-
Notifications
You must be signed in to change notification settings - Fork 2
/
Copy pathreport_algconfigs.py
114 lines (110 loc) · 4.82 KB
/
report_algconfigs.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
#!/usr/bin/env python3
"""
@author: meysamhashemi INS Marseille
"""
import os
import sys
import numpy as np
import re
##########################################################################################################
def convergence_opt(filename):
pass_opt=0
if 'Optimization terminated normally:' in open(filename).read():
pass_opt+=1
return pass_opt
##########################################################################################################
def convergence_advi(filename):
pass_advi=0
if 'COMPLETED.' in open(filename).read():
pass_advi+=1
return pass_advi
##########################################################################################################
def convergence_hmc(filename):
pass_hmc=0
if 'Elapsed Time' in open(filename).read():
pass_hmc+=1
return pass_hmc
##########################################################################################################
def hmc_config(fit_out):
with open(fit_out , 'r') as fd:
lines = fd.readlines()
last_line = lines[-2].strip().split(' ')
with open(fit_out , 'r') as fd:
for line in fd.readlines():
reading_line=line.strip().split(' ')
if reading_line[0]=='num_samples':
if '(Default)' in reading_line:
num_samples=reading_line[-2]
else:
num_samples=reading_line[-1]
if reading_line[0]=='num_warmup':
if '(Default)' in reading_line:
num_warmup=reading_line[-2]
else:
num_warmup=reading_line[-1]
if reading_line[0]=='save_warmup':
if '(Default)' in reading_line:
save_warmup=reading_line[-2]
else:
save_warmup=reading_line[-1]
if reading_line[0]=='delta':
if '(Default)' in reading_line:
adapt_delta=reading_line[-2]
else:
adapt_delta=reading_line[-1]
if reading_line[0]=='max_depth':
if '(Default)' in reading_line:
max_depth=reading_line[-2]
else:
max_depth=reading_line[-1]
if reading_line[0]=='Elapsed':
total_time=last_line[0]
return num_samples, num_warmup, save_warmup, adapt_delta, max_depth, total_time
##########################################################################################################
def advi_config(fit_out):
with open(fit_out , 'r') as fd:
for line in fd.readlines():
reading_line=line.strip().split(' ')
if reading_line[0]=='grad_samples':
if '(Default)' in reading_line:
grad_samples=reading_line[-2]
else:
grad_samples=reading_line[-1]
if reading_line[0]=='elbo_samples':
if '(Default)' in reading_line:
elbo_samples=reading_line[-2]
else:
elbo_samples=reading_line[-1]
if reading_line[0]=='output_samples':
if '(Default)' in reading_line:
output_samples=reading_line[-2]
else:
output_samples=reading_line[-1]
if reading_line[0]=='tol_rel_obj':
if '(Default)' in reading_line:
tol_rel_obj=reading_line[-2]
else:
tol_rel_obj=reading_line[-1]
return grad_samples, output_samples, output_samples, tol_rel_obj
##########################################################################################################
def advi_elbo(fit_out):
with open(fit_out , 'r') as fd:
counter=0
lines=[]
for line in fd.readlines():
reading_line=line.strip().split(' ')
if reading_line[0]=='Success!':
start_iter=counter
if reading_line[0]=='COMPLETED.':
end_iter=counter
counter+=1
lines.append(line.strip().split(','))
Convergence=[]
for row in np.arange(start_iter+4,(end_iter-5)+1):
line=lines[row][0]
Convergence.append([x for x in re.split(',| ',line) if x!=''][0:2])
Convergence=np.asarray(Convergence)
Iter=Convergence[:,0].astype('float')
Elbo=Convergence[:,1].astype('float')
return Iter, Elbo
##########################################################################################################