-
Notifications
You must be signed in to change notification settings - Fork 1
/
inv_cov.py
executable file
·177 lines (144 loc) · 5.55 KB
/
inv_cov.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
#!/usr/bin/python
import sys
import math, numpy as np
import matplotlib.pyplot as plt
import matplotlib.image as mpimg
from numpy import linalg as LA
import numpy as np
#uncomment below if you want to invert the DESC SRD Y10 covariance
# infile =['cov/Y10_3x2pt_clusterN_clusterWL_cov']
# data= ['datav/3x2pt_clusterN_clusterWL_Y10_fid']
# outname=['Y10']
# nggl = 25 # number of ggl power spectra
# ngcl = 11 # number of cluster-source galaxy power spectra
# nlens = 10 # number of lens bins
# nlenscl= 4 # number of cluster redshift bins
# nshear = 15 # number of shear tomographic power spectra
# ncl=20 # number of ell-bins
# nclgcl=5 # number of cluster ell-bins
# nrich=5 # number of richness bins
#uncomment below if you want to invert the DESC SRD Y1 covariance
infile =['cov/Y1_3x2pt_clusterN_clusterWL_cov']
data= ['datav/3x2pt_clusterN_clusterWL_Y1_fid']
outname=['Y1']
nggl = 7 # number of ggl power spectra
ngcl = 6 # number of cluster-source galaxy power spectra
nlens = 5 # number of lens bins
nlenscl= 3 # number of cluster redshift bins
nshear = 15 # number of shear tomographic power spectra
ncl=20 # number of ell-bins
nclgcl=5 # number of cluster ell-bins
nrich=5 # number of richness bins
ndata = (nshear+nggl+nlens)*ncl+nlenscl*nrich+nrich*ngcl*nclgcl
n2pt = (nshear+nggl+nlens)*ncl
ncluster = nlenscl*nrich
n2ptcl=n2pt+ncluster
nclusterN_WL=ncluster+nrich*ngcl*nclgcl
for k in range(0,1):
datafile= np.genfromtxt(data[k])
mask = np.zeros(ndata)
for i in range(0,datafile.shape[0]):
if (datafile[i,1] >1.0e-15):
mask[i]=1.0
covfile = np.genfromtxt(infile[k])
cov = np.zeros((ndata,ndata))
print ndata,n2pt,int(np.max(covfile[:,0])+1)
for i in range(0,covfile.shape[0]):
cov[int(covfile[i,0]),int(covfile[i,1])] = covfile[i,8]+covfile[i,9]
cov[int(covfile[i,1]),int(covfile[i,0])] = covfile[i,8]+covfile[i,9]
cor = np.zeros((ndata,ndata))
for i in range(0,ndata):
for j in range(0,ndata):
if (cov[i,i]*cov[j,j] >0):
cor[i,j] = cov[i,j]/math.sqrt(cov[i,i]*cov[j,j])
a = np.sort(LA.eigvals(cor[:,:]))
print "min+max eigenvalues full cor:"
print np.min(a), np.max(a)
print "neg eigenvalues full cor:"
for i in range(0,a.shape[0]):
if (a[i]< 0.0): print a[i]
# ############### invert shear covariance #################
inv = LA.inv(cov[0:nshear*ncl,0:nshear*ncl])
a = np.sort(LA.eigvals(cov[0:nshear*ncl,0:nshear*ncl]))
print "min+max eigenvalues shear cov:"
print np.min(a), np.max(a)
outfile = "cov/"+outname[k]+"_shear_shear_inv"
f = open(outfile, "w")
for i in range(0,nshear*ncl):
inv[i,i]=inv[i,i]*mask[i]
for j in range(0,nshear*ncl):
f.write("%d %d %e\n" %(i,j, inv[i,j]))
f.close()
# ############### invert clustering covariance #################
inv = LA.inv(cov[(nshear+nggl)*ncl:(nshear+nggl+nlens)*ncl,(nshear+nggl)*ncl:(nshear+nggl+nlens)*ncl])
a = np.sort(LA.eigvals(cov[(nshear+nggl)*ncl:(nshear+nggl+nlens)*ncl,(nshear+nggl)*ncl:(nshear+nggl+nlens)*ncl]))
print "min+max eigenvalues clustering cov:"
print np.min(a), np.max(a)
outfile = "cov/"+outname[k]+"_pos_pos_inv"
f = open(outfile, "w")
for i in range(0,nlens*ncl):
inv[i,i]=inv[i,i]*mask[(nshear+nggl)*ncl+i]
for j in range(0,nlens*ncl):
f.write("%d %d %e\n" %(i,j, inv[i,j]))
f.close()
# ############### invert 2pt covariance #################
a = np.sort(LA.eigvals(cov[0:n2pt,0:n2pt]))
print "min+max eigenvalues 2pt cov:"
print np.min(a), np.max(a)
inv = LA.inv(cov[0:n2pt,0:n2pt])
outfile = "cov/"+outname[k]+"_3x2pt_inv"
f = open(outfile, "w")
for i in range(0,n2pt):
inv[i,i]=inv[i,i]*mask[i]
for j in range(0,n2pt):
f.write("%d %d %e\n" %( i,j, inv[i,j]))
f.close()
# # ############### invert full2pt+clusterN+clusterWL covariance #################
precond = 1.e-7
for i in range(0,ncluster):
cov[n2pt+i,:]*= precond
cov[:,n2pt+i]*= precond
inv = LA.inv(cov)
a = np.sort(LA.eigvals(cov))
print "min+max eigenvalues of full 2ptclusterN+clusterWL pre-conditioned matrix:"
print np.min(a), np.max(a)
if (np.min(a)<0):
print "WARNING WARNING: %s is not positive definite! WARNING!" % (infile[k])
for i in range(0,ncluster):
inv[n2pt+i,:]*= precond
inv[:,n2pt+i]*= precond
outfile = "cov/"+outname[k]+"_3x2pt_clusterN_clusterWL_inv"
f = open(outfile, "w")
for i in range(0,ndata):
inv[i,i]=inv[i,i]*mask[i]
for j in range(0,ndata):
f.write("%d %d %e\n" %( i,j, inv[i,j]))
f.close()
# # ############### invert clusterN+clusterWL covariance #################
inv = LA.inv(cov[n2pt:n2pt+nclusterN_WL,n2pt:n2pt+nclusterN_WL])
a = np.sort(LA.eigvals(cov[n2pt:n2pt+nclusterN_WL,n2pt:n2pt+nclusterN_WL]))
print "min+max eigenvalues of clusterN_WL pre-conditioned matrix:"
print np.min(a), np.max(a)
if (np.min(a)<0):
print "WARNING WARNING: %s is not positive definite! WARNING!" % (infile[k])
for i in range(0,ncluster):
inv[i,:]*= precond
inv[:,i]*= precond
outfile = "cov/"+outname[k]+"_clusterN_clusterWL_inv"
f = open(outfile, "w")
for i in range(0,nclusterN_WL):
inv[i,i]=inv[i,i]*mask[n2pt+i]
for j in range(0,nclusterN_WL):
f.write("%d %d %e\n" %( i,j, inv[i,j]))
f.close()
cor = np.zeros((ndata,ndata))
for i in range(0,ndata):
for j in range(0,ndata):
if (cov[i,i]*cov[j,j] >0):
cor[i,j] = cov[i,j]/math.sqrt(cov[i,i]*cov[j,j])
plt.figure()
#plt.imshow(np.log10(cov[0:1500,2000:]), interpolation="nearest",vmin=-25, vmax=-10)
plt.imshow(np.log10(cov[:,:]), interpolation="nearest",vmin=-25, vmax=-10)
#plt.imshow(cor[n2ptcl:n2ptcl+200,300:nshear*ncl], interpolation="nearest",vmax=0.5)
plt.colorbar()
plt.show()