/
invert_sed.py
353 lines (272 loc) · 12.7 KB
/
invert_sed.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
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
import pdb
import numpy as np
import cPickle as pickle
from utils import black
from utils import loggen
#from astropy.cosmology import FlatLambdaCDM
from astropy.cosmology import Planck15 as cosmo
import astropy.units as u
from lmfit import Parameters, minimize, fit_report
L_sun = 3.839e26 # W
c = 299792458.0 # m/s
def find_nearest_index(array_in,value):
ng = len(value)
#idx = (np.abs(array_in-value)).argmin()
idx = (np.abs(array_in-np.reshape(value,(ng,1)))).argmin(axis=1)
return idx
def sed(p, nu_in, T, betain, alphain):
'''
#m = [A, T, Beta, Alpha] - return SED (array) in Jy
#P[0] = A
#P[1] = T
#P[2] = Beta
#P[3] = Alpha
'''
v = p.valuesdict()
A0= v['Ain']
A=np.asarray(A0)
#T = v['Tin']
#betain = v['betain']
#alphain = v['alphain']
ng = np.size(A)
ns = len(nu_in)
base = 2.0 * (6.626)**(-2.0 - betain - alphain) * (1.38)**(3. + betain + alphain) / (2.99792458)**2.0
expo = 34.0 * (2.0 + betain + alphain) - 23.0 * (3.0 + betain + alphain) - 16.0 + 26.0
K = base * 10.0**expo
w_num = A * K * (T * (3.0 + betain + alphain))**(3.0 + betain + alphain)
w_den = (np.exp(3.0 + betain + alphain) - 1.0)
w_div = w_num/w_den
nu_cut = (3.0 + betain + alphain) * 0.208367e11 * T
graybody = np.reshape(A,(ng,1)) * nu_in**np.reshape(betain,(ng,1)) * black(nu_in, T) / 1000.0
powerlaw = np.reshape(w_div,(ng,1)) * nu_in**np.reshape(-1.0 * alphain,(ng,1))
graybody[np.where(nu_in >= np.reshape(nu_cut,(ng,1)))]=powerlaw[np.where(nu_in >= np.reshape(nu_cut,(ng,1)))]
return graybody
def sed_direct(A, nu_in, T, betain, alphain):
'''
'''
ng = np.size(A)
ns = len(nu_in)
base = 2.0 * (6.626)**(-2.0 - betain - alphain) * (1.38)**(3. + betain + alphain) / (2.99792458)**2.0
expo = 34.0 * (2.0 + betain + alphain) - 23.0 * (3.0 + betain + alphain) - 16.0 + 26.0
K = base * 10.0**expo
w_num = A * K * (T * (3.0 + betain + alphain))**(3.0 + betain + alphain)
w_den = (np.exp(3.0 + betain + alphain) - 1.0)
w_div = w_num/w_den
nu_cut = (3.0 + betain + alphain) * 0.208367e11 * T
#graybody = np.reshape(A,(ng,1)) * nu_in**np.reshape(np.repeat(betain,ng),[ng,1]) * black(nu_in, T) / 1000.0
#powerlaw = np.reshape(w_div,(ng,1)) * nu_in**np.reshape(np.repeat(alphain,ng),[ng,1])
graybody = np.reshape(A,(ng,1)) * nu_in**betain * black(nu_in, T) / 1000.0
powerlaw = np.reshape(w_div,(ng,1)) * nu_in**(-1.0 * alphain)
graybody[np.where(nu_in >= np.reshape(nu_cut,(ng,1)))]=powerlaw[np.where(nu_in >= np.reshape(nu_cut,(ng,1)))]
return graybody
def sedint(p, nu_in, Lir, T, betain, alphain):
'''
#m = [A, T, Beta, Alpha] - return integrated SED flux (one number) in Jy x Hz
#P[0] = A
#P[1] = T
#P[2] = Beta
#P[3] = Alpha
'''
v = p.valuesdict()
A0 = v['Ain']
A=np.asarray(A0)
#pdb.set_trace()
#T = v['Tin']
#betain = v['betain']
#alphain = v['alphain']
#print 'A is ' + str(A)
ns = len(nu_in)
#pdb.set_trace()
ng = np.size(A)
base = 2.0 * (6.626)**(-2.0 - betain - alphain) * (1.38)**(3. + betain + alphain) / (2.99792458)**2.0
expo = 34.0 * (2.0 + betain + alphain) - 23.0 * (3.0 + betain + alphain) - 16.0 + 26.0
K = base * 10.0**expo
w_num = A * K * (T * (3.0 + betain + alphain))**(3.0 + betain + alphain)
w_den = (np.exp(3.0 + betain + alphain) - 1.0)
w_div = w_num/w_den
nu_cut = (3.0 + betain + alphain) * 0.208367e11 * T
#nu_cut_ind = find_nearest_index(nu_in,nu_cut)
graybody = np.reshape(A,(ng,1)) * nu_in**np.reshape(betain,(ng,1)) * black(nu_in, T) / 1000.0
powerlaw = np.reshape(w_div,(ng,1)) * nu_in**np.reshape(-1.0 * alphain,(ng,1))
graybody[np.where(nu_in >= np.reshape(nu_cut,(ng,1)))]=powerlaw[np.where(nu_in >= np.reshape(nu_cut,(ng,1)))]
#pdb.set_trace()
dnu = nu_in[1:ns] - nu_in[0:ns-1]
dnu = np.append(dnu[0],dnu)
return np.ravel([np.sum(graybody * dnu, axis=1) - Lir])
def sedint2(p, nu_in, Lir, ng): # m = [A, T, Beta, Alpha] - return integrated SED flux (one number) in Jy x Hz
#P[0] = A
#P[1] = T
#P[2] = Beta
#P[3] = Alpha
v = p.valuesdict()
A = v['Ain']
T = v['Tin']
betain = v['betain']
alphain = v['alphain']
print 'A is ' + str(A)
ns = len(nu_in)
#ng = len(A)
base = 2.0 * (6.626)**(-2.0 - betain - alphain) * (1.38)**(3. + betain + alphain) / (2.99792458)**2.0
expo = 34.0 * (2.0 + betain + alphain) - 23.0 * (3.0 + betain + alphain) - 16.0 + 26.0
K = base * 10.0**expo
w_num = A * K * (T * (3.0 + betain + alphain))**(3.0 + betain + alphain)
w_den = (np.exp(3.0 + betain + alphain) - 1.0)
w_div = w_num/w_den
nu_cut = (3.0 + betain + alphain) * 0.208367e11 * T
#nu_cut_ind = find_nearest_index(nu_in,nu_cut)
graybody = np.reshape(A,(ng,1)) * nu_in**np.reshape(betain,(ng,1)) * black(nu_in, T) / 1000.0
powerlaw = np.reshape(w_div,(ng,1)) * nu_in**np.reshape(-1.0 * alphain,(ng,1))
graybody[np.where(nu_in >= np.reshape(nu_cut,(ng,1)))]=powerlaw[np.where(nu_in >= np.reshape(nu_cut,(ng,1)))]
#pdb.set_trace()
dnu = nu_in[1:ns] - nu_in[0:ns-1]
dnu = np.append(dnu[0],dnu)
return np.ravel([np.sum(graybody * dnu, axis=1) - Lir])
def simple_flux_from_greybody(lambdavector, Trf = None, b = None, Lrf = None, zin = None, ngal = None):
'''
Return flux densities at any wavelength of interest (in the range 1-10000 micron),
assuming a galaxy (at given redshift) graybody spectral energy distribution (SED),
with a power law replacing the Wien part of the spectrum to account for the
variability of dust temperatures within the galaxy. The two different functional
forms are stitched together by imposing that the two functions and their first
derivatives coincide. The code contains the nitty-gritty details explicitly.
Inputs:
alphain = spectral index of the power law replacing the Wien part of the spectrum, to account for the variability of dust temperatures within a galaxy [default = 2; see Blain 1999 and Blain et al. 2003]
betain = spectral index of the emissivity law for the graybody [default = 2; see Hildebrand 1985]
Trf = rest-frame temperature [in K; default = 20K]
Lrf = rest-frame FIR bolometric luminosity [in L_sun; default = 10^10]
zin = galaxy redshift [default = 0.001]
lambdavector = array of wavelengths of interest [in microns; default = (24, 70, 160, 250, 350, 500)];
AUTHOR:
Lorenzo Moncelsi [moncelsi@caltech.edu]
HISTORY:
20June2012: created in IDL
November2015: converted to Python
'''
nwv = len(lambdavector)
nuvector = c * 1.e6 / lambdavector # Hz
nsed = 1e4
lambda_mod = loggen(1e3, 8.0, nsed) # microns
nu_mod = c * 1.e6/lambda_mod # Hz
#Lorenzo's version had: H0=70.5, Omega_M=0.274, Omega_L=0.726 (Hinshaw et al. 2009)
#cosmo = Planck15#(H0 = 70.5 * u.km / u.s / u.Mpc, Om0 = 0.273)
conversion = 4.0 * np.pi *(1.0E-13 * cosmo.luminosity_distance(zin) * 3.08568025E22)**2.0 / L_sun # 4 * pi * D_L^2 units are L_sun/(Jy x Hz)
Lir = Lrf / conversion # Jy x Hz
Ain = np.zeros(ngal) + 1.0e-36 #good starting parameter
betain = np.zeros(ngal) + b
alphain= np.zeros(ngal) + 2.0
fit_params = Parameters()
fit_params.add('Ain', value= Ain)
#fit_params.add('Tin', value= Trf/(1.+zin), vary = False)
#fit_params.add('betain', value= b, vary = False)
#fit_params.add('alphain', value= alphain, vary = False)
#pdb.set_trace()
#THE LM FIT IS HERE
#Pfin = minimize(sedint, fit_params, args=(nu_mod,Lir.value,ngal))
Pfin = minimize(sedint, fit_params, args=(nu_mod,Lir.value,ngal,Trf/(1.+zin),b,alphain))
#pdb.set_trace()
flux_mJy=sed(Pfin.params,nuvector,ngal,Trf/(1.+zin),b,alphain)
return flux_mJy
def single_simple_flux_from_greybody(lambdavector, Trf = None, b = 2.0, Lrf = None, zin = None):
'''
Return flux densities at any wavelength of interest (in the range 1-10000 micron),
assuming a galaxy (at given redshift) graybody spectral energy distribution (SED),
with a power law replacing the Wien part of the spectrum to account for the
variability of dust temperatures within the galaxy. The two different functional
forms are stitched together by imposing that the two functions and their first
derivatives coincide. The code contains the nitty-gritty details explicitly.
Cosmology assumed: H0=70.5, Omega_M=0.274, Omega_L=0.726 (Hinshaw et al. 2009)
Inputs:
alphain = spectral index of the power law replacing the Wien part of the spectrum, to account for the variability of dust temperatures within a galaxy [default = 2; see Blain 1999 and Blain et al. 2003]
betain = spectral index of the emissivity law for the graybody [default = 2; see Hildebrand 1985]
Trf = rest-frame temperature [in K; default = 20K]
Lrf = rest-frame FIR bolometric luminosity [in L_sun; default = 10^10]
zin = galaxy redshift [default = 0.001]
lambdavector = array of wavelengths of interest [in microns; default = (24, 70, 160, 250, 350, 500)];
AUTHOR:
Lorenzo Moncelsi [moncelsi@caltech.edu]
HISTORY:
20June2012: created in IDL
November2015: converted to Python
'''
nwv = len(lambdavector)
nuvector = c * 1.e6 / lambdavector # Hz
nsed = 1e4
lambda_mod = loggen(1e3, 8.0, nsed) # microns
nu_mod = c * 1.e6/lambda_mod # Hz
#cosmo = Planck15#(H0 = 70.5 * u.km / u.s / u.Mpc, Om0 = 0.273)
conversion = 4.0 * np.pi *(1.0E-13 * cosmo.luminosity_distance(zin) * 3.08568025E22)**2.0 / L_sun # 4 * pi * D_L^2 units are L_sun/(Jy x Hz)
Lir = Lrf / conversion # Jy x Hz
Ain = 1.0e-36 #good starting parameter
betain = b
alphain= 2.0
fit_params = Parameters()
fit_params.add('Ain', value= Ain)
#THE LM FIT IS HERE
Pfin = minimize(sedint, fit_params, args=(nu_mod,Lir.value,Trf/(1.+zin),b,alphain))
flux_mJy=sed(Pfin.params,nuvector,Trf/(1.+zin),b,alphain)
return flux_mJy
def single_simple_rest_frame_flux_from_greybody(lambdavector, Trf = None, b = 2.0, Lrf = None, zin = None):
'''
Return flux densities at the rest-frame wavelength of interest (in the range 1-10000 micron),
assuming a galaxy (at given redshift) graybody spectral energy distribution (SED),
with a power law replacing the Wien part of the spectrum to account for the
variability of dust temperatures within the galaxy. The two different functional
forms are stitched together by imposing that the two functions and their first
derivatives coincide. The code contains the nitty-gritty details explicitly.
Cosmology assumed: H0=70.5, Omega_M=0.274, Omega_L=0.726 (Hinshaw et al. 2009)
Inputs:
alphain = spectral index of the power law replacing the Wien part of the spectrum, to account for the variability of dust temperatures within a galaxy [default = 2; see Blain 1999 and Blain et al. 2003]
betain = spectral index of the emissivity law for the graybody [default = 2; see Hildebrand 1985]
Trf = rest-frame temperature [in K; default = 20K]
Lrf = rest-frame FIR bolometric luminosity [in L_sun; default = 10^10]
zin = galaxy redshift [default = 0.001]
lambdavector = array of wavelengths of interest [in microns; default = (24, 70, 160, 250, 350, 500)];
AUTHOR:
Lorenzo Moncelsi [moncelsi@caltech.edu]
HISTORY:
20June2012: created in IDL
November2015: converted to Python
'''
nwv = len(lambdavector)
nuvector = (c * 1.e6 / lambdavector) / (1.+zin) # Hz
nsed = 1e4
lambda_mod = loggen(1e3, 8.0, nsed) # microns
nu_mod = c * 1.e6/lambda_mod # Hz
#cosmo = Planck15#(H0 = 70.5 * u.km / u.s / u.Mpc, Om0 = 0.273)
conversion = 4.0 * np.pi *(1.0E-13 * cosmo.luminosity_distance(zin) * 3.08568025E22)**2.0 / L_sun # 4 * pi * D_L^2 units are L_sun/(Jy x Hz)
Lir = Lrf / conversion # Jy x Hz
Ain = 1.0e-36 #good starting parameter
betain = b
alphain= 2.0
fit_params = Parameters()
fit_params.add('Ain', value= Ain)
#THE LM FIT IS HERE
Pfin = minimize(sedint, fit_params, args=(nu_mod,Lir.value,Trf/(1.+zin),b,alphain))
flux_mJy=sed(Pfin.params,nuvector,Trf/(1.+zin),b,alphain)
return flux_mJy
def amplitude_of_best_fit_greybody(Trf = None, b = 2.0, Lrf = None, zin = None):
'''
Same as single_simple_flux_from_greybody, but to made an amplitude lookup table
'''
nsed = 1e4
lambda_mod = loggen(1e3, 8.0, nsed) # microns
nu_mod = c * 1.e6/lambda_mod # Hz
#cosmo = Planck15#(H0 = 70.5 * u.km / u.s / u.Mpc, Om0 = 0.273)
conversion = 4.0 * np.pi *(1.0E-13 * cosmo.luminosity_distance(zin) * 3.08568025E22)**2.0 / L_sun # 4 * pi * D_L^2 units are L_sun/(Jy x Hz)
Lir = Lrf / conversion # Jy x Hz
Ain = 1.0e-36 #good starting parameter
betain = b
alphain= 2.0
fit_params = Parameters()
fit_params.add('Ain', value= Ain)
#THE LM FIT IS HERE
Pfin = minimize(sedint, fit_params, args=(nu_mod,Lir.value,Trf/(1.+zin),b,alphain))
#pdb.set_trace()
return Pfin.params['Ain'].value
def invert_sed_neural_net(lam, Trf, Lrf, zin, wpath = '/data/pickles/simstack/ann_function_fits/', wfile = 'SED_amplitude_weights_from_neural_network_logistic_100layers_N8000.p'):
reg = pickle.load( open( wpath + wfile, "rb" ) )
nuvector = c * 1.e6 / lam
rearrange_x = np.transpose(np.array([Trf, Lrf, zin]))
predicted_amplitude = 1e-40 * 10**reg.predict(rearrange_x)
fluxes = sed_direct(predicted_amplitude, np.array([nuvector]), Trf/(1.+zin), betain=2.0, alphain=2.0)
return fluxes