding construction techniques based on random forest or boosting methods. Ex 2: Analysis of champion-challenger model pairs on subsets of observations obtained using data segmentation. The project consists of an extension of the EPP method (Elo for predictive models) based on segmented data (e.g. by k-means or by random splits). In addition to the performance analysis, a PD profile comparison analysis is also performed on the splits. Mid commitment projects (2-2.5/5 FTE) in 2023 Ex 1: Development of techniques for explaining segmentation models for 1d (feature-producing models, e.g. survival models), 2d (image-producing models) or 3d data. (CT, hyperspectral images) Experience and knowledge of classical explanatory techniques (SHAP, LRP, CRP) are desirable. Ex 2: Development of red-teaming techniques for LLaMA 2-type language models. Development of test suites testing the vulnerability of the model to a set of prompting-based attacks. Full-time positions for post-docs in 2023 In addition to the research tasks listed above, responsibilities include mentoring junior staff (students, PhD students) and leading small research groups. Additional information about costs of living in Warsaw From FENS: Warsaw is the capital of Poland, so in principle it is one of the most expensive cities in the country but living in Warsaw is still much cheaper than in the majority of European cities. In general, a single person in Warsaw is able to cover all necessary expenses (flat, media, transport, food) with ca. 5000-5500 PLN per month. A typical cost of renting a 2-room flat (30-40 square meters) varies between 2000-3000 PLN, depending on the part of the city. Electricity and media (TV, Internet) are usually not included in the rent and have to be paid separately. The cost of the lunch in the vicinity of the main campus of WUT is ca. 25-40 PLN, while a monthly ticket for public transport is 110 or 180 PLN depending on the area of validity. Public health insurance is included in the post (this includes also admission to hospitals), however the medicines are paid separately. Research Software Engineer In addition to research positions, we also offer positions for skilled programmers. Required: Background in Computer Science, Mathematics, Statistics or similar. Experience in Computer Vision or NLP models Scope of work: Training of machine learning models for tabular and image data Interpretable solutions for tabular and image data Responsible ML solutions for the healthcare domain "],["contact.html", "Contact", " Contact Our rooms: 44 (DataLab - separate entrance in front of the main entrance) 316 (xLungs) 317 (HOMER) Faculty of Mathematics and Information Science, Warsaw University of Technology, Koszykowa 75, 00-662 Warszawa VAT: PL 5250005834 "],["mi2redteam.html", "MI²RedTeam", " MI²RedTeam MI²RedTeam analyses machine and deep learning predictive models through the lens of AI explainability, fairness, security and human trust. We develop methods and tools for explanatory model analysis and apply them in practice. MI²RedTeam is a group of researchers experienced in XAI who perform a rigorous evaluation of AI solutions in order to improve their transparency and security. We apply state-of-the-art methods and introduce new ones to tailor our analysis to the specific predictive task. We openly collaborate on various topics related to explainable and interpretable machine learning. Feel free to reach out to us with research ideas and development opportunities. We help organizations to better understand the vulnerabilities of their AI systems, and take steps to mitigate them. Our current core research topics of interest include: [ARES] Attack-Resistant Explanations towards Secure AI, i.e. a critical evaluation of the state-of-the-art analysis techniques [xSurvival] Explanatory analysis of machine learning survival models [Large Model Analysis] Explanatory analysis of large models, e.g. transformers Methods and methodologies introduced by our team: Evaluating explanations of vision transformers InteractiveEMA towards human-model interaction in explainable machine learning for tabular data SurvSHAP(t) for time-dependent analysis of machine learning survival models LIMEcraft for human-guided visual explanations of deep neural networks Fooling PD &amp; Manipulating SHAP for stress-testing widely-applied explanation methods Checklist towards responsible deep learning on medical images SAFE for lifting interpretability-performance trade-off via automated feature engineering WildNLP for stress-testing deep learning models in NLP Explanatory Model Analysis towards comprehensive examination of predictive models Tools developed by our team: DALEX, breakDown, auditor &amp; modelStudio for explainable machine learning in R dalex for explainable and fair machine learning in Python survex dedicated to explaining machine learning survival models fairmodels for fairness analysis of machine learning classification models Applications supported by our team: In medicine, we analyzed hundreds of models predicting among others: survival in uveal melanoma eye cancer, survival in sepsis, type of lung cancer, lung cancer risk in screening, lung cancer mortality, COVID-19 mortality, hospital length of stay, progression of Alzheimer’s disease. In credit scoring, we analyzed the transparency, auditability, and explainability of machine learning models. In football analytics, we analyzed expected goal models for performance analysis. … This initiative is generously supported by the following institutions. "],["mi²research.html", "MI²Research", " MI²Research On a mission to responsibly build machine learning predictive models. Responsible and sustainable predictive modelling is still a new and developing area. We are conducting a number of studies in this domain that examine predictive models applied to tabular data, computer vision or natural language processing models. We investigate the stability and robustness of various methods, work on explainability and transparency for simple and complex models. As part of our this effort, we develop open source software packages (usually in R and Python) for model explanatory analysis, publish scientific articles describing new methods or investigating properties of already known methods, and create educational materials, recommendations and examples of application in specific domains. If you want to find out more about what we are working on, check out our seminar, which is always open to those interested in responsible and sustainable data science. "],["papers.html", "Papers", " Papers Consolidated learning: a domain-specific model-free optimization strategy with validation on metaMIMIC benchmarks Katarzyna Woźnica, Mateusz Grzyb, Zuzanna Trafas, Przemysław Biecek Machine Learning (2023) This paper proposes a new formulation of the tuning problem, called consolidated learning, more suited to practical challenges faced by model developers, in which a large number of predictive models are created on similar datasets. We show that a carefully selected static portfolio of hyperparameter configurations yields good results for anytime optimization, while maintaining the ease of use and implementation. We demonstrate the effectiveness of this approach through an empirical study for the XGBoost algorithm and the newly created metaMIMIC benchmarks of predictive tasks extracted from the MIMIC-IV medical database. Towards Evaluating Explanations of Vision Transformers for Medical Imaging Piotr Komorowski, Hubert Baniecki, Przemysław Biecek CVPR Workshop on Explainable AI for Computer Vision (2023) This paper investigates the performance of various interpretation methods on a Vision Transformer (ViT) applied to classify chest X-ray images. We introduce the notion of evaluating faithfulness, sensitivity, and complexity of ViT explanations. The obtained results indicate that Layerwise relevance propagation for transformers outperforms Local interpretable model-agnostic explanations and Attention visualization, providing a more accurate and reliable representation of what a ViT has actually learned. Hospital Length of Stay Prediction Based on Multi-modal Data towards Trustworthy Human-AI Collaboration in Radiomics Hubert Baniecki, Bartlomiej Sobieski, Przemysław Bombiński, Patryk Szatkowski, Przemysław Biecek International Conference on Artificial Intelligence in Medicine (2023) To what extent can the patient’s length of stay in a hospital be predicted using only an X-ray image? We answer this question by comparing the performance of machine learning survival models on a novel multi-modal dataset created from 1235 images with textual radiology reports annotated by humans. We introduce time-dependent model explanations into the human-AI decision making process. For reproducibility, we open-source code and the TLOS dataset at this URL. SurvSHAP(t): Time-dependent explanations of machine learning survival models Mateusz Krzyziński, Mikołaj Spytek, Hubert Baniecki, Przemysław Biecek Knowledge-Based Systems (2023) In this paper, we introduce SurvSHAP(t), the first time-dependent explanation that allows for interpreting survival black-box models. The proposed methods aim to enhance precision diagnostics and support domain experts in making decisions. SurvSHAP(t) is model-agnostic and can be applied to all models with functional output. We provide an accessible implementation of time-dependent explanations in Python at this URL. The grammar of interactive explanatory model analysis Hubert Baniecki, Dariusz Parzych, Przemyslaw Biecek Data Mining and Knowledge Discovery (2023) This paper proposes how different Explanatory Model Analysis (EMA) methods complement each other and discusses why it is essential to juxtapose them. The introduced process of Interactive EMA (IEMA) derives from the algorithmic side of explainable machine learning and aims to embrace ideas developed in cognitive sciences. We formalize the grammar of IEMA to describe human-model interaction. We conduct a user study to evaluate the usefulness of IEMA, which indicates that an interactive sequential analysis of a model may increase the accuracy and confidence of human decision making. Climate Policy Tracker: Pipeline for automated analysis of public climate policies Artur Żółkowski, Mateusz Krzyziński, Piotr Wilczyński, Stanisław Giziński, Emilia Wiśnios, Bartosz Pieliński, Julian Sienkiewicz, Przemysław Biecek NeurIPS Workshop on Tackling Climate Change with Machine Learning (2022) In this work, we use a Latent Dirichlet Allocation-based pipeline for the automatic summarization and analysis of 10-years of national energy and climate plans (NECPs) for the period from 2021 to 2030, established by 27 Member States of the European Union. We focus on analyzing policy framing, the language used to describe specific issues, to detect essential nuances in the way governments frame their climate policies and achieve climate goals. Explainable expected goal models for performance analysis in football analytics Mustafa Cavus, Przemyslaw Biecek International Conference on Data Science and Advanced Analytics (2022) The expected goal provides a more representative measure of the team and player performance which also suit the low-scoring nature of football instead of the score in modern football. This paper proposes an accurate expected goal model trained on 315,430 shots from seven seasons between 2014-15 and 2020-21 of the top-five European football leagues. Moreover, we demonstrate a practical application of aggregated profiles to explain a group of observations on an accurate expected goal model for monitoring the team and player performance. Multi-omics disease module detection with an explainable Greedy Decision Forest Bastian Pfeifer, Hubert Baniecki, Anna Saranti, Przemyslaw Biecek, Andreas Holzinger Scientific Reports (2022) In this work, we demonstrate subnetwork detection based on multi-modal node features using a novel Greedy Decision Forest (GDF) with inherent interpretability. The latter will be a crucial factor to retain experts and gain their trust in such algorithms. To demonstrate a concrete application example, we focus on bioinformatics, systems biology and particularly biomedicine, but the presented methodology is applicable in many other domains as well. Our proposed explainable approach can help to uncover disease-causing network modules from multi-omics data to better understand complex diseases such as cancer. Interpretable meta-score for model performance Alicja Gosiewska, Katarzyna Woźnica, Przemysław Biecek Nature Machine Intelligence (2022) Elo-based predictive power (EPP) meta-score that is built on other performance measures and allows for interpretable comparisons of models. Differences between this score have a probabilistic interpretation and can be compared directly between data sets. Furthermore, this meta-score allows for an assessment of ranking fitness. We prove the properties of the Elo-based predictive power meta-score and support them with empirical results on a large-scale benchmark of 30 classification data sets. Additionally, we propose a unified benchmark ontology that provides a uniform description of benchmarks. fairmodels: a Flexible Tool for Bias Detection, Visualization, and Mitigation in Binary Classification Models Jakub Wiśniewski, Przemyslaw Biecek The R Journal (2022) This article introduces an R package fairmodels that helps to validate fairness and eliminate bias in binary classification models quickly and flexibly. It offers a model-agnostic approach to bias detection, visualization, and mitigation. The implemented functions and fairness metrics enable model fairness validation from different perspectives. In addition, the package includes a series of methods for bias mitigation that aim to diminish the discrimination in the model. The package is designed to examine a single model and facilitate comparisons between multiple models. A robust framework to investigate the reliability and stability of explainable artificial intelligence markers of Mild Cognitive Impairment and Alzheimer’s Disease Angela Lombardi, Domenico Diacono, Nicola Amoroso, Przemysław Biecek, Alfonso Monaco, Loredana Bellantuono, Ester Pantaleo, Giancarlo Logroscino, Roberto De Blasi, Sabina Tangaro, Roberto Bellotti Brain Informatics (2022) In this work, we present a robust framework to (i) perform a threefold classification between healthy control subjects, individuals with cognitive impairment, and subjects with dementia using different cognitive indexes and (ii) analyze the variability of the explainability SHAP values associated with the decisions taken by the predictive models. We demonstrate that the SHAP values can accurately characterize how each index affects a patient’s cognitive status. Furthermore, we show that a longitudinal analysis of SHAP values can provide effective information on Alzheimer’s disease progression. LIMEcraft: handcrafted superpixel selection and inspection for Visual eXplanations Weronika Hryniewska, Adrianna Grudzień, Przemysław Biecek Machine Learning (2022) LIMEcraft enhances the process of explanation by allowing a user to interactively select semantically consistent areas and thoroughly examine the prediction for the image instance in case of many image features. Experiments on several models show that our tool improves model safety by inspecting model fairness for image pieces that may indicate model bias. The code is available at: this URL. Fooling Partial Dependence via Data Poisoning Hubert Baniecki, Wojciech Kretowicz, Przemyslaw Biecek ECML PKDD (2022) We showcase that PD can be manipulated in an adversarial manner, which is alarming, especially in financial or medical applications where auditability became a must-have trait supporting black-box machine learning. The fooling is performed via poisoning the data to bend and shift explanations in the desired direction using genetic and gradient algorithms. Manipulating SHAP via Adversarial Data Perturbations (Student Abstract) Hubert Baniecki, Przemyslaw Biecek AAAI Conference on Artificial Intelligence (2022) We introduce a model-agnostic algorithm for manipulating SHapley Additive exPlanations (SHAP) with perturbation of tabular data. It is evaluated on predictive tasks from healthcare and financial domains to illustrate how crucial is the context of data distribution in interpreting machine learning models. Our method supports checking the stability of the explanations used by various stakeholders apparent in the domain of responsible AI; moreover, the result highlights the explanations’ vulnerability that can be exploited by an adversary. A Signature of 14 Long Non-Coding RNAs (lncRNAs) as a Step towards Precision Diagnosis for NSCLC Anetta Sulewska, Jacek Niklinski, Radoslaw Charkiewicz, Piotr Karabowicz, Przemyslaw Biecek, Hubert Baniecki, Oksana Kowalczuk, Miroslaw Kozlowski, Patrycja Modzelewska, Piotr Majewski et al. Cancers (2022) The aim of the study was the appraisal of the diagnostic value of 14 differentially expressed long non-coding RNAs (lncRNAs) in the early stages of non-small-cell lung cancer (NSCLC). We established two classifiers. The first recognized cancerous from noncancerous tissues, the second successfully discriminated NSCLC subtypes (LUAD vs. LUSC). Our results indicate that the panel of 14 lncRNAs can be a promising tool to support a routine histopathological diagnosis of NSCLC. dalex: Responsible Machine Learning with Interactive Explainability and Fairness in Python Hubert Baniecki, Wojciech Kretowicz, Piotr Piątyszek, Jakub Wiśniewski, Przemyslaw Biecek Journal of Machine Learning Research (2021) We introduce dalex, a Python package which implements a model-agnostic interface for interactive explainability and fairness. It adopts the design crafted through the development of various tools for explainable machine learning; thus, it aims at the unification of existing solutions. This library’s source code and documentation are available under open license at this URL. Checklist for responsible deep learning modeling of medical images based on COVID-19 detection studies Weronika Hryniewska, Przemysław Bombiński, Patryk Szatkowski, Paulina Tomaszewska, Artur Przelaskowski, Przemysław Biecek Pattern Recognition (2021) Our analysis revealed numerous mistakes made at different stages of data acquisition, model development, and explanation construction. In this work, we overview the approaches proposed in the surveyed Machine Learning articles and indicate typical errors emerging from the lack of deep understanding of the radiography domain. The final result is a proposed checklist with the minimum conditions to be met by a reliable COVID-19 diagnostic model. Towards explainable meta-learning Katarzyna Woźnica, Przemyslaw Biecek ECML PKDD Workshop on eXplainable Knowledge Discovery in Data Mining (2021) To build a new generation of meta-models we need a deeper understanding of the importance and effect of meta-features on the model tunability. In this paper, we propose techniques developed for eXplainable Artificial Intelligence (XAI) to examine and extract knowledge from black-box surrogate models. To our knowledge, this is the first paper that shows how post-hoc explainability can be used to improve the meta-learning. Prevention is better than cure: a case study of the abnormalities detection in the chest Weronika Hryniewska, Piotr Czarnecki, Jakub Wiśniewski, Przemysław Bombiński, Przemysław Biecek CVPR Workshop on “Beyond Fairness: Towards a Just, Equitable, and Accountable Computer Vision” (2021) In this paper, we analyze in detail a single use case - a Kaggle competition related to the detection of abnormalities in X-ray lung images. We demonstrate how a series of simple tests for data imbalance exposes faults in the data acquisition and annotation process. Complex models are able to learn such artifacts and it is difficult to remove this bias during or after the training. Simpler is better: Lifting interpretability-performance trade-off via automated feature engineering Alicja Gosiewska, Anna Kozak, Przemysław Biecek Decision Support Systems (2021) We propose a framework that uses elastic black boxes as supervisor models to create simpler, less opaque, yet still accurate and interpretable glass box models. The new models were created using newly engineered features extracted with the help of a supervisor model. We supply the analysis using a large-scale benchmark on several tabular data sets from the OpenML database. The first SARS-CoV-2 genetic variants of concern (VOC) in Poland: The concept of a comprehensive approach to monitoring and surveillance of emerging variants Radosław Charkiewicz, Jacek Nikliński, Przemysław Biecek, Joanna Kiśluk, Sławomir Pancewicz, Anna Moniuszko-Malinowska, Robert Flisiak, Adam Krętowski, Janusz Dzięcioł, Marcin Moniuszko, Rafał Gierczyński, Grzegorz Juszczyk, Joanna Reszeć Advances in Medical Sciences (2021) This study shows the first confirmed case of SARS-CoV-2 in Poland with the lineage B.1.351 (known as 501Y.V2 South African variant), as well as another 18 cases with epidemiologically relevant lineage B.1.1.7, known as British variant. Responsible Prediction Making of COVID-19 Mortality (Student Abstract) Hubert Baniecki, Przemyslaw Biecek AAAI Conference on Artificial Intelligence (2021) During the literature review of COVID-19 related prognosis and diagnosis, we found out that most of the predictive models are not faithful to the RAI principles, which can lead to biassed results and wrong reasoning. To solve this problem, we show how novel XAI techniques boost transparency, reproducibility and quality of models. Models in the Wild: On Corruption Robustness of Neural NLP Systems Barbara Rychalska, Dominika Basaj, Alicja Gosiewska, Przemyslaw Biecek International Conference on Neural Information Processing (2019) In this paper we introduce WildNLP - a framework for testing model stability in a natural setting where text corruptions such as keyboard errors or misspelling occur. We compare robustness of deep learning models from 4 popular NLP tasks: Q&amp;A, NLI, NER and Sentiment Analysis by testing their performance on aspects introduced in the framework. In particular, we focus on a comparison between recent state-of-the-art text representations and non-contextualized word embeddings. In order to improve robustness, we perform adversarial training on selected aspects and check its transferability to the improvement of models with various corruption types. We find that the high performance of models does not ensure sufficient robustness, although modern embedding techniques help to improve it. auditor: an R Package for Model-Agnostic Visual Validation and Diagnostics Alicja Gosiewska, Przemyslaw Biecek The R Journal (2019) This paper describes methodology and tools for model-agnostic auditing. It provides functinos for assessing and comparing the goodness of fit and performance of models. In addition, the package may be used for analysis of the similarity of residuals and for identification of outliers and influential observations. The examination is carried out by diagnostic scores and visual verification. The code presented in this paper are implemented in the auditor package. Its flexible and consistent grammar facilitates the validation models of a large class of models. Explanations of Model Predictions with live and breakDown Packages Mateusz Staniak, Przemyslaw Biecek The R Journal (2018) Complex models are commonly used in predictive modeling. In this paper we present R packages that can be used for explaining predictions from complex black box models and attributing parts of these predictions to input features. We introduce two new approaches and corresponding packages for such attribution, namely live and breakDown. We also compare their results with existing implementations of state-of-the-art solutions, namely, lime (Pedersen and Benesty, 2018) which implements Locally Interpretable Model-agnostic Explanations and iml (Molnar et al., 2018) which implements Shapley values. DALEX: Explainers for Complex Predictive Models in R Przemyslaw Biecek Journal of Machine Learning Research (2018) This paper describes a consistent collection of explainers for predictive models, a.k.a. black boxes. Each explainer is a technique for exploration of a black box model. Presented approaches are model-agnostic, what means that they extract useful information from any predictive method irrespective of its internal structure. Each explainer is linked with a specific aspect of a model. Every explainer presented here works for a single model or for a collection of models. In the latter case, models can be compared against each other. Presented explainers are implemented in the DALEX package for R. They are based on a uniform standardized grammar of model exploration which may be easily extended. archivist: An R Package for Managing, Recording and Restoring Data Analysis Results Przemyslaw Biecek, Marcin Kosiński Journal of Statistical Software (2017) Everything that exists in R is an object (Chambers 2016). This article examines what would be possible if we kept copies of all R objects that have ever been created. Not only objects but also their properties, meta-data, relations with other objects and information about context in which they were created. We introduce archivist, an R package designed to improve the management of results of data analysis. "],["software.html", "Software", " Software DALEX XAI with DALEX for R and Python survex Explainable machine learning in survival analysis Arena Interactive tool for the exploration and comparison of models’ explanations fairmodels Fairness with fairmodels COVID-19 COVID-19 Risk Score archivist Model governance with R (MLOps) "],["books.html", "Books", " Books Explanatory Model Analysis Explore, Explain, and Examine Predictive Models. With examples in R and Python Przemysław Biecek, Tomasz Burzykowski Chapman and Hall/CRC, New York (2021) Odkrywać! Ujawniać! Objaśniać! Zbiór esejów o sztuce prezentowania danych Essays on the art of data visualisation Przemysław Biecek University of Warsaw Press (2016) Analiza danych z programem R. Modele liniowe z efektami stałymi, losowymi i mieszanymi Data analysis in R. Linear models with fixed, random and mixed effects Przemysław Biecek Polish Scientific Publishers PWN (2013) "],["seminars.html", "Seminars", " Seminars We meet every Monday, at 10 am online or in MI2DataLab (room 044, Faculty of Mathematics and Information Science, Warsaw University of Technology). Join us at https://meet.google.com/nno-okiz-bxy (or http://meet.drwhy.ai/) List of topics and materials from past seminars: https://github.com/MI2DataLab/MI2DataLab_Seminarium "],["research-grants.html", "Research grants", " Research grants ARES 2022-2026 ARES: Attack-resistant Explanations toward Secure and trustworthy AI Machine learning explainability, fairness, robustness, and security are key elements of trustworthy Artificial Intelligence, an area of strategic importance. In this context, the main goals of the ARES project are: Develop adversarial attacks on state-of-the-art explanations to investigate vulnerabilities and limitations of the existing explainability and fairness approaches in machine learning. Introduce novel robust explanations that are stable against manipulation and intuitive to evaluate. Achieving the first goal primarily impacts various domains of research, which currently use (and explain) black-box models for knowledge discovery and decision-making, by highlighting vulnerabilities and limitations of their explanations. Achieving the second goal impacts more the broad machine learning domain as it aims at improving state-of-the-art by introducing robust explanations toward secure and trustworthy AI. Work on this project is financially supported by the Polish National Science Centre PRELUDIUM BIS grant 2021/43/O/ST6/00347. DARLING 2022-2024 DARLING: Deep Analysis of Regulations with Language Inference, Network analysis and institutional Grammar Aim of the project Developing the tools for automated analysis of content of legal documents leveraging Natural Language Processing, that will help understand the dynamic of change in public policies and variables influencing those changes. Those tools will be firstly used to analyse the case of development of policy subsystem regulating usage of AI in the European Union. Specific goals of the project Developing and evaluating multilingual models for issue classification for legal and public policy documents. Developing embedding-based topic modeling methods for legal and public policy documents suited for analysis of change of the topics between documents. Institutional grammar based analysis of changes in topics between different public policy documents, regulations and public consultation documents. Agent-based models predicting diffusion of issues in public policy documents. Methodology The core of the DARLING project is the issues and topic analysis in documents connected with regulations development using NLP tools. Issues analysis shall allow tracking how different options of AI operationalisation, ways the AI-connected threats are perceived as well as ideas regarding AI regulations are shared among three different types of texts: scientific, expert and legal ones. The extracted issues will then be subject to complex networks analysis and institutional grammar approach. The network analysis, backed by agent-based modeling, will be used to examine the flow of issues among the documents based on their vector-formed characteristics. On the other hand, the Institutional Grammar (IG) will be used to analyze the modality of issues, e.g., the tendency to regulate a specific aspect of AI in a given issue, its deontic character or its conditionality. In result the DARLING project will effect in the development of new methods to analyze legal documents connected to regulation based on deep text processing and links among the documents. An inter-institutional and interdisciplinary team of computer, political sciences and physics of complex systems scientists will elaborate new machine learning approaches to examine the regulation corpora, issues recognition, issues analysis by the means of IG as well as propose new methods of modeling the flow/changes of regulations based on complex networks tools. X-LUNGS 2021-2024 X-LUNGS: Responsible Artificial Intelligence for Lung Diseases The aim of the project is to support the process of identification of lesions visible on CT and lung x-rays. We intend to achieve this goal by building an information system based on artificial intelligence (AI) that will support the radiologist’s work by enriching the images with additional information. The unique feature of the proposed system is a trustworthy artificial intelligence module that: will reduce the image analysis time needed to detect lesions, will make the image evaluation process more transparent, will provide image and textual explanations indicating the rationale behind the proposed recommendation, will be verified for effective collaboration with the radiologist. Work on this project is financially supported from the INFOSTRATEG-I/0022/2021-00 grant funded by Polish National Centre for Research and Development (NCBiR). HOMER 2020-2025 HOMER: Human Oriented autoMated machinE leaRning One of the biggest challenges in the state-of-the-art machine learning is dealing with the complexity of predictive models. Recent techniques like deep neural networks, gradient boosting or random forests create models with thousands or even millions of parameters. This makes decisions generated by these black-box models completely opaque. Model obscurity undermines trust in model decisions, hampers model debugging, blocks model auditability, exposes models to problems with concept drift or data drift. Recently, there has been a huge progress in the area of model interpretability, which results in the first generation of model explainers, methods for better understanding of factors that drive model decisions. Despite this progress, we are still far from methods that provide deep explanations, confronted with domain knowledge that satisfies our ,,Right to explanation’’ as listed in the General Data Protection Regulation (GDPR). In this project I am going to significantly advance next generation of explainers for predictive models. This will be a disruptive change in the way how machine learning models are created, deployed, and maintained. Currently to much time is spend on handcrafted models produced in a tedious and laborious try-and-error process. The proposed Human-Oriented Machine Learning will focus on the true bottleneck in development of new algorithms, i.e. on model-human interfaces. The particular directions I consider are (1) developing an uniform grammar for visual model exploration, (2) establishing a methodology for contrastive explanations that describe similarities and differences among different models, (3) advancing a methodology for non-additive model explanations, (4) creating new human-model interfaces for effective communication between models and humans, (5) introducing new techniques for training of interpretable models based on elastic surrogate black-box models, (6) rising new methods for automated auditing of fairness, biases and performance of predictive models. Work on this project is financially supported from the SONATA BIS grant 2019/34/E/ST6/00052 funded by Polish National Science Centre (NCN). DeCoviD 2020-2022 DeCoviD: Detection of Covid-19 related markers of pulmonary changes using Deep Neural Networks models supported by eXplainable Artificial Intelligence and Cognitive Compressed Sensing Covid-19 is an infectious respiratory disease. A coronavirus infection leaves permanent ramifications in the respiratory system and beyond. In this situation, tools supporting diagnosis and assessment of lung damage after infection and during Covid-19 treatment are crucial. Preliminary results of analysis of CT images and lung xrays suggest that they can help to quickly assess even asymptomatic cases and facilitate prognosis of response to treatment. There are also reports of usefulness of ultrasound images. The aim of the DeCoviD project is to develop methods and tools to support radiologists in the assessment of lung imaging data for the occurrence of changes caused by Covid-19 disease. The developed solution will allow to automate the identification of pathological changes and will support the diagnosis of coexisting lung diseases as well as diseases of other organs visible on chest images. It will also allow to quantify the severity of lung damage caused by the disease Responsible decision support for radiologists requires models based on interpretable features. Such features will be stored in a hybrid knowledge base powered by two research teams from WUT, working on the basis of two, seemingly opposite, paradigms of image data analysis. The eXplainable Artificial Intelligence (XAI) team will use trained deep networks to automatically extract features that are essential for effective disease detection. Cognitive Compressed Sensing (CCS) will build a set of interpretable semantic features using sparse cognitive representations agreed with a group of cooperating radiologists. Combining these two approaches will achieve high effectiveness of the constructed models, combined with high transparency, clarity and stability of the solution. The DeCoviD project is a part of a broader strategy of competence development in the area of deep learning + XAI + medical applications at the Warsaw University of Technology. More information: https://github.com/MI2DataLab/DeCoviD. Work on this project is financially supported by the IDUB against COVID PW. DALEX 2018-2022 DALEX: Descriptive and model Agnostic Local EXplanations Research project objectives. Black boxes are complex machine learning models, for example deep neural network, an ensemble of trees of high-dimensional regression model. They are commonly used due to they high performance. But how to understand the structure of a black-box, a model in which decision rules are too cryptic for humans? The aim of the project is to create a methodology for such exploration. To address this issue we will develop methods, that: (1) identify key variables that mostly determine a model response, (2) explain a single model response in a compact visual way through local approximations, (3) enrich model diagnostic plots. Research project methodology. This project is divided into three subprojects - local approximations od complex models (called LIVE), explanations of particular model predictions (called EXPLAIN) and conditional explanations (called CONDA). Expected impact on the development of science. Explanations of black boxes have fundamental implications for the field of predictive and statistical modelling. The advent of big data forces imposes usage of black boxes that are easily able to overperform classical methods. But the high performance itself does not imply that the model is appropriate. Thus, especially in applications to personalized medicine or some regulated fields, one should scrutinize decision rules incorporated in the model. New methods and tools for exploration of black-box models are useful for quick identification of problems with the model structure and increase the interpretability of a black-box Work on this project is financially supported from the OPUS grant 2017/27/B/ST6/01307 funded by Polish National Science Centre (NCN). MLGenSig 2017-2021 MLGenSig: Machine Learning Methods for building of Integrated Genetic Signatures Research project objectives. The main scientific goal of this project is to develop a methodology for integrated genetic signatures based on data from divergent high-throughput techniques used in molecular biology. Integrated signatures base on ensembles of signatures for RNA-seq, DNA-seq, data as well for methylation profiles and protein expression microarrays. The advent of high throughput methods allows to measure dozens of thousands or even millions features on different levels like DNA / RNA / protein. And nowadays in many large scale studies scientists use data from mRNA seq to assess the state of transcriptome, protein microarrays to asses the state of proteome and DNA-seq / bisulfide methylation to assess genome / methylome. Research methodology. Genetic signatures are widely used in different applications, among others: for assessing genes that differentiate cells that are chemo resistant vs. cells that are not, assess the stage of cell pluripotency, define molecular cancer subtypes. For example, in database Molecular Signatures Database v5.0 one can find thousands of gene sets - genetic signatures for various conditions. There are signatures that characterize some cancer cells, pluripotent cells and other groups. But they usually contain relatively small number of genes (around 100), results with them are hard to replicate and they are collection of features that were found significant when independently tested. In most cases signatures are derived from measurements of the same type. Like signatures based of expression of transcripts based on data from microarrays or RNA-seq, or methylation profile or DNA variation. We are proposing a very different approach. First we are going to use machine-learning techniques to create large collections of signatures. Such signatures base on ensembles of small sub-signatures, are more robust and usually have higher precision. Then out of such signatures we are going to develop methodology for meta-signatures, that integrate information from different types of data (transcriptome, proteome, genome). Great examples of such studies are: Progenitor Cell Biology Consortium (PCBC) and The Cancer Genome Atlas (TCGA) studies. For thousands of patients in different cohorts (for PCBC cohorts based on stemness phenotype, for TCGA based on cancer type) measurements of both mRNA, miRNA, DNA and methylation profiles are available. New, large datasets require new methods that take into account high and dense structure of dependencies between features. The task that we are going to solve is to develop methodology that will create genetic signatures that integrate information from different levels of cell functioning. Then we are going to use data from TCGA and PBCB project to assess the quality of proposed methodology. As a baseline we are going to use following methodologies: DESeq, edgeR (for mRNA), casper (for lternative splicing), MethylKit (for RRBS data) and RPPanalyzer for protein arrays. Here is the skeleton for our approach: (1) Use ensembles in order to building a genetic signature. The first step would be to use random forests to train a new signature. Ensembles of sub-signtures are build on bootstrap subsamples and they votes if given sample fit given signature or not. (2) In order to improve signatures we are going to consider various normalization of raw counts. We start with log and rank transformation. (3) In order to improve the process of training an ensemble we are going to use pre-filtering of genes. (4) Another approach is to use Bayesian based methods, that may incorporate the expert knowledge, like belief-based gaussian modelling Research project impact. Genetic profiling is more and more important and has number of application starting from basic classification up to personalized medicine in which patients are profiled against different signatures. Existing tools for genetic signatures have many citations. This we assume that the methodology for integrated genetic profiling will be a very useful for many research groups. It is hard to overestimate the impact of better genetic profiling on medicine. Moreover we build a team of people with knowledge in cancer genetic profiling Work on this project is financially supported from the OPUS grant 2016/21/B/ST6/02176 funded by Polish National Science Centre (NCN). "],["mi²solutions.html", "MI²Solutions", " MI²Solutions Hire a team of experienced researchers. The blue team will help you develop good predictive models, create a responsible solution tailored to your needs. The red team will help you find and analyse any weaknesses in your predictive models. It will help you confront them with domain knowledge and make sure they are resilient to future changes in the data. If you need tailor-made solutions for your individual needs, we are happy to help you too. Contact us, we can develop software for you, deploy it, provide training, discuss your needs, verify the quality of your existing solutions. Below you will find a sample offer for trainings or deployments. Research as a service Our team has experience not only in groundbreaking research, but also in deploying these research into business. There are many ways we can help, for example help in delivery of champion-challenger evaluations in which we look for potential to increase the effectiveness of predictive models in your company. take care of the whole life cycle of the predictive models, from reproducibility of results to constant monitoring and continuous improvement of the model. audit models and analyse the sensitivity and vulnerability of the model to incorrect or unexpected behaviours. We would be happy to discuss how we could help with your organisation! Trainings Based on our experience in the area of Responsible Machine Learning, developed a unique two-day hands-on training. Jump into the topic of eXplainable Artificial Intelligence with our trainers. Responsible Machine Learning images/training_xai The training is conducted once a month in small groups online. Small groups encourage questions and the interactions within the group. Language Depending on the group’s preference, the hands-on part can be carried out in R (mlr3 + DALEX) or Python (scikit-learn + dalex). The methodology part does not depend on the language. Book your training To book a training please contact with trainings(at)solutions42.ai. "],["mi²education.html", "MI²Education", " MI²Education The demand for predictive modelling skills is growing at a furious rate. Part of our mission is to develop human capital so that predictive modelling is applied responsibly and safely. We take social responsibility seriously and as part of our activities we support the development of data analysis skills among pupils, students and senior professionals alike. "],["teaching.html", "Teaching", " Teaching Programming R 22/23 Summer Programming and data analysis advanced in R lectures, labs, projects - Anna Kozak Exploratory Data Analysis 22/23 Summer Introduction to exploratory data analysis for Mathematics and data analysis studies lectures, labs, projects - Anna Kozak labs - Hubert Ruczyński, Bartłomiej Sobieski Data Visualization 22/23 Winter Data Visualization Techniques for Data Science studies lectures, labs, projects - Anna Kozak labs, projects - Mateusz Krzyziński, Hubert Ruczyński, Mikołaj Spytek Exploratory Data Analysis 21/22 Summer Introduction to exploratory data analysis for Mathematics and data analysis studies lectures, labs, projects - Anna Kozak labs - Katatzyna Woźnica Interpretable Machine Learning 21/22 Summer Interpretable Machine Learning lectures, projects - Przemysław Biecek Case Studies 21/22 Summer Case Studies for Data Science studies lectures - Weronika Hryniewska ML-1 - labs, projects - Anna Kozak ML-2 - labs, projects - Bartłomiej Eljasiak XAI-tabular - labs, projects - Mustafa Cavus AutoML - labs, projects - Katarzyna Woźnica XIC - labs, projects - Hubert Baniecki TL - labs, projects - Paulina Tomaszewska Data - labs, projects - Weronika Hryniewska NLP - labs, projects - Stanisław Giziński Data Visualization 21/22 Winter Data Visualization Techniques for Data Science studies lectures, labs, projects - Anna Kozak labs, projects - Hubert Baniecki Exploratory Data Analysis 20/21 Summer Introduction to exploratory data analysis for Mathematics and data analysis studies lectures, labs, projects - Anna Kozak labs - Krzysztof Spaliński Case Studies 20/21 Summer Case Studies for Data Science studies lectures - Katarzyna Woźnica XAI1 - labs, projects - Anna Kozak XAI2 - labs, projects - Szymon Maksymiuk DL1 - labs, projects - Weronika Hryniewska DL2 - labs, projects - Paulina Tomaszewska ML - labs, projects - Hubert Baniecki RashomonML - labs, projects - Katarzyna Woźnica Interpretable Machine Learning 20/21 Summer Interpretable Machine Learning for Data Science studies XAI stories 2 lectures, projects - Przemysław Biecek Data Visualization 20/21 Winter Data Visualization Techniques for Data Science studies lectures, labs - Alicja Gosiewska projects - Hubert Baniecki Case Studies 19/20 Summer Case Studies for Data Science studies lectures - Alicja Gosiewska Imputation - labs, projects - Katarzyna Woźnica Reproducibility of scientific papers - labs, projects - Alicja Gosiewska Interpretability - labs, projects - Katarzyna Kobylińska Interpretable Machine Learning 19/20 Summer Interpretable Machine Learning for Data Science studies lectures, projects - Przemysław Biecek Data Visualization 19/20 Summer Data Visualization for Data Science studies lectures, labs, projects - Michał Burdukiewicz "],["beta-bit.html", "Beta Bit", " Beta Bit Chaos Game EN: Are you curious about fractals? The Chaos Game is the book for you. You will learn the mathematical basis behind these figures, find out what algorithm can be used to code them, write code in your favourite programming language (Python, R, Julia?) and also explore the bibliographies of three mathematicians associated with the development of mathematics around these shapes. This is the next book in the Beta Bit series for anyone interested in computational mathematics and data analysis. PL: Jesteś ciekawy czym są fraktale? Gra w Chaos to książka dla Ciebie. Poznasz matematyczne podstawy tych figur, dowiesz się, jaki algorytm można wykorzystać do ich zaprogramowania, napiszesz kod w swoim ulubionym języku programowania (Python, R, Julia?), a także poznasz bibliografie trzech matematyków związanych z rozwojem matematyki wokół tych kształtów. To kolejna książka z serii Beta Bit dla wszystkich zainteresowanych matematyką obliczeniową i analizą danych. Flipbook online [ENG] Flipbook online [POL] Wykresy od kuchni PL: Jak tworzyć dobre wykresy? Dobre, czyli takie, które z przyjemnością się ogląda, z których można wyciągnąć wiele informacji, które są zrozumiałe dla szerokiego odbiorcy, a jednocześnie docenią je smakosze. Na bazie doświadczeń z prowadzenia tych zajęć powstały Wykresy od kuchni. To zbiór krótkich wykładów omawiających różne wątki przydatne w lepszym zrozumieniu tego, jak działa komunikacja z użyciem wykresów statystycznych. Na kolejnych stronach pojawi się wiele analogii do przyrządzania posiłków, ponieważ zarówno w kuchni, jak i w przygotowaniu wykresów statystycznych potrzebna jest praktyka, znajomość pewnych fundamentalnych prawideł, garść sprawdzonych przepisów i dużo zapału do eksperymentowania. Będąc tak uzbrojonym, każdy adept sztuki kulinarnej jest skazany na sukces. Flipbook online [POL] The Hitchhiker’s Guide to Responsible Machine Learning EN: A one-of-a-kind 52-page story about responsible machine learning. Beta and Bit use decision trees, random forests, and AutoML tools to build a risk model after a covid infection, and then use explainable artificial intelligence tools to analyze the behavior of that model. The description of the data analysis process is intertwined with descriptions of ML tools and code snippets. All examples are fully reproducible! PL: Jedyna w swoim rodzaju 52-stronicowa opowieść o odpowiedzialnym uczeniu maszynowym. Beta i Bit używają drzew decyzyjnych, lasów losowych i narzędzi AutoML do budowy modelu ryzyka po zakażeniu covid, a następnie używają narzędzi wyjaśnialnej sztucznej inteligencji by przeanalizować działanie tego modelu. Opis procesu analizy danych przeplata się na opisem kolejnych narzędzi i przykładami kodu. Wszystkie wyniki są całkowicie odtwarzalne! Flipbook online Przemysław Biecek, Anna Kozak, Aleksander Zawada Fundacja Naukowa SmarterPoland.pl. 2022 W pogoni za nieskończonością. Szeregi EN: What does hiking in the mountains have to do with the convergence of series? Quite a lot! We start with the paradoxes related to infinity, but step by step we learn the techniques of geometric series. In this book, the conditions for convergence are explained, together with numerous examples. The comic ends with a collection of exercises with different levels of difficulty. PL: Co wspólnego ma chodzenie po górach ze zbieżnością szeregów? Otóż całkiem sporo! Zaczynamy od paradoksów związanych z nieskończonością, ale krok po kroku poznajemy techniki szeregów geometrycznych. W tej pozycji wyjaśnione są warunki zbieżności wraz z licznymi przykładami. Komiks kończy zbiór zadań o różnych poziomach trudności. Flipbook online Przemysław Biecek, Łukasz Maciejewski, Aleksander Zawada Fundacja Naukowa SmarterPoland.pl. 2022 Przewodnik po pakiecie R EN: The Guide to the R package was the first published Polish book focused on the R language. The current fourth edition consists of four parts: Basics of using R (+tidyverse, shiny, knitr and other goodies), Programming in R (object-oriented, package development, class system), Statistics with R (statistical tests, models, exploration techniques) and Visualization with R (graphics, lattice and ggplot2 packages). PL: Przewodnik po pakiecie R był pierwszą wydaną polskojęzyczną książką poświęconą językowi R. Aktualne czwarte wydanie składa się z czterech części: Podstaw posługiwania się językiem R (+tidyverse, shiny, knitr i inne smaczki), Programowanie w R (obiektowe, tworzenie pakietów, system klas), Statystyka z R (testy statystyczne, modele, techniki eksploracji) i Wizualizacja z R (pakiety graphics, lattice i ggplot2). Wersja online, Książka w księgarnii. Przemysław Biecek Wydawnictwo GiS. 2008-2021 Analiza danych z programem R EN: An academic textbook describing estimation and testing topics for linear models with fixed effects, random effects and mixed effects. The theoretical introduction is complemented by numerous examples for one-way and multivariate ANOVA, one and multiple random components. The examples focus on biological and medical applications and are based on real analyses of real data. PL: Podręcznik akademicki opisujący zagadnienia estymacji i testowania dla modeli liniowych z efektami stałymi, losowymi i mieszanymi. Wprowadzenie teoretyczne jest uzupełnione o liczne przykłady dla jednokierunkowej i wielokierunkowej ANOVA, jednym i wieloma komponentami losowymi. Przykłady dotyczą głównie zastosowań biologicznych i medycznych i bazują na prawdziwych analizach rzeczywistych danych. Książka w księgarnii. Przemysław Biecek Wydawnictwo Naukowe PWN 2013-2018 Eseje o sztuce wizualizacji danych EN: Discover! Reveal! Explain! These three roles can be fulfilled by good statistical graphics. Good means understandable, faithful to the data, aesthetic. How to create such graphics? A collection of essays on the art of displaying data systematises knowledge useful in designing and producing good data visualisations. It is not easy. On the one hand, we can fall into the trap of a colourful mush full of numbers, which is sometimes proudly called infographics. On the other hand, we can fall into the trap of graphics that perfectly reproduce the complexity of numbers, and thus completely incomprehensible. Somewhere in the middle is a graphic that explains, that informs, that is aesthetically pleasing and informative. PL: Odkrywać! Ujawniać! Objaśniać! Te trzy role może spełniać dobra grafika statystyczna. Dobra czyli zrozumiała, wierna danym, estetyczna. Jak tworzyć taką grafikę? Zbiór esejów o sztuce pokazywania danych systematyzuje wiedzę przydatną do projektowania i wykonania dobrej wizualizacji danych. Nie jest to proste. Z jednej strony możemy wpaść w pułapkę pstrokatej papki najeżonej liczbami, którą czasem dumnie nazywa się infografiką. Z drugiej strony wpaść można w pułapkę grafiki idealnie odwzorowującej złożoność liczb a przez to zupełnie niezrozumiałej. Gdzieś po środku jest grafika, która wyjaśnia, która informuje, która jest estetyczna i informatywna. Książka online, Książka w księgarnii. Przemysław Biecek Wydawnictwo SmarterPoland 2008-2021 Pogromcy Danych EN: Data Crunchers is the first MOOC (Massive Open Online Course) developed in Polish for data scientists. Two modules were developed in 2015: the first one is an introduction to R, with loading data, overview of syntax, basic data types, descriptive statistics and pipelined processing. The second module is dedicated to data visualisation and statistical modelling. More than 8,000 people have registered on the Data Crunchers platform. PL: Pogromcy Danych to pierwszy MOOC (Massive Open Online Course) opracowany w języku polskim do analizy danych. W roku 2015 powstały dwa moduły: pierwszy jest wprowadzeniem do programu R, przez wczytywanie danych, omówienie składni, podstawowych typów danych, statystyk opisowych oraz przetwarzania potokowego. Drugi moduł poświęcony jest wizualizacji danych oraz modelowaniu statystycznemu. W platformie Pogromców Danych zarejestrowało się ponad 8000 osób. Przetwarzanie danych w programie R, Wizualizacja i modelowanie, Strona WWW. Przemysław Biecek ICM UW. 2015 Wykresy unplugged EN: Can you create clear charts without any electricity? An illustrated collection of exercises showing eight of the most popular ways to visualise data, with do-it-yourself challenges. Grab your crayons and start creating fantastic charts. PL: Czy można tworzyć czytelne wykresy bez użycia prądu? Ilustrowany zbiór ćwiczeń przedstawiających osiem najpopularniejszych sposobów wizualizacji danych, wraz z zadaniami do samodzielnego wykonania. Weź kredki i zacznij tworzyć fantastyczne wykresy. Flipbook online, Komiks w księgarnii. Przemysław Biecek, Ewa Baranowska, Piotr Sobczyk Fundacja Naukowa SmarterPoland.pl. 2018 W pogoni za nieskończonością EN: Two mathematicians share stories about infinity. In the first Beta attends a lecture on the properties of prime numbers. In the second, Bit breaks into the Palace of Culture and Science. How should we talk about mathematics? PL: Dwójka matematyków wymienia się opowiadaniami o nieskończoności. W pierwszym Beta bierze udział w wykładzie o właściwościach liczb pierwszych. W drugim Bit włamuje się do Pałacu Kultury i Nauki. Jak opowiadać o matematyce? Flipbook online, Komiks w księgarnii. Przemysław Biecek, Łukasz Maciejewski, Tomasz Samojlik, Sebastian Szpakowski Fundacja Naukowa SmarterPoland.pl. 2018 Jak długo żyją Muffinki? EN: A collection of three stories for children showing statistical relationships in the world around us. Beautifully illustrated stories about the distribution of height according to age, the life span of dogs or measuring the weight of trees. PL: Zbiór trzech opowiadań dla dzieci pokazującym zależności statystyczne w świecie wokół nas. Pięknie ilustrowane opowiadania o rozkładzie wzrostu w zależności od wieku, czasie życia psów czy pomiarze wagi drzew. Online: Jak szybko urosnę, Jak długo żyją Muffinki. Przemysław Biecek Fundacja Naukowa SmarterPoland.pl.2016 Pieczara Pietraszki EN: How linear regression can help in getting home, and why it’s not worth hacking into a mad mathematician’s office. A short story describing the adventures of two teenagers Beta and Bit moving around historic Warsaw. PL: W jaki sposób regresja liniowa może pomóc w powrocie do domu, oraz dlaczego nie warto włamywać się do pokoju szalonego matematyka? Lekkie opowiadanie opisujące przygody dwójki nastolatków Bety i Bita w historycznej Warszawie. Online: W jezyku Polskim, In English, По-Русски. Magda Chudzian, Przemysław Biecek Fundacja Naukowa SmarterPoland.pl. 2015 How to weight a dog with a ruler? EN: Workshop materials for children aged 8-10. Kids measure different parameters of their body, such as arm span or height. Then they create a graph summarizing the collected data and look for relations between the measured features. It just so happens that parts of the human body are proportional to each other and you can use a ruler to find this relationship. Part of the StatTub project. PL: Materiały do warsztaty dla dzieci w wieku 8-10. Dzieci mierzą różne parametry swojego ciała, takie jak rozpiętość ramion lub wzrost. Następnie tworzą wykres podsumowujący zebrane dane i szukają zależności pomiędzy zmierzonymi cechami. Tak się składa, że części ciała ludzkiego są do siebie proporcjonalne i można z użyciem linijki znaleźć tę relację. Część projektu StatTuba. Online: English, Polish, Chinese, Simplified Chinese, Czech, German, Spanish, Spanish (Latin America), French, Dutch, Vietnamese. Przemysław Biecek, Klaudia Korniluk Fundacja Naukowa SmarterPoland.pl.2016-2021 "],["responsibleml-blog.html", "ResponsibleML Blog", " ResponsibleML Blog Read more about the research, solutions and education on our blog: Tools for Explainable, Fair and Responsible ML BASIC XAI with DALEX— Part 1: Introduction Anna Kozak In this post, we will take a closer look at some algorithms used in explainable artificial intelligence. You will find here an introduction to methods of global and local model evaluation. Each description will include a technical introduction, example analysis, and code in R and Python. R packages for eXplainable Artificial Intelligence Przemysław Biecek We have prepared an overview of the most popular R-packages, which can be used to build interpretable models or to explore complex ones. Examples of knitr notebooks for more than 30 packages are available at http://xai-tools.drwhy.ai/. Adversarial attacks on Explainable AI Hubert Baniecki There are various adversarial attacks on machine learning models; hence, ways of defending, e.g. by using Explainable AI methods. Nowadays, attacks on model explanations come to light, so does the defense to such adversary. Here, we introduce fundamental concepts related to the domain. A further reference list is available at https://github.com/hbaniecki/adversarial-explainable-ai. "],["404.html", "Page not found", " Page not found The page you requested cannot be found (perhaps it was moved or renamed). You may want to try searching to find the page's new location, or use the table of contents to find the page you are looking for. "]]
diff --git a/docs/the-team.html b/docs/the-team.html
index a9286f7..a6044f0 100644
--- a/docs/the-team.html
+++ b/docs/the-team.html
@@ -204,10 +204,11 @@ <h4>Members<a href="the-team.html#members" class="anchor-section" aria-label="An
 <li><a href="the-team.html#przemysław-biecek">Przemysław Biecek</a>, PhD, DSc (Team Leader)</li>
 <li><a href="the-team.html#hubert-baniecki">Hubert Baniecki</a>, PhD student</li>
 <li><a href="the-team.html#mustafa-cavus">Mustafa Cavus</a>, PhD</li>
-<li><a href="the-team.html#maciej-chrabąszcz">Maciej Chrabąszcz</a>, PhD student</li>
+<li><a href="the-team.html#maciej-chrabąszcz">Maciej Chrabąszcz</a>, MSc student</li>
 <li><a href="the-team.html#mateusz-grzyb">Mateusz Grzyb</a>, MSc student</li>
 <li><a href="the-team.html#stanisław-giziński">Stanisław Giziński</a>, MSc student</li>
 <li><a href="the-team.html#weronika-hryniewska-guzik">Weronika Hryniewska-Guzik</a>, PhD student</li>
+<li><a href="the-team.html#paulina-kaczyńska">Paulina Kaczyńska</a>, MSc student</li>
 <li><a href="the-team.html#piotr-komorowski">Piotr Komorowski</a>, MSc student</li>
 <li><a href="the-team.html#mateusz-krzyziński">Mateusz Krzyziński</a>, MSc student</li>
 <li><a href="the-team.html#tymoteusz-kwiecinski">Tymoteusz Kwieciński</a>, BSc student</li>
@@ -301,7 +302,7 @@ <h3>Hubert Baniecki<a href="the-team.html#hubert-baniecki" class="anchor-section
 <div id="mustafa-cavus" class="section level3 unnumbered hasAnchor">
 <h3>Mustafa Cavus<a href="the-team.html#mustafa-cavus" class="anchor-section" aria-label="Anchor link to header"></a></h3>
 <p><img src="images/mcavus.jpg" ></p>
-<p>I work as an assistant professor at <a href="http://www.mini.pw.edu.pl/">Warsaw University of Technology</a> and the <a href="https://www.eskisehir.edu.tr/">Eskisehir Technical University</a>. I joined the MI² DataLab as a post-doc researcher in 2021. I work on explainable artificial intelligence and AutoML.</p>
+<p>I work as an assistant professor at the <a href="https://www.eskisehir.edu.tr/">Eskisehir Technical University</a>. I joined the MI² DataLab as a post-doc researcher in 2021. I work on glocal explanations and imbalanced learning.</p>
 <p>Google Scholar: <a href="https://scholar.google.com.tr/citations?user=I63d1WIAAAAJ&amp;hl">I63d1WIAAAAJ&amp;hl</a></p>
 <p>GitHub: <a href="https://github.com/mcavs">mcavus</a></p>
 <p>LinkedIn: <a href="https://www.linkedin.com/in/mustafacavusphd">mcavus</a></p>
@@ -317,7 +318,7 @@ <h3>Julian Sienkiewicz<a href="the-team.html#julian-sienkiewicz" class="anchor-s
 <div id="maciej-chrabąszcz" class="section level3 unnumbered hasAnchor">
 <h3>Maciej Chrabąszcz<a href="the-team.html#maciej-chrabąszcz" class="anchor-section" aria-label="Anchor link to header"></a></h3>
 <p><img src="images/maciej_chrabaszcz.jpg"></p>
-<p>I am pursuing a PhD in Computer Science at Warsaw University of Technology, where I also obtained my MSc in Mathematical Statistics in 2023. My main research interests lie in the fields of responsible and explainable machine learning, with a focus on Red Teaming foundation models.</p>
+<p>Master’s student in mathematical statisctics at Warsaw University of Technology. Interested in deep learning on text and images, explainable and responsible AI.</p>
 <p>GitHub: <a href="https://github.com/maciejchrabaszcz">maciejchrabaszcz</a></p>
 <p>LinkedIn: <a href="https://www.linkedin.com/in/maciej-chrabaszcz/">maciej-chrabaszcz</a></p>
 </div>
@@ -343,6 +344,12 @@ <h3>Weronika Hryniewska-Guzik<a href="the-team.html#weronika-hryniewska-guzik" c
 <p>GitHub: <a href="https://github.com/Hryniewska">Hryniewska</a></p>
 <p>LinkedIn: <a href="https://www.linkedin.com/in/weronikahryniewska">weronikahryniewska</a></p>
 </div>
+<div id="paulina-kaczyńska" class="section level3 unnumbered hasAnchor">
+<h3>Paulina Kaczyńska<a href="the-team.html#paulina-kaczyńska" class="anchor-section" aria-label="Anchor link to header"></a></h3>
+<p><img src="images/paulina_kaczy%C5%84ska.jpg" ></p>
+<p>I am working towards a Master’s degree in Machine Learning at University of Warsaw. I am interested in Natural Language Processing and ML applications in social sciences</p>
+<p>GitHub: <a href="https://github.com/Kaczyniec">Kaczyniec</a></p>
+</div>
 <div id="piotr-komorowski" class="section level3 unnumbered hasAnchor">
 <h3>Piotr Komorowski<a href="the-team.html#piotr-komorowski" class="anchor-section" aria-label="Anchor link to header"></a></h3>
 <p><img src="images/pkomorowski.jpg" ></p>