-
Notifications
You must be signed in to change notification settings - Fork 0
/
bandbyband_all_dlam_control.py
205 lines (157 loc) · 8.67 KB
/
bandbyband_all_dlam_control.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
import matplotlib
matplotlib.use('Agg') # Solves tkagg problem
import matplotlib.pyplot as plt
import numpy as np
import astropy.units as u
from emcee import EnsembleSampler
# from emcee.mpi_pool import MPIPool
# import sys
import os
import h5py
from corner import corner
from astropy.modeling.blackbody import blackbody_lambda
from toolkit import (get_slices_dlambdas, bands_TiO,
SimpleSpectrum, model_known_lambda,
plot_posterior_samples_for_paper)
archive = h5py.File('/Users/bmmorris/git/aesop/notebooks/spectra.hdf5', 'r+')
from json import load, dump
star_temps = load(open('star_temps.json', 'r'))
colors = load(open('colors.json', 'r'))
stars = load(open('stars_control.json', 'r'))
# Set additional width in angstroms centered on band core,
# used for wavelength calibration
roll_width = 15
bands = bands_TiO#[:-1]
yerr = 0.001
color_error = 2 * 0.003 #0.003
force_refit = False #True
# Set width where fitting will occur
fit_width = 0*u.Angstrom
path = 'bandbyband_control_results.json'
if os.path.exists(path) and not force_refit:
results = load(open(path, 'r'))
else:
results = dict()
for star in sorted(stars.keys()):
if star not in results or force_refit:
star_results = dict()
phot_temp = star_temps[star]
comparison_temp_high = star_temps[stars[star][0][0]]
comparison_temp_low = star_temps[stars[star][1][0]]
mixture_coefficient = 1 - ((phot_temp - comparison_temp_low) /
(comparison_temp_high - comparison_temp_low))
mixture_coefficient = np.max([0, mixture_coefficient])
print(star, phot_temp, comparison_temp_high, comparison_temp_low, mixture_coefficient)
# Book keeping:
target_name = star
comp1_name = stars[star][0][0]
comp1_time = stars[star][0][1]
comp2_name = stars[star][1][0]
comp2_time = stars[star][1][1]
target_temp = star_temps[target_name]
comp1_temp = star_temps[comp1_name]
comp2_temp = star_temps[comp2_name]
target_color = colors[target_name]
comp1_color = colors[comp1_name]
comp2_color = colors[comp2_name]
# Load spectra from database
times = list(archive[target_name])
for time in times[:1]:
if time not in star_results or force_refit:
spectrum1 = archive[target_name][time]
spectrum2 = archive[comp1_name][comp1_time]
spectrum3 = archive[comp2_name][comp2_time]
wavelength1 = spectrum1['wavelength'][:]
flux1 = spectrum1['flux'][:]
wavelength2 = spectrum2['wavelength'][:]
flux2 = spectrum2['flux'][:]
wavelength3 = spectrum3['wavelength'][:]
flux3 = spectrum3['flux'][:]
target = SimpleSpectrum(wavelength1, flux1, dispersion_unit=u.Angstrom)
source1 = SimpleSpectrum(wavelength2, flux2, dispersion_unit=u.Angstrom)
source2 = SimpleSpectrum(wavelength3, flux3, dispersion_unit=u.Angstrom)
# Slice the spectra into chunks centered on each TiO band:
slicesdlambdas = get_slices_dlambdas(bands, roll_width, target, source1, source2)
target_slices, source1_slices, source2_slices, source1_dlambdas, source2_dlambdas = slicesdlambdas
# source1_dlambdas = [0]*len(bands)
# source2_dlambdas = [0]*len(bands)
time_results = dict()
for inds, band in zip(target_slices.wavelength_splits, bands):
band_results = dict()
R_lambda = (blackbody_lambda(band.core, comp2_temp) /
blackbody_lambda(band.core, comp1_temp)).value
def random_in_range(min, max):
return (max-min)*np.random.rand(1)[0] + min
def lnprior(theta):
area, f = theta
# f_S = area * R_lambda
#net_color = (1 - f_S) * target_color + f_S * comp2_color
# net_color = (1 - area) * comp1_color + area * comp2_color
W_Q = (1 - area)/( area*R_lambda + (1 - area) )
W_S = (area * R_lambda)/( area*R_lambda + (1 - area) )
net_color = 2.5 * np.log10(W_Q * 10**(comp1_color/2.5) + W_S * 10**(comp2_color/2.5))
# print(net_color, target_color)
if 0 <= area <= 1 and 0 < f: # and -1 < dlam < 1:
return -0.5 * (net_color - target_color)**2/color_error**2
return -np.inf
def lnlike(theta, target, source1, source2):
area, f = theta
model, residuals = model_known_lambda(target, source1, source2,
mixture_coefficient, area,
# [i - dlam for i in source1_dlambdas],
# [i - dlam for i in source2_dlambdas],
source1_dlambdas, source2_dlambdas,
band, inds, R_lambda, width=fit_width,
uncertainty=f)
return residuals
def lnprob(theta, target, source1, source2):
lp = lnprior(theta)
if not np.isfinite(lp):
return -np.inf
# print(lp, lnlike(theta, target, source1, source2))
return lp + lnlike(theta, target, source1, source2)
ndim, nwalkers = 2, 6
pos = []
counter = -1
while len(pos) < nwalkers:
realization = [random_in_range(0, 1), random_in_range(0.01, 1)]# ,
#random_in_range(-0.01, 0.01)]
if np.isfinite(lnprior(realization)):
pos.append(realization)
sampler = EnsembleSampler(nwalkers, ndim, lnprob, threads=8,
args=(target_slices, source1_slices,
source2_slices))
p0 = sampler.run_mcmc(pos, 1000)[0]
sampler.reset()
sampler.run_mcmc(p0, 1000)
sampler.pool.close()
samples = sampler.chain[:, 500:, :].reshape((-1, ndim))
#samples[:, 0] *= R_lambda
lower, m, upper = np.percentile(samples[:, 0], [16, 50, 84])
band_results['f_S_lower'] = m - lower
band_results['f_S'] = m
band_results['f_S_upper'] = upper - m
band_results['yerr'] = np.median(samples[:, 1])
corner(samples, labels=['$f_S$', '$f$'])#, '$\Delta \lambda$'])
plt.savefig('plots_control/{0}_{1}_{2}.pdf'.format(star, int(band.core.value),
time.replace(':', '_')),
bbox_inches='tight')
plt.close()
fig, ax = plot_posterior_samples_for_paper(samples, target_slices, source1_slices,
source2_slices, mixture_coefficient,
source1_dlambdas, source2_dlambdas,
band, inds, fit_width, star, R_lambda)
# fig, ax = plot_posterior_samples(samples, target_slices, source1_slices,
# source2_slices, mixture_coefficient,
# source1_dlambdas, source2_dlambdas,
# band, inds, fit_width, star)
plt.savefig('plots_control/{0}_{1}_{2}_fit.pdf'.format(star, int(band.core.value),
time.replace(':', '_')),
bbox_inches='tight')
plt.close()
time_results[int(band.core.value)] = band_results
star_results[time] = time_results
results[star] = star_results
dump(results, open(path, 'w'), indent=4, sort_keys=True)
# Old sigmaDraconis
#