-
Notifications
You must be signed in to change notification settings - Fork 0
/
crosscheck_onlymanu.py
265 lines (191 loc) · 12.1 KB
/
crosscheck_onlymanu.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
import universe
#reload(universe)
from universe import *
import halo_fit
#reload(halo_fit)
from halo_fit import *
import weight
#reload(weight)
from weight import *
import pn_2d
#reload(pn_2d)
from pn_2d import *
import cmb
#reload(cmb)
from cmb import *
import flat_map
#reload(flat_map)
from flat_map import *
from mpi4py import MPI
comm = MPI.COMM_WORLD
rank = comm.Get_rank()
size = comm.Get_size()
mock_numb = 80#
delta = mock_numb/size #1 I am using size > mock_numb #mock_numb/size
start = 0
rank_ass = rank
print('Size', size)
print('Rank', rank)
iMax = (rank_ass+1)*delta+start #number 3 cmbset does not have temperature lensed
iMin = rank_ass*delta+start
#iMin, iMax = 0, 2
iMax = int(iMax)
iMin = int(iMin)
lMin = 30.; lMax = 4.5e3
# ell bins for power spectra
nBins = 21 # number of bins
lRange = (1., 2.*lMax) # range for power spectra
nu0 = 148
#cmb = StageIVCMB(beam = 1.4, noise = 6., lMin=lMin, lMaxT=lMax, lMaxP=lMax, atm=False)
cmb = CMB(beam=1.4, noise=6., nu1=148.e9, nu2=148.e9, lMin=lMin, lMaxT=lMax, lMaxP=1.e4, fg = True, atm = False, name = "cmbs4")
cmb.nu1 = 148.e9
cmb.nu2 = 148.e9
forCtotal = lambda l: cmb.ftotalTT(l)
# reinterpolate: gain factor 10 in speed
L = np.logspace(np.log10(lMin/2.), np.log10(2.*lMax), 1001, 10.)
F = np.array(list(map(forCtotal, L)))
cmb.fCtotal = interp1d(L, F, kind='linear', bounds_error=False, fill_value=0.)
ell = np.arange(1, 10000, 1)
if rank == 0:
ffunl = np.vectorize(lambda l: cmb.funlensedTT(l))
fflen = np.vectorize(lambda l: cmb.flensedTT(l))
ffdetectornoise = np.vectorize(lambda l: cmb.fdetectorNoise(l))
fftot = np.vectorize(forCtotal)
ffg = np.vectorize(lambda l: cmb.fkSZ(l) + cmb.fCIB(l) + cmb.ftSZ(l) + cmb.ftSZ_CIB(l) + cmb.fradioPoisson(l))
fftsz = np.vectorize(lambda l: cmb.ftSZ(l))
np.savetxt('spectra_lensqest_un_len_detectnoise_fftot_fg_ftSZ.txt', np.c_[ell, ffunl(ell), fflen(ell), ffdetectornoise(ell), fftot(ell), ffg(ell), fftsz(ell)])
np.savetxt('all_fgs_tsz_cib_cross_ksz_radio.txt', np.c_[ell, np.vectorize(lambda l: cmb.ftSZ(l))(ell), np.vectorize(lambda l: cmb.fCIB(l))(ell), np.vectorize(lambda l: cmb.ftSZ_CIB(l))(ell), np.vectorize(lambda l: cmb.fkSZ(l))(ell), np.vectorize(lambda l: cmb.fradioPoisson(l))(ell)])
##################################################################################
u = UnivPlanck15()
halofit = Halofit(u, save=False)
w_cmblens = WeightLensSingle(u, z_source=1100., name="cmblens")
p2d_cmblens = P2dAuto(u, halofit, w_cmblens, save=False)
nu = nu0*1.e9
print("My conversion agrees with the Lambda website recommendation:", 1.e-26 / cmb.dBdT(148.e9, cmb.Tcmb), cmb.Tcmb / 1.072480e9)
# cutout dimensions
# map side in lon and lat
dLon = 20.#10.# [deg]
dLat = 20.#10.# [deg]
lonRange = np.array([-dLon/2., dLon/2.]) # [deg]
latRange = np.array([-dLat/2., dLat/2.]) # [deg]
pixRes = 1./60. #0.5 / 60. # [arcmin] to [deg]
# number of pixels on the side
xSize = np.int(np.ceil(dLon / pixRes))
ySize = np.int(np.ceil(dLat / pixRes))
baseMap = FlatMap(nX=xSize, nY=ySize, sizeX=dLon*np.pi/180., sizeY=dLat*np.pi/180.)
if rank == 0:
fNqCmb_fft = baseMap.forecastN0Kappa(cmb.funlensedTT, cmb.fCtotal, lMin=lMin, lMax=lMax, test=False)
np.savetxt('QEnoise.txt', np.c_[ell, fNqCmb_fft(ell), p2d_cmblens.fPinterp(ell)])
for iPatch in range(iMin, iMax):
pp = "../manusmaps/"
# read Sehgal maps [Jy/sr]
path = pp+"flat_maps_large/newmaps11022021/148/sehgal_cib_148_large_cutout_"+str(iPatch)+".txt"
cib148 = np.genfromtxt(path)
path = pp+"flat_maps_large/newmaps11022021/148/sehgal_tsz_148_large_cutout_"+str(iPatch)+".txt"
#path = pp+"flat_maps_large/latestmaps13092020/sehgal_tsz_148_large_cutout_"+str(iPatch)+".txt"
tsz = np.genfromtxt(path)
path = pp+"flat_maps_large/newmaps11022021/148/sehgal_ksz_148_large_cutout_"+str(iPatch)+".txt"
ksz = np.genfromtxt(path)
path = pp+"flat_maps_large/newmaps11022021/148/sehgal_radiops_148_large_cutout_"+str(iPatch)+".txt"
radiops = np.genfromtxt(path)
# Rescale the maps to match Dunkley power spectra after masking
cib148 *= 0.38 # 0.35 * np.sqrt(1.2)
tsz *= 0.7 #0.82 # 0.68 * np.sqrt(1.45)
ksz *= 0.82 # 0.8 * np.sqrt(1.05)
radiops *= 1.1
# Sum of all the foregrounds
fTot = cib148 + tsz + ksz + radiops
#np.savetxt(pp+'flat_maps_large/latestmaps13092020/'+'sehgal_total_148_large_cutout_'+str(iPatch)+".txt", fTot)
# Fourier transform
tsz = tsz
tszFourier = baseMap.fourier(tsz)
fTotFourier = baseMap.fourier(fTot)
# Create a mask with the sum of the foregrounds as the input
fluxCut = 0.005 # in Jy
maskPatchRadius = 3. * np.pi/(180.*60.) # in rad
#cmb.ftotal
#forCtotal = lambda l: cmb.flensedTT(l)+cmb.fkSZ(l) + cmb.fCIB(l) + cmb.ftSZ(l) + cmb.ftSZ_CIB(l) + cmb.fradioPoisson(l)#cmb.ftotalTT(l)
#F = np.array(list(map(forCtotal, L)))
#forCtotal = interp1d(L, F, kind='linear', bounds_error=False, fill_value=0.)
profile = None #cmb.fbeam
maskTot = baseMap.pointSourceMaskMatchedFilterIsotropic(cmb.fCtotal, fluxCut, fprof=profile, dataFourier=fTotFourier, maskPatchRadius=maskPatchRadius, test=False)
np.savetxt(f"prova{iPatch}.txt", maskTot)
#np.savetxt(pp+'flat_maps_large/latestmaps13092020/'+'ps_mask_'+str(np.int(round(fluxCut*1000)))+"mJy_T_patch"+str(iPatch)+".txt", maskTot)
#maskTot = np.loadtxt(pp+'flat_maps_large/latestmaps13092020/'+'ps_mask_'+str(np.int(round(fluxCut*1000)))+"mJy_T_patch"+str(iPatch)+".txt")
#maskTot = 0.*maskTot+1.
tsz -= np.mean(tsz.flatten())
# convert from Jy/sr to muK
# consistent with Sehgal's conversion: https://lambda.gsfc.nasa.gov/toolbox/tb_sim_info.cfm
conversion = 1.e6 * 1.e-26 / cmb.dBdT(cmb.nu1, cmb.Tcmb)
tsz *= conversion
# mask all the maps
tsz *= maskTot
# Fourier transform
tszFourier = baseMap.fourier(tsz)
# randomize the phases to create "Gaussian" version of tSZ map
f = lambda lx,ly: np.exp(1j*np.random.uniform(0., 2.*np.pi))
tszGFourier = baseMap.filterFourier(f, dataFourier=tszFourier)
tszG = baseMap.inverseFourier(tszGFourier)
tszG *= maskTot
path = pp+"flat_maps_large/latestmaps13092020/gaussian_sehgal_tsz_148_large_cutout_"+str(iPatch)+".txt"
#np.savetxt(path, tszG)
tszGFourier = baseMap.fourier(tszG)
QE = 'QE'
SH = 'SH'
estimator = QE
# calculate the estimators
if estimator == QE:
pQFourier = baseMap.computeQuadEstKappaNorm(cmb.funlensedTT, cmb.fCtotal, lMin=lMin, lMax=lMax, dataFourier=tszFourier, test=False)
pQGFourier = baseMap.computeQuadEstKappaNorm(cmb.funlensedTT, cmb.fCtotal, lMin=lMin, lMax=lMax, dataFourier=tszGFourier, test=False)
elif estimator == SH:
pQFourier = baseMap.computeQuadEstKappaShearNormCorr(cmb.funlensedTT, cmb.fCtotal, lMin=lMin, lMax=lMax, dataFourier=tszFourier, test=False)
pQGFourier = baseMap.computeQuadEstKappaShearNormCorr(cmb.funlensedTT, cmb.fCtotal, lMin=lMin, lMax=lMax, dataFourier=tszGFourier, test=False)
#pQFourier = baseMap.computeQuadEstKappaPointSourceHardenedNorm(cmb.flensedTT, cmb.fCtotal, lMin=lMin, lMax=lMax, dataFourier=tszFourier, test=False)
#pQGFourier = baseMap.computeQuadEstKappaPointSourceHardenedNorm(cmb.flensedTT, cmb.fCtotal, lMin=lMin, lMax=lMax, dataFourier=tszGFourier, test=False)
# get the power spectra
lCen, ClQ, sCl = baseMap.powerSpectrum(pQFourier, plot=False, save=False, nBins=nBins)
lCen, ClQG, sCl = baseMap.powerSpectrum(pQGFourier, plot=False, save=False, nBins=nBins)
# this is the trispectrum bias
ClQ -= ClQG
#i = iPatch
#cmb0 = np.loadtxt('/global/cscratch1/sd/omard/extract_sehgal/manusmaps/flat_maps_large/cmb0/cmb0_'+str(iPatch)+'.txt')
#cmb1 = np.loadtxt('/global/cscratch1/sd/omard/extract_sehgal/manusmaps/flat_maps_large/cmb1/cmb1_'+str(iPatch)+'.txt')
#cmb0 = np.loadtxt(pp+'flat_maps_large/latestmaps13092020/cmb0_'+str(iPatch)+'.txt')#*maskTot
#cmb1 = np.loadtxt(pp+'flat_maps_large/latestmaps13092020/cmb1_'+str(iPatch)+'.txt')#*maskTot
#cmb0Fourier = baseMap.fourier(cmb0)
cmb0Fourier = baseMap.genGRF(cmb.funlensedTT) #baseMap.fourier(cmb0)
cmb0 = baseMap.inverseFourier(cmb0Fourier)
manusdir = pp+'flat_maps_large/'
kCmb = np.genfromtxt(manusdir+'sehgal_kcmb/sehgal_kcmb_large_cutout_'+str(iPatch)+'.txt')#*maskTot
kCmbFourier = baseMap.fourier(kCmb)
cmb1 = baseMap.doLensingTaylor(cmb0, kappaFourier=kCmbFourier, order=1) - cmb0
cmb1Fourier = baseMap.fourier(cmb1)
#np.savetxt(pp+'flat_maps_large/latestmaps13092020/cmb0_'+str(iPatch)+'.txt', cmb0)
#np.savetxt(pp+'flat_maps_large/latestmaps13092020/cmb1_'+str(iPatch)+'.txt', cmb1)
totalCmbFourier = tszFourier
if estimator == QE:
qCmbFourier0f = baseMap.computeQuadEstKappaNorm(cmb.funlensedTT, cmb.fCtotal, lMin=lMin, lMax=lMax, dataFourier = cmb0Fourier, dataFourier2 = totalCmbFourier, test=False)
qCmbFourierf0 = baseMap.computeQuadEstKappaNorm(cmb.funlensedTT, cmb.fCtotal, lMin=lMin, lMax=lMax, dataFourier = totalCmbFourier, dataFourier2 = cmb0Fourier, test=False)
qCmbFourier1f = baseMap.computeQuadEstKappaNorm(cmb.funlensedTT, cmb.fCtotal, lMin=lMin, lMax=lMax, dataFourier = cmb1Fourier, dataFourier2 = totalCmbFourier, test=False)
qCmbFourierf1 = baseMap.computeQuadEstKappaNorm(cmb.funlensedTT, cmb.fCtotal, lMin=lMin, lMax=lMax, dataFourier = totalCmbFourier, dataFourier2 = cmb1Fourier, test=False)
elif estimator == SH:
qCmbFourier0f = baseMap.computeQuadEstKappaShearNormCorr(cmb.funlensedTT, cmb.fCtotal, lMin=lMin, lMax=lMax, dataFourier = cmb0Fourier, dataFourier2 = totalCmbFourier, test=False)
qCmbFourierf0 = baseMap.computeQuadEstKappaShearNormCorr(cmb.funlensedTT, cmb.fCtotal, lMin=lMin, lMax=lMax, dataFourier = totalCmbFourier, dataFourier2 = cmb0Fourier, test=False)
qCmbFourier1f = baseMap.computeQuadEstKappaShearNormCorr(cmb.funlensedTT, cmb.fCtotal, lMin=lMin, lMax=lMax, dataFourier = cmb1Fourier, dataFourier2 = totalCmbFourier, test=False)
qCmbFourierf1 = baseMap.computeQuadEstKappaShearNormCorr(cmb.funlensedTT, cmb.fCtotal, lMin=lMin, lMax=lMax, dataFourier = totalCmbFourier, dataFourier2 = cmb1Fourier, test=False)
#qCmbFourier0f = baseMap.computeQuadEstKappaPointSourceHardenedNorm(cmb.funlensedTT, cmb.fCtotal, lMin=lMin, lMax=lMax, dataFourier = cmb0Fourier, dataFourier2 = totalCmbFourier, test=False)
#qCmbFourierf0 = baseMap.computeQuadEstKappaPointSourceHardenedNorm(cmb.funlensedTT, cmb.fCtotal, lMin=lMin, lMax=lMax, dataFourier = totalCmbFourier, dataFourier2 = cmb0Fourier, test=False)
#qCmbFourier1f = baseMap.computeQuadEstKappaPointSourceHardenedNorm(cmb.funlensedTT, cmb.fCtotal, lMin=lMin, lMax=lMax, dataFourier = cmb1Fourier, dataFourier2 = totalCmbFourier, test=False)
#qCmbFourierf1 = baseMap.computeQuadEstKappaPointSourceHardenedNorm(cmb.funlensedTT, cmb.fCtotal, lMin=lMin, lMax=lMax, dataFourier = totalCmbFourier, dataFourier2 = cmb1Fourier, test=False)
lCen, Clcrossf0f1, sCl = baseMap.crossPowerSpectrum(qCmbFourierf0, qCmbFourierf1, plot=False, save=False, nBins=nBins)
lCen, Clcross0f1f, sCl = baseMap.crossPowerSpectrum(qCmbFourier0f, qCmbFourier1f, plot=False, save=False, nBins=nBins)
lCen, Clcross0ff1, sCl = baseMap.crossPowerSpectrum(qCmbFourier0f, qCmbFourierf1, plot=False, save=False, nBins=nBins)
lCen, Clcross1ff0, sCl = baseMap.crossPowerSpectrum(qCmbFourier1f, qCmbFourierf0, plot=False, save=False, nBins=nBins)
lCen, clkk, sCl = baseMap.crossPowerSpectrum(kCmbFourier, kCmbFourier, plot=False, save=False, nBins=nBins)
#np.savetxt('out_txt/'+estimator+'_sec_0f1f_f0f1_0ff1_1ff0.txt'+str(iPatch)+'.txt', np.c_[Clcross0f1f, Clcrossf0f1, Clcross0ff1, Clcross1ff0])
#Clcross = (2*Clcross0f1f)+(2*Clcrossf0f1)+2*(Clcross0ff1+Clcross1ff0)
lCen, Clcross, sCl = baseMap.crossPowerSpectrum((qCmbFourier0f+qCmbFourierf0), (qCmbFourier1f+qCmbFourierf1), plot=False, save=False, nBins=nBins)
Clcross *= 2
lCen, Clprim, sCL = baseMap.crossPowerSpectrum(pQFourier, kCmbFourier, plot=False, save=False, nBins=nBins)
Clprim *= 2.
np.savetxt('out_txt/'+estimator+'_'+str(lMax)+'_tris_prim_sec_'+str(iPatch)+'.txt', np.c_[lCen, ClQ, Clprim, Clcross, clkk, p2d_cmblens.fPinterp(lCen)])