PyMultiNest/multinest_marginals.py at master · jbrinchmann/PyMultiNest · GitHub

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
#!/usr/bin/env python
from __future__ import absolute_import, unicode_literals, print_function
__doc__ = """
Script that does default visualizations (marginal plots, 1-d and 2-d).

Author: Johannes Buchner (C) 2013
"""
import numpy
from numpy import exp, log
import matplotlib.pyplot as plt
import sys, os
import json
import pymultinest

if len(sys.argv) != 2:
	sys.stderr.write("""SYNOPSIS: %s <output-root>

	output-root: 	Where the output of a MultiNest run has been written to.
	            	Example: chains/1-
%s""" % (sys.argv[0], __doc__))
	sys.exit(1)

prefix = sys.argv[1]
print('model "%s"' % prefix)
parameters = json.load(open(prefix + 'params.json'))
n_params = len(parameters)

a = pymultinest.Analyzer(n_params = n_params, outputfiles_basename = prefix)
s = a.get_stats()

json.dump(s, open(prefix + 'stats.json', 'w'), indent=4)

print('  marginal likelihood:')
print('    ln Z = %.1f +- %.1f' % (s['global evidence'], s['global evidence error']))
print('  parameters:')
for p, m in zip(parameters, s['marginals']):
	lo, hi = m['1sigma']
	med = m['median']
	sigma = (hi - lo) / 2
	i = max(0, int(-numpy.floor(numpy.log10(sigma))) + 1)
	fmt = '%%.%df' % i
	fmts = '\t'.join(['    %-15s' + fmt + " +- " + fmt])
	print(fmts % (p, med, sigma))

print('creating marginal plot ...')
p = pymultinest.PlotMarginal(a)

values = a.get_equal_weighted_posterior()
assert n_params == len(s['marginals'])
modes = s['modes']

dim2 = os.environ.get('D', '1' if n_params > 20 else '2') == '2'
nbins = 100 if n_params < 3 else 20
if dim2:
	plt.figure(figsize=(5*n_params, 5*n_params))
	for i in range(n_params):
		plt.subplot(n_params, n_params, i + 1)
		plt.xlabel(parameters[i])

		m = s['marginals'][i]
		plt.xlim(m['5sigma'])

		oldax = plt.gca()
		x,w,patches = oldax.hist(values[:,i], bins=nbins, edgecolor='grey', color='grey', histtype='stepfilled', alpha=0.2)
		oldax.set_ylim(0, x.max())

		newax = plt.gcf().add_axes(oldax.get_position(), sharex=oldax, frameon=False)
		p.plot_marginal(i, ls='-', color='blue', linewidth=3)
		newax.set_ylim(0, 1)

		ylim = newax.get_ylim()
		y = ylim[0] + 0.05*(ylim[1] - ylim[0])
		center = m['median']
		low1, high1 = m['1sigma']
		#print(center, low1, high1)
		newax.errorbar(x=center, y=y,
			xerr=numpy.transpose([[center - low1, high1 - center]]),
			color='blue', linewidth=2, marker='s')
		oldax.set_yticks([])
		#newax.set_yticks([])
		newax.set_ylabel("Probability")
		ylim = oldax.get_ylim()
		newax.set_xlim(m['5sigma'])
		oldax.set_xlim(m['5sigma'])
		#plt.close()

		for j in range(i):
			plt.subplot(n_params, n_params, n_params * (j + 1) + i + 1)
			p.plot_conditional(i, j, bins=20, cmap = plt.cm.gray_r)
			for m in modes:
				plt.errorbar(x=m['mean'][i], y=m['mean'][j], xerr=m['sigma'][i], yerr=m['sigma'][j])
			plt.xlabel(parameters[i])
			plt.ylabel(parameters[j])
			#plt.savefig('cond_%s_%s.pdf' % (params[i], params[j]), bbox_tight=True)
			#plt.close()

	plt.savefig(prefix + 'marg.pdf')
	plt.savefig(prefix + 'marg.png')
	plt.close()
else:
	from matplotlib.backends.backend_pdf import PdfPages
	sys.stderr.write('1dimensional only. Set the D environment variable \n')
	sys.stderr.write('to D=2 to force 2d marginal plots.\n')
	pp = PdfPages(prefix + 'marg1d.pdf')

	for i in range(n_params):
		plt.figure(figsize=(3, 3))
		plt.xlabel(parameters[i])
		plt.locator_params(nbins=5)

		m = s['marginals'][i]
		iqr = m['q99%'] - m['q01%']
		xlim = m['q01%'] - 0.3 * iqr, m['q99%'] + 0.3 * iqr
		#xlim = m['5sigma']
		plt.xlim(xlim)

		oldax = plt.gca()
		x,w,patches = oldax.hist(values[:,i], bins=numpy.linspace(xlim[0], xlim[1], 20), edgecolor='grey', color='grey', histtype='stepfilled', alpha=0.2)
		oldax.set_ylim(0, x.max())

		newax = plt.gcf().add_axes(oldax.get_position(), sharex=oldax, frameon=False)
		p.plot_marginal(i, ls='-', color='blue', linewidth=3)
		newax.set_ylim(0, 1)

		ylim = newax.get_ylim()
		y = ylim[0] + 0.05*(ylim[1] - ylim[0])
		center = m['median']
		low1, high1 = m['1sigma']
		#print center, low1, high1
		newax.errorbar(x=center, y=y,
			xerr=numpy.transpose([[center - low1, high1 - center]]),
			color='blue', linewidth=2, marker='s')
		oldax.set_yticks([])
		newax.set_ylabel("Probability")
		ylim = oldax.get_ylim()
		newax.set_xlim(xlim)
		oldax.set_xlim(xlim)
		plt.savefig(pp, format='pdf', bbox_inches='tight')
		plt.close()
	pp.close()