forked from IshitaTakeshi/RoadDamageDetector
-
Notifications
You must be signed in to change notification settings - Fork 0
/
showreport.py
86 lines (63 loc) · 1.82 KB
/
showreport.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
import sys
import json
import random
from os.path import basename
import numpy as np
from matplotlib import pyplot as plt
from matplotlib import rcParams
rcParams['figure.figsize'] = 16, 20
rcParams['figure.dpi'] = 240
rcParams['font.size'] = 18
linestyles = ['-', '--', ':', '-.']
def extract(logs, keys):
def _extract(log):
items = []
for key in keys:
try:
item = log[key]
except:
item = None
items.append(item)
return items
L = list(zip(*[_extract(log) for log in logs]))
return np.array(L, dtype=np.double)
if len(sys.argv) < 2:
print("Usage $python3 showreport.py <path to log file>")
exit(0)
logfile_path = sys.argv[1]
with open(logfile_path, "r") as f:
logs = json.load(f)
L = extract(logs, keys=[
"iteration",
"main/loss",
"main/loss/conf",
"main/loss/loc"
])
labels = ["confidence loss", "location loss", "overall loss"]
iteration, loss = L[0], L[1:]
ax1 = plt.subplot(211)
for loss_, label in zip(loss, labels):
plt.plot(iteration, loss_, label=label)
plt.legend(prop={'size': 16})
keys = [
'validation/main/ap/D00',
'validation/main/ap/D01',
'validation/main/ap/D10',
'validation/main/ap/D11',
'validation/main/ap/D20',
'validation/main/ap/D40',
'validation/main/ap/D43',
'validation/main/ap/D44',
'validation/main/map'
]
L = extract(logs, keys=["iteration"] + keys)
labels = [basename(key) for key in keys]
iteration, aps = L[0], L[1:]
plt.subplot(212, sharex=ax1)
plt.ylim([0, 1])
for ap, label in zip(aps, labels):
masks = np.logical_not(np.isnan(ap))
plt.plot(iteration[masks], ap[masks], linestyle=random.choice(linestyles), label=label)
plt.xlabel("iteration")
plt.legend(loc='center left', bbox_to_anchor=(1, 0.815), numpoints=1)
plt.show()