-
Notifications
You must be signed in to change notification settings - Fork 2
/
evaluation.py
340 lines (313 loc) · 12.9 KB
/
evaluation.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
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
import os
import pandas as pd
import numpy as np
from PIL import Image
import argparse
import torch
import torch.nn.functional as F
import joblib
from pathlib import Path
categories = ['background', 'aeroplane', 'bicycle', 'bird', 'boat', 'bottle', 'bus', 'car', 'cat', 'chair', 'cow',
'diningtable', 'dog', 'horse', 'motorbike', 'person', 'pottedplant', 'sheep', 'sofa', 'train',
'tvmonitor']
categories_coco = ['background',
'person',
'bicycle',
'car',
'motorcycle',
'airplane',
'bus',
'train',
'truck',
'boat',
'traffic light',
'fire hydrant',
'street sign',
'stop sign',
'parking meter',
'bench',
'bird',
'cat',
'dog',
'horse',
'sheep',
'cow',
'elephant',
'bear',
'zebra',
'giraffe',
'hat',
'backpack',
'umbrella',
'shoe',
'eye glasses',
'handbag',
'tie',
'suitcase',
'frisbee',
'skis',
'snowboard',
'sports ball',
'kite',
'baseball bat',
'baseball glove',
'skateboard',
'surfboard',
'tennis racket',
'bottle',
'plate',
'wine glass',
'cup',
'fork',
'knife',
'spoon',
'bowl',
'banana',
'apple',
'sandwich',
'orange',
'broccoli',
'carrot',
'hot dog',
'pizza',
'donut',
'cake',
'chair',
'couch',
'potted plant',
'bed',
'mirror',
'dining table',
'window',
'desk',
'toilet',
'door',
'tv',
'laptop',
'mouse',
'remote',
'keyboard',
'cell phone',
'microwave',
'oven',
'toaster',
'sink',
'refrigerator',
'blender',
'book',
'clock',
'vase',
'scissors',
'teddy bear',
'hair drier',
'toothbrush']
def do_python_eval(predict_folder, gt_folder, name_list, num_cls=21, input_type='png', threshold=1.0, printlog=False,
out_crf=False, n_jobs=10, img_dir=None, cam_type="cam", out_dir=None):
def compare(idx):
name = name_list[idx]
if input_type == 'png':
predict_file = os.path.join(predict_folder, '%s.png' % name)
predict = np.array(Image.open(predict_file)) # cv2.imread(predict_file)
if num_cls == 81:
predict = predict - 91
elif input_type == 'npy':
predict_file = os.path.join(predict_folder, '%s.npy' % name)
predict_dict = np.load(predict_file, allow_pickle=True).item()
cam = np.array([predict_dict[key] for key in predict_dict.keys()])
label_key = np.array([key+1 for key in predict_dict.keys()]).astype(np.uint8)
cam = np.pad(cam, ((1, 0), (0, 0), (0, 0)), mode='constant', constant_values=threshold)
label_key = np.pad(label_key, (1, 0), mode='constant', constant_values=0)
cam = F.softmax(torch.tensor(cam).float(), dim=0).numpy()
predict = np.argmax(cam, axis=0).astype(np.uint8)
if out_crf:
orig_image = np.array(Image.open(os.path.join(img_dir, name + '.jpg')).convert("RGB"))
# crf use predict label
if "cam" in cam_type:
from tool.imutils import crf_inference_label
predict = crf_inference_label(orig_image, predict, n_labels=cam.shape[0])
elif "coco" in cam_type:
from tool.imutils import crf_inference_label_coco
predict = crf_inference_label_coco(orig_image, predict, n_labels=cam.shape[0])
predict = label_key[predict]
if out_dir is not None:
predict_img = Image.fromarray(predict)
predict_img.save(os.path.join(out_dir, name + ".png"))
gt_file = os.path.join(gt_folder, '%s.png' % name)
gt = np.array(Image.open(gt_file))
cal = gt < 255
mask = (predict == gt) * cal
p_list, t_list, tp_list = [0] * num_cls, [0] * num_cls, [0] * num_cls
for i in range(num_cls):
p_list[i] += np.sum((predict == i) * cal)
t_list[i] += np.sum((gt == i) * cal)
tp_list[i] += np.sum((gt == i) * mask)
return p_list, t_list, tp_list
results = joblib.Parallel(n_jobs=n_jobs, verbose=10, pre_dispatch="all")(
[joblib.delayed(compare)(j) for j in range(len(name_list))]
)
p_lists, t_lists, tp_lists = zip(*results)
TP = [0] * num_cls
P = [0] * num_cls
T = [0] * num_cls
for idx in range(len(name_list)):
p_list = p_lists[idx]
t_list = t_lists[idx]
tp_list = tp_lists[idx]
for i in range(num_cls):
TP[i] += tp_list[i]
P[i] += p_list[i]
T[i] += t_list[i]
IoU = []
T_TP = []
P_TP = []
FP_ALL = []
FN_ALL = []
Prediction = []
Recall = []
for i in range(num_cls):
IoU.append(TP[i] / (T[i] + P[i] - TP[i]))
T_TP.append(T[i] / (TP[i]))
P_TP.append(P[i] / (TP[i]))
FP_ALL.append((P[i] - TP[i]) / (T[i] + P[i] - TP[i]))
FN_ALL.append((T[i] - TP[i]) / (T[i] + P[i] - TP[i]))
Prediction.append(TP[i] / P[i])
Recall.append(TP[i] / T[i])
loglist = {}
for i in range(num_cls):
if num_cls == 21:
loglist[categories[i]] = IoU[i] * 100
else:
loglist[categories_coco[i]] = IoU[i] * 100
miou = np.nanmean(np.array(IoU))
loglist['mIoU'] = miou * 100
fp = np.nanmean(np.array(FP_ALL))
loglist['FP'] = fp * 100
fn = np.nanmean(np.array(FN_ALL))
loglist['FN'] = fn * 100
prediction = np.nanmean(np.array(Prediction))
loglist['Prediction'] = prediction * 100
recall = np.nanmean(np.array(Recall))
loglist['Recall'] = recall * 100
if printlog:
for i in range(num_cls):
if num_cls == 21:
if i % 2 != 1:
print('%11s:%7.3f%%' % (categories[i], IoU[i] * 100), end='\t')
else:
print('%11s:%7.3f%%' % (categories[i], IoU[i] * 100))
else:
if i % 2 != 1:
print('%11s:%7.3f%%' % (categories_coco[i], IoU[i] * 100), end='\t')
else:
print('%11s:%7.3f%%' % (categories_coco[i], IoU[i] * 100))
print('\n======================================================')
print('%11s:%7.3f%%' % ('mIoU', miou * 100))
print('\n')
print(f'FP = {fp * 100}, FN = {fn * 100}')
print(f'Prediction = {prediction * 100}, Recall = {recall * 100}')
return loglist
def writedict(file, dictionary):
s = ''
for key in dictionary.keys():
sub = '%s:%s ' % (key, dictionary[key])
s += sub
s += '\n'
file.write(s)
def writelog(filepath, metric, comment):
filepath = filepath
logfile = open(filepath, 'a')
import time
logfile.write(time.strftime("%Y-%m-%d %H:%M:%S", time.localtime()))
logfile.write('\t%s\n' % comment)
writedict(logfile, metric)
logfile.write('=====================================\n')
logfile.close()
def whole_eval(list, predict_dir, gt_dir, logfile, comment='comment', type='npy',
t=None, curve=True, num_classes=21, start=0, end=60):
if type == 'npy':
assert t is not None or curve
# converters 处理coco数据集前缀零被省略的问题
df = pd.read_csv(list, names=['filename'], converters={"filename": str})
name_list = df['filename'].values
if not curve:
loglist = do_python_eval(predict_dir, gt_dir, name_list, num_classes, type, t, printlog=True)
writelog(logfile, loglist, comment)
else:
l = []
max_mIoU = 0.0
best_thr = 0.0
for i in range(start, end):
t = i / 100.0
loglist = do_python_eval(predict_dir, gt_dir, name_list, num_classes, type, t, printlog=True)
l.append(loglist['mIoU'])
print('%d/%d background score: %.3f\tmIoU: %.3f%%' % (i, end, t, loglist['mIoU']))
if loglist['mIoU'] > max_mIoU:
max_mIoU = loglist['mIoU']
best_thr = t
else:
break
print('Best background score: %.3f\tmIoU: %.3f%%' % (best_thr, max_mIoU))
writelog(logfile, {'mIoU': l, 'Best mIoU': max_mIoU, 'Best threshold': best_thr}, comment)
return max_mIoU, best_thr
if __name__ == '__main__':
parser = argparse.ArgumentParser()
parser.add_argument("--list", default='voc12/train_id.txt', type=str)
parser.add_argument("--data-path", default='data/voc12')
parser.add_argument("--predict_dir", default='./out_rw', type=str)
parser.add_argument("--gt_dir", default='SegmentationClass', type=str)
parser.add_argument("--img_dir", type=str, default=None)
parser.add_argument("--out-crf", action="store_true")
parser.add_argument("--out-dir", default=None, type=str)
parser.add_argument('--logfile', default='./evallog.txt', type=str)
# parser.add_argument('--comment', required=True, type=str)
parser.add_argument('--type', default='png', choices=['npy', 'png'], type=str)
parser.add_argument('--t', default=None, type=float)
parser.add_argument('--curve', default=False, type=bool)
parser.add_argument('--num_classes', default=21, type=int)
parser.add_argument('--start', default=0, type=int)
parser.add_argument('--end', default=60, type=int)
parser.add_argument('--n_jobs', default=10, type=int)
parser.add_argument('--cam-type', default="cam", type=str)
args = parser.parse_args()
if "voc12" in args.list:
args.data_path = Path(args.data_path) / "voc12" if "voc12" not in args.data_path else args.data_path
args.data_path = Path(args.data_path) / "VOCdevkit" / "VOC2012"
args.gt_dir = args.data_path / "SegmentationClassAug"
args.img_dir = args.data_path / "JPEGImages"
if "coco" in args.list:
args.data_path = Path(args.data_path) / "coco" if "coco" not in args.data_path else args.data_path
args.gt_dir = args.data_path / "voc_format" / "class_labels"
args.img_dir = args.data_path / "images"
args.cam_type="coco"
if args.type == 'npy':
assert args.t is not None or args.curve
df = pd.read_csv(args.list, names=['filename'], converters={"filename": str})
name_list = df['filename'].values
if args.t is not None:
args.t = args.t / 100.0
if args.out_dir is not None:
Path(args.out_dir).mkdir(parents=True, exist_ok=True)
if not args.curve:
loglist = do_python_eval(args.predict_dir, args.gt_dir, name_list, args.num_classes, args.type, args.t,
printlog=True, cam_type=args.cam_type,
out_crf=args.out_crf, n_jobs=args.n_jobs, img_dir=args.img_dir,
out_dir=args.out_dir)
# writelog(args.logfile, loglist, args.comment)
else:
l = []
max_mIoU = 0.0
best_thr = 0.0
for i in range(args.start, args.end):
t = i / 100.0
loglist = do_python_eval(args.predict_dir, args.gt_dir, name_list, args.num_classes, args.type, t,
printlog=True, cam_type=args.cam_type,
out_crf=args.out_crf, n_jobs=args.n_jobs, img_dir=args.img_dir)
l.append(loglist['mIoU'])
print('%d/%d background score: %.3f\tmIoU: %.3f%%' % (i, args.end, t, loglist['mIoU']))
if loglist['mIoU'] > max_mIoU:
max_mIoU = loglist['mIoU']
best_thr = t
else:
break
print('Best background score: %.3f\tmIoU: %.3f%%' % (best_thr, max_mIoU))
# writelog(args.logfile, {'mIoU': l, 'Best mIoU': max_mIoU, 'Best threshold': best_thr}, args.comment)