-
Notifications
You must be signed in to change notification settings - Fork 0
/
ccspectools.py
239 lines (187 loc) · 9.05 KB
/
ccspectools.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
from scipy import stats
import shutil
import importlib
import xspec
import os
import numpy as np
from IPython.display import Image
def view_group_spectra(r, reference_instrument, subcases_pattern, systematic_fraction=0.):
# I add response and arf keywords for easier reading
import astropy.io.fits as pf
if reference_instrument.lower() == 'spi':
ff=pf.open(r["reference_spec"], mode='update')
try:
reference_exposure=ff[2].header['EXPOSURE']
ff[2].header['RESPFILE']=r["reference_rmf"]
except:
try:
reference_exposure=ff[1].header['EXPOSURE']
ff[1].header['RESPFILE']=r["reference_rmf"]
except:
raise ValueError("Exposure should be in SPI spectrum!")
reference_times='N/A'
ff.flush()
ff.close()
elif reference_instrument.lower() == 'nustar':
for x in ['A', 'B']:
ff=pf.open(r["reference_spec_"+x], mode='update')
ff[1].header['RESPFILE']=r["reference_rmf_"+x]
ff[1].header['ANCRFILE']=r["reference_arf_"+x]
ff[1].header['BACKFILE']=r["reference_bkg_"+x]
reference_exposure=ff[1].header['EXPOSURE']
mjdref=float(ff[1].header['MJDREFI'])+float(ff[1].header['MJDREFF'])
tstart=float(ff[1].header['TSTART'])/86400. + mjdref
tstop=float(ff[1].header['TSTOP'])/86400. + mjdref
reference_times='%.4f--%.4f'%(tstart,tstop)
#reference_date=ff[1].header['DATE-OBS'] + '--'+ ff[1].header['DATE-END']
ff.flush()
ff.close()
elif reference_instrument.lower().replace("'","") == 'none':
reference_exposure='N/A'
reference_times='N/A'
else:
print("Undefined reference instrument")
raise ValueError
ff=pf.open(r["isgri_spec"], mode='update')
ff[1].header['RESPFILE']=r["isgri_rmf"]
ff[1].header['ANCRFILE']=r["isgri_arf"]
ff[1].data['SYS_ERR']=systematic_fraction
mjdref=51544.
tstart=float(ff[1].header['TSTART']) + mjdref
tstop=float(ff[1].header['TSTOP']) + mjdref
isgri_times='%.4f--%.4f'%(tstart,tstop)
isgri_exposure=ff[1].header['EXPOSURE']
ff.flush()
ff.writeto('isgri_spec'+subcases_pattern+'_'+isgri_times+'.fits', overwrite=True)
ff.close()
shutil.copy(r["isgri_rmf"],'isgri_rmf'+subcases_pattern+'_'+isgri_times+'.fits' )
shutil.copy(r["isgri_arf"],'isgri_arf'+subcases_pattern+'_'+isgri_times+'.fits' )
print("reference exposure",reference_exposure,"isgri exposure",isgri_exposure, "isgri time range", isgri_times, reference_times)
return dict(
reference_exposure=reference_exposure,
isgri_exposure=isgri_exposure,
isgri_times=isgri_times,
reference_times=reference_times,
)
def basic_consistency(fit_by_lt, nh_sig_limit):
# marginally more adcanced choice of low thresholds
print("lt\tchi2_r\tprob\tsigmas")
for lt,d in fit_by_lt.items():
d['nh_prob']=stats.chi2(d['ndof']).sf(d['chi2'])
d['nh_sig']=stats.norm().isf(d['nh_prob'])
print("%.1f\t%.2f\t%.2f\t%.2f"%(float(lt), d['chi2_red'], d['nh_prob'], d['nh_sig']) )
try:
good_lt = min([p for p in fit_by_lt.items() if p[1]['nh_sig'] < nh_sig_limit])
except:
good_lt = None
best_lt = min([p for p in fit_by_lt.items()], key=lambda x:x[1]['chi2_red'])
if good_lt != best_lt:
print("\033[31mNote that in this case, best LT does not coincde with the good LT. We want broader energy range\033[0m")
if good_lt is not None:
return good_lt
else:
return best_lt
def fit(data, reference_instrument, model_setter, emin_values, fn_prefix="", systematic_fraction=0):
importlib.reload(xspec)
fit_by_lt = {}
fn_by_lt={}
xspec.AllModels.systematic=systematic_fraction
for c_emin in emin_values:
xspec.AllData.clear()
xspec.AllModels.clear()
isgri, ref_ind = model_setter(data, reference_instrument, c_emin)
max_chi=np.ceil(xspec.Fit.statistic / xspec.Fit.dof)
m1=xspec.AllModels(1)
if ref_ind is not None:
n_spec = 2
if isinstance(ref_ind, list):
n_spec = len(ref_ind) + 1
xspec.Fit.error("1.0 max %.1f 1-%d"%(max_chi, n_spec*m1.nParameters))
ref = ref_ind[0]
else:
xspec.Fit.error("1.0 max %.1f 1-%d" % (max_chi,m1.nParameters))
models={}
lt_key='%.10lg'%c_emin
fit_by_lt[lt_key]=dict(
emin=c_emin,
chi2_red=xspec.Fit.statistic/xspec.Fit.dof,
chi2=xspec.Fit.statistic,
ndof=xspec.Fit.dof,
models=models,
)
for m, ss in (isgri, 'isgri'), (ref, 'ref'):
if m is None: continue
#initialize dictionaries
models[ss]={}
#models[ss]['flux']={}
#fills dictionaries
for i in range(1,m.nParameters+1):
if (not m(i).frozen) and (not bool(m(i).link)):
#use the name plus position because there could be parameters with same name from multiple
#model components (e.g., several gaussians)
print(m(i).name, "%.2f"%(m(i).values[0]), m(i).frozen,bool(m(i).link) )
models[ss][m(i).name+"_%02d"%(i)]=[ m(i).values[0], m(i).error[0], m(i).error[1] ]
# for flux_e1, flux_e2 in [(30,80), (80,200)]:
# xspec.AllModels.calcFlux("{} {} err".format(flux_e1, flux_e2))
# #print ( xspec.AllData(1).flux)
# models['isgri']['flux'][(flux_e1, flux_e2)] = xspec.AllData(1).flux
# models['ref']['flux'][(flux_e1, flux_e2)] = xspec.AllData(2).flux
xcmfile="saved"+fn_prefix+"_"+reference_instrument+".xcm"
if os.path.exists(xcmfile):
os.remove(xcmfile)
xspec.XspecSettings.save(xspec.XspecSettings, xcmfile, info='a')
xspec.Plot.device="/png"
#xspec.Plot.addCommand("setplot en")
xspec.Plot.xAxis="keV"
xspec.Plot("ldata del")
xspec.Plot.device="/png"
fn="fit_lt%.5lg.png"%c_emin
fn_by_lt[lt_key] = fn
shutil.move("pgplot.png_2", fn)
_=display(Image(filename=fn,format="png"))
return fit_by_lt, fn_by_lt
def parameter_comparison(good_lt, ng_sig_limit, reference_instrument, flux_tolerance=0.1):
#compare parameters of best fit
parameter_comparison={}
best_result=good_lt[1]
for par_name, par_values_ref in best_result['models']['ref'].items():
try:
par_values_isgri=best_result['models']['isgri'][par_name]
except:
print("parameter %s is not in isgri parameter list"%(par_name))
continue
isgri_value = par_values_isgri[0]
ref_value = par_values_ref[0]
if ref_value > isgri_value:
err_ref = ref_value - par_values_ref[1]
err_isgri = par_values_isgri[2] - isgri_value
else:
err_ref = - (ref_value - par_values_ref[2])
err_isgri = - (par_values_isgri[1] - isgri_value)
par_diff_sigmas = (isgri_value - ref_value) / np.sqrt(err_ref**2 + err_isgri**2)
print(par_name + " %.2f +/- %.2f ; %.2f +/- %.2f ; %.1f" % (
isgri_value, err_isgri, ref_value, err_ref, par_diff_sigmas))
#We do not compare the normalization with NuSTAR strictly, because of non-simultaneity issues
if (('g10Flux' in par_name) or ('norm' in par_name)) and (reference_instrument.lower() == 'nustar') :
print("Not comparing %s for %s"%(par_name, reference_instrument))
success = True
elif 'g10Flux' in par_name and np.abs(par_diff_sigmas) > ng_sig_limit:
frac_difference = np.abs( 1 - 10**(isgri_value - ref_value))
print("Using a fractional difference of %f to compare %s for %s" % (frac_difference,
par_name, reference_instrument))
success = bool(frac_difference < flux_tolerance)
elif 'norm' in par_name and np.abs(par_diff_sigmas) > ng_sig_limit:
frac_difference = np.abs( 1 - isgri_value / ref_value)
print("Using a fractional difference of %f to compare %s for %s" % (frac_difference,
par_name, reference_instrument))
success = bool(frac_difference < flux_tolerance)
else:
#print("standard comparison for %s"%par_name)
success = bool(np.abs(par_diff_sigmas) < ng_sig_limit)
parameter_comparison[par_name] = dict(
ref = [ref_value, err_ref],
isgri = [isgri_value, err_isgri ],
significance = np.abs(par_diff_sigmas),
success=success
)
return parameter_comparison