From 90149463cd38822ab160415939a8db4a14e2a9e9 Mon Sep 17 00:00:00 2001
From: Dhvani Patel <dhvanipatel_2000@yahoo.ca>
Date: Thu, 27 Jul 2017 13:53:45 -0600
Subject: [PATCH] Update formatting for testing model

---
 __pycache__/toCheck.pypy-41.pyc | Bin 2337 -> 2604 bytes
 predict_file.py                 |  99 ++++++++++++++++++++++++++------
 py_mutations_hub.pyc            | Bin 10233 -> 10628 bytes
 test_file.py                    |   2 +-
 4 files changed, 81 insertions(+), 20 deletions(-)

diff --git a/__pycache__/toCheck.pypy-41.pyc b/__pycache__/toCheck.pypy-41.pyc
index 0c426ee9297581ca5bd9cd370b7f8ff18c0d319a..462cb2f93cfeef90406e38e7c6a5f31b6924e241 100644
GIT binary patch
literal 2604
zcmcIm+in|G6x}nv#CDP<ZRsVXrIr9;gwpf@DkP#Jl!P`Cvcx1&sb5AD&v82Gc*e{*
z7NknVL-_>W_#xci;R9f;Jx-iNh)PJ<o;~|q_ie3n-1>8|_2|~?7k!z2YSXv<1xk#c
zC#gtJ<aqh4=}FI*=h=O!nu@$?N%56<o;RKAb1?Jd2ey4lt1_%fuO>QPk+d$uIqA*G
zupzw$daGhYj+!#?B(BM?B5y=q^jfsL+o{8c<S%@>9S^FbbK`Lmf8O!gtUyGgkz*r1
zaM$sAW}?C!n8@uJ>oy~6&I98HoM^v*fXyBZ3Jt<v7^705$5g5kS5&D<Tveqm!ET4<
z``m|570)IfdjjF`^W;c(+JIlvhA$h?CRKSWf@z`05L}%xzCr&S`(dfgv4$i%>8L@Q
zkCBYHF7cck!mEZn=9DI<#7&7?WgA%%WN97_bln`}-uCX^PWP!>g#OL`-m@^+{(gTa
z*xpJSbdu@Ji{voLG<AbK*Y^QZ##A0euN|Xu(V?}=P>|a+=|?Wfv-KmJXLbRdn|bC;
z=GNTt$f#*uvsa8wpf#eAex41ILoL_0QEH8&iw+eD`)Pv3YR<YMK=-t&qmd~Bu60d_
zmuF?vu+WN4VM`3qLu=AO2X9f~TF_j+`O@@XJscgYS=ihkEph4%&hr<&MQ_Q!=U2Rz
zce_&Y&dW$hXc@0TK^Z3=i$b*Ss=cm4El|ACPK+pu{5XscbU3()zL_h*d)!5njR>~#
zpRzVOELV%KHt9Y~r8a*@#!%QJH46{|egN~NqMbkudP=<p-9Po%KdH+Jpf*R8y!FN7
zf`)cAWdcNSDu1W`*=(S)E-Qf23T@5nDsHqOC)1%uIRw~4<E9*F%F1bJ{eN_CQBGiE
z>IN*T@iHUVh;xkIVK7%hFN1hT=>M|!(mVH701)q8%d)S}5XlP0k&6N%E15u8J(t{I
z?aP2?kit0NDm|H!*b@G;9oCaS(C(07C1tqdD7Q|TM4~B#<MD80%MEEsITTrI3b*~^
zI7+o`UF60#?B}u3HEABjHn_{_<Wg)nQqqcKD3^bi*y$FG@4}W=d3k8uba85&>Cibv
zxx%_Ic7%h#ls35jJTYMsUnPQainKqU^buCrHiT5xyqn%Cz7PD?RKHuft$lneS;!mO
zSF(Hx<sw;TumIA*2tB=tpP1a`f!4KmATK07&W+7-XMRQolHWkSGCH1H!G{$0DL#NG
zSx<5YAG3+46eYto!<PaJv2QrY$65R}rdoO}{*5vW`ZD7@8W+R(FPxvQ$4#MZzsipK
z`N7cC%(P@O&#UCp!BDavd@nDQx0Pw)^ftMId9C&ZykEw%RXoo~ph$GNi&9)y$c3JX
zY_`u~VB1gHk2<`jlmWr})K)3(Q7DTD`dLBB`aEDClOMDp^qwp6`?SEFZwf8DqE*YL
zQ?J&JMb*0qK%W8AZlPN*I7KE(M|5^rHU)kdX3@}up<Xs&ILza5ie_*N69U34_=JMS
zg3lmkR#j8j1Ox(u*7Vl>=JjT^x!SCi*JuL+<1BIXO|95uI7$uA15J$*erGt!aR3wr
zBr(Q+Ljm5RE20#qL-Y{$c9~YawYcxuzajPDpIp-zH_E7<aG51{7jR!L<K|rUmvL+U
E4btY(;Q#;t

literal 2337
zcmZuxSyLNF5bjwa0Wz4&#`t7?#FkA=QgPy!ToxcwiiIGnb%ggtS!qUit<~(-&d4B_
z{p7qSzbaMuf8-bB2PEB%WK7Bu8v33-zMkSgv&FsK>*pO9eF~%h#&-ZD{t1Wyf)BKS
zK(O3_*nxfyf*kZGAeaC#vPMy!EpiYSz#mB`A((`r&x#3%r=UL#LGgprIe0k(!7Ru;
z1m{5(Ko#K~K(7GVpDdgMT>yzu<{-jkra(@^<6s_|`bU^h3!oOE>p-sv+20_9z_ckE
z+G8~KQcDyNJ=6s_z6i1i?*#}gU|*ht-%+y+!*9kFUxMHgsAclV<^?6S0(yyEFN2(6
zR|>m=_ut|B>E@sDVlfIo`;iS@qH=J21>|`W2UlUl&M|upB&L<4YJ=-z)dH(-jM+tI
zSI6uUv+kI^!0b&>tC0RB)O9|u@p%J+Tc9RDUW6Xb={BgFkll8MYoP9c%ER$z@E+hD
z4s{uNdC2}1605(<?Oq|GsV`xL_qeyK4)vDBRFuZ_0FFx(At-^m3fZ@gybL-A@(TH?
zyAa$5d6gUdfWXu!T|<OL|Kg`!LWqH#!Xu4C+fA~*okXS|9+~L1DhZ?w9S@ka=a4*5
zw$%<hJ*8ztDoc2cbR21guo&5|v9up%J(Za+P`VlFd)AHmX_8sDAKEldY#g=M-DX0H
zwsOOEtX!M8q*>=>pxh{%O_&Gs9Esr*uy~<>mkt&3&VfHl@D?RlEJ|{4fae6fL3HN9
zPQc4NcGY<+powt)WxuJX2$+~!WB(0ydlGsMG*HzjQ3uTDczDVVGHu+b>+TKr-N?9|
zB+FFC#zQyK#)fgMq?=@}POPN_58Eg0lU;0}Ix^EJd-(7bmK`OUyI`zL26jxUEUR;$
zN>myq36P{pbF9R0_oaHJ2XSndQD+A2G)p?lm?Z1GP@P_@gDKeJ85PHF_*=Wre49tj
zb{HGQqxEQF*f3n85+1+@d)s7_!mX!6&fh9yTj}uV1{#?eGIes|qF56-k$22C^3)Eg
zFjc8^CsM`@Geuqas4X|$y;y}tjYY1nj%v{>Mg7`|dcN9UH{WBZGho2DpFqI?qohU-
z`f2Gtxu>#w!mM+Jh_=BIH;bCrHj?RXrWDbHNUiLclwo!lY4$S%jZLN=q2$q5RV0AA
zbMS~i?h|thxt*w!B_`=w_vy|<ztZsCT9Tv$dbcdY)Z%cjoH6^qnx~04&Y@MDe934=
z5@_0+_%lm*Ae4^<pAKy=zDf<=vU-W!xFAGairBN~QGk1gFj84J3;Q@Mno(zf(A1VC
zr{iVHTc8+oc%(QdZCl-e?pTCm9YuzK3!=1!r$<oo=tbac&CqyAgU72)my92V7OpEl
z0HzbB!*#@Ql6kY_M-fg}+N0o%-BVfXqesJ1W~V9yAZ<L)Y(!G|X>2*}PTAk9?fSLq
zhFwJQDch>nYt?#X%U(Y#-l}YupVj;pd9|LFy`LJKbd_J83MI|0^9I^ReRU$BvB(Jf
zOxqGC2dy{>Wuy;yJ0smqJi43vgDb-S_nz<bgbu?@lO->t4Kr&FBm068hUq;!G1}d$
z)HfPi?6B=sDnt`T>lJ@YZk4@ewayle-CEhJHkd}))^4-GSCT2cF4acs@z)zwzrnY3
zF;~mBnd0TMG?Uuy?%qau^J%MI-l=eMzlQN(R4&1Ppb<sf(hsK9x?l0O%bOK%6%9U#
zrzuo?*H7diRzFYy<~KZ6#Il$XMP~+IjCnC93iz9L3Sz-oLYltQNKY4YXf=(RSy2?T
e_$}t<&~gTUC*Ku}P{8+Q1#Q^EnGuV*g7Y73J^8o*

diff --git a/predict_file.py b/predict_file.py
index fb68c1a..a83ff5f 100644
--- a/predict_file.py
+++ b/predict_file.py
@@ -15,9 +15,11 @@
 import keyword
 import json
 import numpy as np
+from numpy import newaxis
 
 global all_tokens
 global indexed_tokens
+global check_tokens
 
 START_TOKEN = '<s>'
 END_TOKEN = '</s>'
@@ -183,8 +185,8 @@ def handle_token(type, token, (srow, scol), (erow, ecol), line):
     send = Token(tokenize.tok_name[type], val, srow, scol, erow, ecol, line)
     global all_tokens
     all_tokens.append(send)
-    #print "%d,%d-%d,%d:\t%s\t%s" % \
-    #    (srow, scol, erow, ecol, tokenize.tok_name[type], repr(token))
+    print "%d,%d-%d,%d:\t%s\t%s" % \
+        (srow, scol, erow, ecol, tokenize.tok_name[type], repr(token))
 
 def getFileTokens(fileName):
 	with open(fileName, 'r') as myfile:
@@ -199,20 +201,31 @@ def getFileTokens(fileName):
 	except tokenize.TokenError:
    		pass
 	one_hot_file = vocabularize_tokens(all_tokens, False)
+	global check_tokens
+	check_tokens = []
 
+	
+	#print len(one_hot_file)-10
+	#print "GOTCH U"
 	windowInd = 0
 	while windowInd <= int((int(len(one_hot_file)) - 10)):
 		toPass = []
 		for x in range(10):
 			y = x + windowInd
-			print y
-			print len(one_hot_file)
+			#print y
+			#print len(one_hot_file)
 			toPass.append(one_hot_file[y])
+		assert len(toPass) > 0
 		a = np.array(toPass).astype(int)
-		assert len(a) == 10
-		print a
-		yield a
+		b = a[newaxis, :]
+		#print b
+		check_tokens.append(toPass)
+		assert len(b[0]) == 10
+		#print a
+		yield b
 		windowInd += 1
+		#print "WINDOW"
+		#print windowInd
 	print "done radha"
 	#print len(one_hot_file)
 
@@ -232,8 +245,17 @@ def predict(fileName):
 	opt = optimizers.RMSprop(lr=0.001, rho=0.9, epsilon=1e-08, decay=0.5)
 	loaded_model.compile(loss='categorical_crossentropy', optimizer=opt, metrics=['accuracy'])
 
-	outPredict = loaded_model.predict_generator(getFileTokens(fileName), 225, 3, verbose=1)
+	'''
+	gen = getFileTokens(fileName)
+	arrs = []	
+	for x in gen:
+		arrs.append(x)
+	print len(arrs)
+	print type(radha)
+	'''
 
+	outPredict = loaded_model.predict_generator(getFileTokens(fileName), 227, 1, verbose=1)
+	print "HERE"
 	print outPredict
 	print len(outPredict)	
 
@@ -246,26 +268,65 @@ def predict(fileName):
         countGood = -1
         countIns = -1
 	countDel = -1
+	countWhat = -1
+	iterInd = 0
+	print inds
+	print zip(*(iter(inds),) * 10)
+
+
+	global check_tokens
+	check = check_tokens[:]
+	print len(check)
+	#print check[225]
 	iterInd = 0
+	with open('vocabulary.json') as data_file:    
+    		data = json.load(data_file)
+	print "-------------------------------"
+	for window in check:
+		errType = inds[iterInd]
+		if errType == 0:
+			msg = "NO ERROR: "
+		elif errType == 2:
+			msg = "DELETION: "
+		elif errType == 3:
+			msg = "INSERTION: "
+		else:
+			msg = "IDEK: "
+		errLine = ""
+		for toks in window:
+			getInd = toks.index(1.0)
+			actualToken = data["indexes"][getInd]
+			errLine = errLine + ' ' + actualToken
+		print msg + errLine
+		#print c	
+		#print type(radha)
+		iterInd += 1
+	print "-------------------------------"
+	sys.exit()
+	#print type(radha)
 	for b in inds:
-		if iterInd == 3:
-			iterInd = 0
+		#if iterInd == 3:
+		#	iterInd = 0
 		
-		if iterInd == 0:
-               		if b == 0:
-				countGood += 1
-		if iterInd == 1: 		
-			if b == 3:
-				countIns += 1
-		if iterInd == 2:
-			if b == 2:
-				countDel += 1
+		#if iterInd == 0:
+               	if b == 0:
+			countGood += 1
+		if b == 1:
+			countWhat += 1
+		#if iterInd == 1: 		
+		if b == 3:
+			countIns += 1
+		#if iterInd == 2:
+		if b == 2:
+			countDel += 1
+		#print b
 		iterInd += 1
 		#print b
         print len(inds)
 	print countGood
 	print countIns
 	print countDel	
+	print countWhat
 
 if __name__ == '__main__':
 	fileName = sys.argv[1]
diff --git a/py_mutations_hub.pyc b/py_mutations_hub.pyc
index 317011ad63863bef5d96bc80a0b46ef5c37562da..abeee6aaeb2ca8ea417b775ebdfd78466945578e 100644
GIT binary patch
delta 1062
zcmZuvIcyU_6n%euBykLw;3UpLF2@J=k#NKuf+XT-La>X#O0byB5ZQ6!#NI%-lp+XG
zAR)9=6cp45B(_782#JCSXppEV=qNyf2GQ_mk{}9pHE-Vh{qx_O`TOhCtz4V$$C^+u
za_#ImRvsVEM|e~p-EaG{07yb91}g>`g4F_9533b2466+jfYokrqrncSPFP)#YhZOl
zu7%ZOFm58fkn3P2AS1AnCZ2*^4=WAX0BaNE2HfI;n<1lyY=LZCFl;MilOfw6n+@3x
z88c)DWQ!p?AzKaE1=(iEZm2%Uc1#LX9t!QXAM;0{_CRjLB4PMkAFRDldvW{(g9Ijh
zsC*-}OUF3D+J^<zkK%r)eR@sP!0byJQv;A4keyf*urg3F6c2zCRYpgMr{X~@g4yon
zsTsvXhRshH`OvcbSuHQSmgSc{bZZagMWOxlFdtl9#Q*uw`wvfOUU9`=)5c^*jImel
zTrQpT=ESg|e(#0<t)L<APR$6>Mek$n(*VtSx7Q8|dhET4yrJMH-Snip6lwLIH@u>B
zn0OuLpbw4IL`e~+gc#=_GCb;d4DqO5bst4f)13P~+DX!jG=8n^o~Xmk0o)isj(d9i
z#0@aWI^?t`u&1co)bZO(b?zj}cxSr$eDzL{OF8Aye6{M%`iI=5csLl-`qKLxKTiu=
zy2JipfC9q&rrLvlRli;D1d0Sl=p##^yTI+bxuT|j=#grKF<&a#GZkgayp!MV%sadR
zs9a3z45x-+e6%n=SyI^+=GkV3w3c#eLX5n5nD=RW#~gk7iNa-d(os{@*IcHmOM#qP
zM%$js7p85y@Sbb_Us#vD;*KTablAO~`00PgYj-;N;KCA<><5OGO4Io%W!uiGd+pqI
z%JvLjRJLB{Ht1Y)fbwWo--YpQ3|=afRL&_^l%p?8&P~l^@41Q8CA#e1P4&1>QvLB~
bb^b5~X#)i)%x{#|QG`NNBSIz?5C`cG!;jd^

delta 711
zcmYk3%WG3X6vn@EuQxW2rZ!DXle9_N#59SInum{yf>fc!MM@i`WWZjcb6e7qM3V>=
zgf7~Dpp1JF1i^(McjMBH%Tl@&+=zmLE76U(sB@=PIxye)aps)w44kd%@41-%v)`i+
zJg(ix;W;Mbb$PpYwqsv5JYd7%Ua$zb4=jomfW?%KC>@0r2TOqWgC)WJU}H+hRVD>K
z0G0+n0d`8|GvI?@6W{@`Ecg(%q+t#`s2~p>Qg9mlq=XHyGvHyBItxCm;2d~F!6bN8
z!FgB(@EF!Kbmlc@SP|=&VU@r~ut~UiHxE{ZRmRM$6b`KE=<JbIge7~+PR%9R(*hp8
zq(zyQbz!Vt*uW-0>nqC2b6{0-toZ1G)q!8!@oQhv9Ph*tP9EX0BRuYK=a%+=I60Gl
z(4q{SP|c>~0vh}K`}TY7nnopCH+D3dw)c)#i0XFI{o0@h_O5?gb3J!fl^34^QM)ks
zis*y=G4PqlE8Ybo{wewBmiLBJ)D>TY^VAkop#&}2wb1vz*s>c{xq_M5Cy>H}s5&2j
z{6I)p@joG_w?R@WmCC<ZvCTw*Xv*GBmUOY3nsbF^MfaYaPA^d;m(UF|j9#B?GR<ba
z(K5}(Hysd9Gby?#-e!J@HxnYYBi|2;TJ@%7nmw1ybRRWuw|fg-i3<VoBD+kh!kZfx
YleuE#nU7pFL<adNOor}JY-m^MFC6HV2mk;8

diff --git a/test_file.py b/test_file.py
index 20542cd..d60f3bc 100644
--- a/test_file.py
+++ b/test_file.py
@@ -13,7 +13,7 @@
     "entity_id": "%s/idp.xml" % BASE,
     "interaction": [
         {
-            "matches":
+            "matches": 
                 "url": "%s/login" % BASE,
                 "title": 'IDP test login'
             },