-
Notifications
You must be signed in to change notification settings - Fork 2
/
make_mock.py
206 lines (164 loc) · 8.24 KB
/
make_mock.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
from __init__ import *
from main import *
import numpy as np
import os
def make_mock(numbener, numbtime, strgmult):
gdat = gdatstrt()
gdatnotp = gdatstrt()
gdat.boolspre = True
np.random.seed(0)
gdat.sizeregi = None
gdat.diagmode = False
gdat.booltile = True
print 'Generating mock data...'
for strgtype in ['sing', 'nomi', 'feww']:
if strgtype == 'sing':
boolcent = True
minf = np.float32(10000.)
gdat.numbstar = 1
elif strgtype == 'nomi':
boolcent = False
minf = np.float32(1000.)
gdat.numbstar = 20
elif strgtype == 'nomi':
boolcent = False
minf = np.float32(1000.)
gdat.numbstar = 4
gdat.stdvlcpr = 1e-6
#gdat.stdvcolr = np.array([0.05])
#gdat.meancolr = np.array([0.])
gdat.stdvcolr = np.array([0.05, 0.05])
gdat.meancolr = np.array([0., 0.])
slop = np.float32(2.0)
gdat.sizeimag = [100, 100]
gdat.numbsidexpos = gdat.sizeimag[0]
gdat.numbsideypos = gdat.sizeimag[1]
if boolcent and gdat.numbstar != 1:
raise Exception('')
#listdims = [[1, 1], [3, 1], [1, 3], [2, 2]]
listdims = [[1, 3]]
strgdata = 'sdss0921'
strgpsfn = 'sdss0921'
gdat.pathlion = os.environ['LION_PATH'] + '/'
pathdata = os.environ['LION_DATA_PATH'] + '/'
# setup
gdat.boolplotsave = False
setp(gdat)
setp_clib(gdat, gdatnotp, gdat.pathlion)
gdat.verbtype = 1
# get the 3-band PSF
filepsfn = open(pathdata + 'idR-002583-2-0136-psfg.txt')
numbsidepsfn, factusam = [np.int32(i) for i in filepsfn.readline().split()]
filepsfn.close()
numbsidepsfnusam = numbsidepsfn * factusam
gdat.cntppsfn = np.empty((3, numbsidepsfnusam, numbsidepsfnusam))
gdat.cntppsfn[0, :, :] = np.loadtxt(pathdata + 'idR-002583-2-0136-psfg.txt', skiprows=1).astype(np.float32)
gdat.cntppsfn[1, :, :] = np.loadtxt(pathdata + 'idR-002583-2-0136-psfr.txt', skiprows=1).astype(np.float32)
gdat.cntppsfn[2, :, :] = np.loadtxt(pathdata + 'idR-002583-2-0136-psfi.txt', skiprows=1).astype(np.float32)
for numbener, numbtime in listdims:
if numbener > 3:
raise Exception('')
gdat.numbener = numbener
gdat.numbtime = numbtime
gdat.indxener = np.arange(gdat.numbener)
gdat.indxtime = np.arange(gdat.numbtime)
gdat.numbcolr = gdat.numbener - 1
gdat.numblcpr = gdat.numbtime - 1
print 'numbener'
print numbener
print 'numbtime'
print numbtime
cntppsfn = gdat.cntppsfn[:gdat.numbener, :, :]
coefspix = psf_poly_fit(gdat, cntppsfn, factusam)
print 'coefspix'
summgene(coefspix)
indxstar = np.arange(gdat.numbstar)
# background
trueback = np.float32(179.)
# gain
gain = np.float32(4.62)
# position
if boolcent:
xpos = (np.array([0.5]) * (gdat.sizeimag[0] - 1)).astype(np.float32)
ypos = (np.array([0.5]) * (gdat.sizeimag[0] - 1)).astype(np.float32)
else:
xpos = (np.random.uniform(size=gdat.numbstar)*(gdat.sizeimag[0]-1)).astype(np.float32)
ypos = (np.random.uniform(size=gdat.numbstar)*(gdat.sizeimag[1]-1)).astype(np.float32)
# flux
fluxsumm = minf * np.exp(np.random.exponential(scale=1./(slop-1.), size=gdat.numbstar).astype(np.float32))
flux = np.ones((gdat.numbener, gdat.numbtime, gdat.numbstar), dtype=np.float32)
# temp
flux *= fluxsumm
if gdat.numbener > 1:
# spectral parameters
colr = gdat.stdvcolr[:gdat.numbener-1, None] * np.random.randn(gdat.numbcolr * gdat.numbstar).reshape((gdat.numbcolr, gdat.numbstar)).astype(np.float32) + \
gdat.meancolr[:gdat.numbener-1, None]
#print 'colr'
#summgene(colr)
#print 'gdat.numbener'
#print gdat.numbener
#print 'fluxsumm'
#summgene(fluxsumm)
#print 'flux'
#summgene(flux)
#print 'colr[:, None, :]'
#summgene(colr[:, None, :])
#print 'flux[1:, :, :]'
#summgene(flux[1:, :, :])
# temp
for t in gdat.indxtime:
flux[1:, t, :] *= 10**(0.4*colr)
if gdat.numbtime > 1:
# temporal parameters
#arry = np.linspace(0., 1. - 1. / gdat.numblcpr, gdat.numblcpr)
#temp = np.hstack(arry, gdat.numblcpr)
#temp = (1e-6 * np.random.randn((gdat.numblcpr * gdat.numbstar)) + np.tile(np.linspace(0., 1. - 1. / gdat.numblcpr, gdat.numblcpr), (1, gdat.numbstar))) % 1.
#temp = temp.reshape((gdat.numblcpr, gdat.numbstar)).astype(np.float32)
temp = np.random.random((gdat.numblcpr * gdat.numbstar)).reshape((gdat.numblcpr, gdat.numbstar)).astype(np.float32)
temp = np.sort(temp, axis=0)
temptemp = np.concatenate([np.zeros((1, gdat.numbstar), dtype=np.float32)] + [temp] + [np.ones((1, gdat.numbstar), dtype=np.float32)], axis=0)
difftemp = temptemp[1:, :] - temptemp[:-1, :]
#for k in range(10):
# if (np.sum(difftemp[:, k]) != 1).any():
# print 'temp[:, k]'
# print temp[:, k]
# print 'temptemp[:, k]'
# print temptemp[:, k]
# print 'np.sum(difftemp[:, k])'
# print np.sum(difftemp[:, k])
#
#assert (np.sum(difftemp, axis=0) == 1.).all()
flux[:, :, :] *= difftemp[None, :, :]
# inject transits
indxstartran = np.random.choice(indxstar, size=gdat.numbstar/2, replace=False)
for k in indxstartran:
indxinit = np.random.choice(gdat.indxtime)
indxtemp = np.arange(indxinit, indxinit + 4) % gdat.numblcpr
flux[:, indxtemp, k] *= np.random.rand()
#flux[:, 1:, :] = fluxsumm[None, None, :] * lcpr[None, :, :]
print 'flux'
for a in range(flux.shape[2]):
print flux[:, :, a].flatten()
print
# evaluate model
cntpdata = eval_modl(gdat, xpos, ypos, flux, trueback, numbsidepsfn, coefspix, clib=gdatnotp.clib.clib_eval_modl, sizeimag=gdat.sizeimag)
if not np.isfinite(cntpdata).all():
raise Exception('')
cntpdata[cntpdata < 1] = 1.
# add noise
vari = cntpdata / gain
cntpdata += (np.sqrt(vari) * np.random.normal(size=(gdat.numbener, gdat.sizeimag[1], gdat.sizeimag[0], gdat.numbtime))).astype(np.float32)
if not np.isfinite(cntpdata).all():
raise Exception('')
# write to file
path = pathdata + strgdata + '_%04d%04d_%s_mock.h5' % (gdat.numbener, gdat.numbtime, strgtype)
print 'Writing to %s...' % path
filetemp = h5py.File(path, 'w')
filetemp.create_dataset('cntpdata', data=cntpdata)
filetemp.create_dataset('numb', data=gdat.numbstar)
filetemp.create_dataset('xpos', data=xpos)
filetemp.create_dataset('ypos', data=ypos)
filetemp.create_dataset('flux', data=flux)
filetemp.create_dataset('gain', data=gain)
filetemp.close()
print