-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy pathpltupper.py
executable file
·206 lines (181 loc) · 7.82 KB
/
pltupper.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
#!/usr/bin/python
import numpy as np
import matplotlib.pyplot as plt
import matplotlib as mpl
import sys
from mpl_toolkits.mplot3d import Axes3D
from matplotlib.ticker import LinearLocator, FormatStrFormatter
import pymultinest
def getargv(argv, key):
for i in range(0,len(argv)):
arg=argv[i]
if (arg==key):
return argv[i+1]
def chkargv(argv, key):
for i in range(0,len(argv)):
arg=argv[i]
if (arg==key):
return True
return False
def plot2dposterier_withconf(dat, likli, indx=[], labels=[], rate=0.5,
par=np.array([]),
parm=np.array([]), rangedat=np.array([]), levels=[0.68]):
'''Plot a 2D posterier with likelihood burning curve
dat: a n x m numpy matrix, each data point is a row
indx: a array of integer indicating wich column (parameter) to plot
labels: the LaTex label of the given parameter
rate: the percentage of data to plot, e.g. 0.3 means plot data from 70% to
the end.
'''
#rc('font',**{'family':'sans-serif','sans-serif':['Helvetica']})
## for Palatino and other serif fonts use:
#rc('font', **{'family': 'serif', 'serif': ['Palatino'], 'size': 8})
#rc('text', usetex=True)
row, col = dat.shape;
if len(indx) == 0:
indx = np.arange(0, col);
if len(labels) == 0:
labels = ['a'];
labels = labels * (col);
dat=dat.copy()
dat = dat[np.ix_(np.arange(int(row - row * rate), row), indx)];
likli = likli[np.ix_(np.arange(int(row - row * rate), row))];
row, col = dat.shape;
#max_yticks = 3
max_yticks = 4
if len(rangedat) == 0:
rangedat = np.zeros((col, 2));
for i in range(0, col):
rangedat[i, 0] = np.min(dat[:, i]);
rangedat[i, 1] = np.max(dat[:, i]);
npar = col;
#index of y
for vari in range(0, npar):
# plot the histogram
#ax = plt.subplot2grid((npar, npar), (vari, vari))
ax=plt.subplot(npar, npar, vari*npar+vari+1)
ind = ((dat[:, vari] < rangedat[vari, 1]) & (dat[:, vari] > rangedat[vari, 0]))
n, bins, patches = plt.hist(dat[ind, vari], 100, normed=True, \
histtype='stepfilled', range=(rangedat[vari, 0], rangedat[vari, 1]))
plt.setp(patches, 'facecolor', 'y', 'alpha', 0.75)
yloc = plt.MaxNLocator(max_yticks)
ax.xaxis.set_major_locator(yloc)
xloc = plt.MaxNLocator(max_yticks)
ax.yaxis.set_major_locator(xloc)
plt.xlabel(labels[vari])
if len(levels)>0:
for ls in [0.68, 0.95]:
histc, bin_edges = np.histogram(dat[:, vari], density=True, bins=100)
vom=(bin_edges[:-1]+bin_edges[1:])*0.5
ind=np.argsort(histc)
ind=ind[::-1]
v=histc.copy()
v[0]=histc[ind[0]]
for i in range(1, len(ind)):
v[i] = v[i-1]+histc[ind[i]]
v=v/float(np.sum(histc))
thre= histc[ind[ (v[:-1] < ls) & (v[1:]>=ls)][0]]
lv=np.min(vom[histc>=thre])
rv=np.max(vom[histc>=thre])
print vari,'-th parameter', 'sigma=', ls, 'lv=', lv, 'rv=', rv, thre
#if vari == 2 and ls == 0.95:
# plt.plot([rv, rv], [0, max(n)], ls='dashed', color='k', linewidth=1)
if vari != 2:
plt.plot([lv, lv], [0, max(n)], ls='dashed', color='k', linewidth=1)
plt.plot([rv, rv], [0, max(n)], ls='dashed', color='k', linewidth=1)
print rv-par[vari], lv-par[vari]
elif ls == 0.95:
print rv
plt.plot([rv, rv], [0, max(n)], ls='solid', color='k', linewidth=2)
if par.size > 0 and vari != 2:
plt.plot([par[vari], par[vari]], [0, max(n)], ls='solid',
color='k', linewidth=2)
if len(parm) > 0:
for i in range(len(parm)):
parm0=parm[i]
plt.plot([parm0[vari], parm0[vari]], [0, max(n)], ls='dashed',
color='k', linewidth=2)
plt.xlim(rangedat[vari, :])
#index of x
for varj in range(vari + 1, npar):
#ax = plt.subplot2grid((npar, npar), (vari, varj))
ax=plt.subplot(npar, npar, (vari)*npar+varj+1)
x = dat[:, varj]
y = dat[:, vari]
ind = (
(x < rangedat[varj, 1]) & (x > rangedat[varj, 0]) & (y < rangedat[vari, 1]) & (y > rangedat[vari, 0]))
liklisub = likli[ind]
x = dat[ind, varj]
y = dat[ind, vari]
ngridx = 20
ngridy = 30
#generate 2D histogram
H, xedges, yedges = np.histogram2d(x, y, bins=(ngridx, ngridy),
range=(rangedat[varj, :], rangedat[vari, :]))
extent = [xedges[0], xedges[-1], yedges[0], yedges[-1]]
#hisogram is row leading need transpose to plot with contourf
H = H.transpose()
#get the center of bin
xedges = (xedges[:-1] + xedges[1:]) / 2
yedges = (yedges[:-1] + yedges[1:]) / 2
mxx, mxy = np.meshgrid(xedges, yedges)
plt.contourf(mxx, mxy, H, 100, cmap=plt.cm.summer_r);
plt.title("%(aa)s-%(bb)s" % {'aa': labels[vari], 'bb': labels[varj]})
#Prepare to count the confidence level
indx, indy = np.meshgrid(np.arange(0, ngridx), np.arange(0, ngridy))
vmx2 = np.squeeze(np.reshape(mxx, (-1, 1)));
vmy2 = np.squeeze(np.reshape(mxy, (-1, 1)));
vm = np.squeeze(np.reshape(H, (-1, 1)));
vx = np.squeeze(np.reshape(indx, (-1, 1)));
vy = np.squeeze(np.reshape(indy, (-1, 1)));
#Sort according to likelihood
vm2 = np.sort(vm)[::-1];
#Get index and convert everything to array
ix = np.argsort(vm, axis=0)[::-1];
ix = np.ix_(ix);
vx2 = vx[ix];
vy2 = vy[ix];
vmx2 = vmx2[ix];
vmy2 = vmy2[ix];
#get the cumulative of the hisogram
vm2 = np.cumsum(vm2 / np.sum(vm2));
cmxx2 = H;
mxx2 = mxx;
mxy2 = mxy;
#Form 2D cumulative plot
for ki in range(0, len(vm2)):
mxx2[vy2[ki], vx2[ki]] = vmx2[ki];
mxy2[vy2[ki], vx2[ki]] = vmy2[ki];
cmxx2[vy2[ki], vx2[ki]] = vm2[ki];
conls = plt.contour(mxx2, mxy2, cmxx2, levels, colors='k');
plt.clabel(conls, inline=1, fontsize=10)
ax.get_yaxis().set_visible(False)
ax.get_xaxis().set_visible(False)
yloc = plt.MaxNLocator(max_yticks)
ax.xaxis.set_major_locator(yloc)
xloc = plt.MaxNLocator(max_yticks)
ax.yaxis.set_major_locator(xloc)
a = pymultinest.Analyzer(n_params = 8, outputfiles_basename=getargv(sys.argv, "-f"))
b = a.get_equal_weighted_posterior()
#allres=a.get_data()
#vlik=-allres[:,1]
vlik=b[:,-1]
allres=b[:,:-1]
mxchain = allres
rdat = np.array([[41, 46], [-3.0, 3.0], [38, 43]])
vpar = mxchain[np.argmax(vlik),:]
#print vlik
print vpar
mpl.rcParams['font.size']=18
mpl.rcParams['xtick.labelsize']=16
mpl.rcParams['ytick.labelsize']=16
mpl.rcParams['axes.labelsize']=20
if chkargv(sys.argv, '-o'):
plt.figure(figsize=(10,7))
plot2dposterier_withconf(mxchain, vlik, par=vpar, indx=[], rangedat=rdat, rate=1.0, levels=[0.68, 0.95], labels=[r'$\log L^*$', r'$\alpha$', r'$\log L_0$'])
plt.suptitle(getargv(sys.argv, "-title"))
plt.savefig(getargv(sys.argv, "-o"))
else:
plot2dposterier_withconf(mxchain, vlik, par=vpar, indx=[], rangedat=rdat, rate=1.0, levels=[0.68, 0.95], labels=[r'$\log L^*$', r'$\alpha$', r'$\log L_0$'])
plt.suptitle(getargv(sys.argv, "-title"))
plt.show()