-
Notifications
You must be signed in to change notification settings - Fork 13
/
utils.py
143 lines (123 loc) · 3.76 KB
/
utils.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
import numpy as np
CLASSES = [
'background',
'aeroplane',
'bicycle',
'bird',
'boat',
'bottle',
'bus',
'car',
'cat',
'chair',
'cow',
'diningtable',
'dog',
'horse',
'motorbike',
'person',
'potted plant',
'sheep',
'sofa',
'train',
'tv/monitor',
]
def _fast_hist(label_true, label_pred, n_class):
mask = (label_true >= 0) & (label_true < n_class)
hist = np.bincount(n_class * label_true[mask].astype(int) +
label_pred[mask],
minlength=n_class**2).reshape(n_class, n_class)
return hist
def label_accuracy_score(label_trues, label_preds, n_class):
"""Returns accuracy score evaluation result.
- overall accuracy
- mean accuracy
- mean IU
- fwavacc
"""
hist = np.zeros((n_class, n_class))
for lt, lp in zip(label_trues, label_preds):
hist += _fast_hist(lt.flatten(), lp.flatten(), n_class)
acc = np.diag(hist).sum() / hist.sum()
acc_cls = np.diag(hist) / hist.sum(axis=1)
acc_cls = np.nanmean(acc_cls)
iu = np.diag(hist) / (hist.sum(axis=1) + hist.sum(axis=0) - np.diag(hist))
mean_iu = np.nanmean(iu)
freq = hist.sum(axis=1) / hist.sum()
fwavacc = (freq[freq > 0] * iu[freq > 0]).sum()
return acc, acc_cls, mean_iu, fwavacc
def get_log_dir(model_name, config_id, cfg):
# load config
# import datetime
# import pytz
import os
import yaml
import os.path as osp
name = 'MODEL-%s' % (model_name)
# now = datetime.datetime.now(pytz.timezone('America/Bogota'))
# name += '_TIME-%s' % now.strftime('%Y%m%d-%H%M%S')
# create out
log_dir = osp.join('logs', name)
if not osp.exists(log_dir):
os.makedirs(log_dir)
with open(osp.join(log_dir, 'config.yaml'), 'w') as f:
yaml.safe_dump(cfg, f, default_flow_style=False)
return log_dir
def get_config():
return {
# same configuration as original work
# https://github.com/shelhamer/fcn.berkeleyvision.org
1:
dict(
max_iteration=100000,
lr=1.0e-10,
momentum=0.99,
weight_decay=0.0005,
interval_validate=4000,
)
}
def get_cuda(cuda, _id):
import torch
if not cuda:
return torch.device('cpu')
else:
return torch.device('cuda:{}'.format(_id))
def imshow_label(label_show, alpha=1.0):
import matplotlib
import matplotlib.pyplot as plt
import numpy as np
cmap = plt.cm.jet
# extract all colors from the .jet map
cmaplist = [cmap(i) for i in range(cmap.N)]
cmaplist[0] = (0.0, 0.0, 0.0, 1.0)
cmap = cmap.from_list('Custom cmap', cmaplist, cmap.N)
# define the bins and normalize
bounds = np.arange(0, len(CLASSES))
norm = matplotlib.colors.BoundaryNorm(bounds, cmap.N)
plt.imshow(label_show, cmap=cmap, norm=norm, alpha=alpha)
cbar = plt.colorbar(ticks=bounds)
cbar.ax.set_yticklabels(CLASSES)
def fileimg2model(img_file, transform):
import PIL
img = PIL.Image.open(img_file).convert('RGB')
img = np.array(img, dtype=np.uint8)
return transform(img, img)[0]
def run_fromfile(model, img_file, cuda, transform, val=False):
import matplotlib.pyplot as plt
import torch
if not val:
img_torch = torch.unsqueeze(fileimg2model(img_file, transform), 0)
else:
img_torch = img_file
img_torch = img_torch.to(cuda)
model.eval()
with torch.no_grad():
if not val:
img_org = plt.imread(img_file)
else:
img_org = transform(img_file[0], img_file[0])[0]
score = model(img_torch)
lbl_pred = score.data.max(1)[1].cpu().numpy()
plt.imshow(img_org, alpha=.9)
imshow_label(lbl_pred[0], alpha=0.5)
plt.show()