-
Notifications
You must be signed in to change notification settings - Fork 0
/
big_sim_plot.py
146 lines (108 loc) · 5.87 KB
/
big_sim_plot.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
import matplotlib.pyplot as plt
import matplotlib.lines as mlines
from astropy.cosmology import FlatLambdaCDM
import pickle
import numpy as np
import corner
mcmc_point_estimates = {'AV':[], 'mu':[], 'theta':[]}
vi_point_estimates = {'AV':[], 'mu':[], 'theta':[]}
mcmc_uncertainties = {'AV':[], 'mu':[], 'theta':[]}
vi_uncertainties = {'AV':[], 'mu':[], 'theta':[]}
fiducial_cosmology={"H0": 73.24, "Om0": 0.28}
cosmo = FlatLambdaCDM(**fiducial_cosmology)
num_to_plot = 200
dataset_number=23
av_stat = 'median'
title_str = ""
if dataset_number == 6:
title_str = "epsilons off for sims, on for fitting"
if dataset_number == 5 or dataset_number == 7:
title_str = "no epsilons for sim/fitting"
if dataset_number == 4:
title_str = "using epsilons for both sims and fitting"
true_avs = np.loadtxt("sim_population_AV_" + str(dataset_number) + ".txt")[:num_to_plot]
true_thetas = np.loadtxt("sim_population_theta_" + str(dataset_number) + ".txt")[:num_to_plot]
true_z = np.loadtxt("sim_population_z_" + str(dataset_number) + ".txt")[:num_to_plot]
true_mus = np.array([cosmo.distmod(z).value for z in true_z])
def get_mode_from_samples(samples):
hist, bin_edges = np.histogram(samples, bins=50)
max_index = np.argmax(hist)
mode = (bin_edges[max_index] + bin_edges[max_index + 1])/2
return mode
for i in range(num_to_plot):
dataset = 'sim_population_' + str(dataset_number) + '/' + str(i)
with (open("results/" + dataset + "_vi/chains.pkl", "rb")) as openfile:
vi_objects = pickle.load(openfile)
with (open("results/" + dataset + "_mcmc/chains.pkl", "rb")) as openfile:
mcmc_objects = pickle.load(openfile)
vi_params = np.load("results/" + dataset + "_vi/vi_params.npz")
for var in ['AV', 'mu', 'theta']:
mcmc_samples = mcmc_objects[var].reshape((1000,))
vi_samples = np.squeeze(vi_objects[var])
if var == "AV":
if av_stat == 'median':
vi_point_estimates[var].append(np.median(vi_samples))
mcmc_point_estimates[var].append(np.median(mcmc_samples))
elif av_stat == 'mode':
mcmc_point_estimates[var].append(get_mode_from_samples(mcmc_samples))
vi_point_estimates[var].append(get_mode_from_samples(vi_samples))
else: # Use sample median for mu point estimate
vi_point_estimates[var].append(np.median(vi_samples))
mcmc_point_estimates[var].append(np.median(mcmc_samples))
vi_uncertainties[var].append(np.std(vi_samples))
mcmc_uncertainties[var].append(np.std(mcmc_samples))
for var in ['AV', 'mu', 'theta']:
mcmc_point_estimates[var] = np.array(mcmc_point_estimates[var])
vi_point_estimates[var] = np.array(vi_point_estimates[var])
vi_uncertainties[var] = np.array(vi_uncertainties[var])
mcmc_uncertainties[var] = np.array(mcmc_uncertainties[var])
# VI vs true subplots
alpha = 0.2
fig, ax = plt.subplots(3,3, figsize = (9,9))
mcmc_color = 'r'
vi_color = 'b'
axis_fontsize = 12
true_values = {'mu': true_mus, 'theta': true_thetas, 'AV': true_avs}
latex_version = {'mu': '$\\mu$', 'theta': '$\\theta$', 'AV': '$A_V$'}
for i, var in enumerate(['mu', 'theta', 'AV']):
ax[0][i].plot(true_values[var], vi_point_estimates[var], '.', alpha=alpha, c=vi_color, label='VI')
ax[0][i].plot(true_values[var], mcmc_point_estimates[var], '.', c=mcmc_color, alpha=alpha, label='MCMC')
ax[0][i].errorbar(true_values[var], vi_point_estimates[var], yerr = vi_uncertainties[var], alpha=alpha, c=vi_color, linestyle='None')
ax[0][i].errorbar(true_values[var], mcmc_point_estimates[var], yerr = mcmc_uncertainties[var], c=mcmc_color, alpha=alpha, linestyle='None')
linspace_vals = np.linspace(min(true_values[var]), max(true_values[var]))
ax[0][i].plot(linspace_vals, linspace_vals, c='k')
ax[0][i].set_ylabel('Fit '+ latex_version[var], fontsize = axis_fontsize)
ax[0][i].legend()
ax[1][i].plot(true_values[var], vi_point_estimates[var] - true_values[var], '.', alpha=alpha, c = vi_color, label='VI')
ax[1][i].plot(true_values[var], mcmc_point_estimates[var] - true_values[var], '.', c=mcmc_color, alpha=alpha, label='MCMC')
ax[1][i].errorbar(true_values[var], vi_point_estimates[var] - true_values[var], yerr = vi_uncertainties[var], alpha=alpha, c=vi_color, linestyle='None')
ax[1][i].errorbar(true_values[var], mcmc_point_estimates[var] - true_values[var], yerr = mcmc_uncertainties[var], c=mcmc_color, alpha=alpha, linestyle='None')
ax[1][i].axhline(0, color = 'k')
ax[1][i].set_ylabel('Residual ' + latex_version[var] + ' (Fit - true)', fontsize = axis_fontsize)
ax[1][i].legend()
ax[2][i].plot(true_values[var], vi_point_estimates[var] - mcmc_point_estimates[var], 'o', color='gray')
# ax[2][i].errorbar(true_values[var], vi_point_estimates[var] - mcmc_point_estimates[var], yerr = np.max([vi_uncertainties[var], mcmc_uncertainties[var]]), c='gray', linestyle='None')
ax[2][i].axhline(0, color = 'k')
ax[2][i].set_ylabel('Residual ' + latex_version[var] + ' (VI - MCMC)', fontsize = axis_fontsize)
ax[2][i].set_xlabel('True ' + latex_version[var], fontsize = axis_fontsize)
ax[0][0].annotate("$\\mu$: median", (36, 34.5), fontsize = 16)
ax[0][1].annotate("$\\theta$: median", (0,-1.5), fontsize = 16)
if av_stat == 'median':
ax[0][2].annotate("$A_V$: median", (0.5,0.2), fontsize = 16)
elif av_stat == 'mode':
ax[0][2].annotate("$A_V$: mode", (0.7,0.2), fontsize = 16)
for axis in ax.flatten():
axis.tick_params(axis='x', labelsize=12)
axis.tick_params(axis='y', labelsize=12)
plt.tight_layout()
plt.show()
fig, ax = plt.subplots(1,3)
for i, var in enumerate(['mu', 'theta', 'AV']):
ax[i].plot(vi_uncertainties[var], mcmc_uncertainties[var], 'o')
min_value = np.min(np.concatenate((vi_uncertainties[var], mcmc_uncertainties[var])))
max_value = np.max(np.concatenate((vi_uncertainties[var], mcmc_uncertainties[var])))
linspace_vals = np.linspace(min_value, max_value)
ax[i].plot(linspace_vals, linspace_vals, 'k')
ax[i].set_ylabel('MCMC ' + latex_version[var] + ' uncertainty', fontsize = axis_fontsize)
ax[i].set_xlabel('VI ' + latex_version[var] + ' uncertainty', fontsize = axis_fontsize)
plt.show()