-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathtest.py
executable file
·261 lines (219 loc) · 11.7 KB
/
test.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
import os
import torch
import json
import cv2
from opts import get_opts
from tqdm import tqdm
from utils import utils
cuda = True if torch.cuda.is_available() else False
device = torch.device("cuda:0" if cuda else "cpu")
FloatTensor = torch.cuda.FloatTensor if cuda else torch.FloatTensor
LongTensor = torch.cuda.LongTensor if cuda else torch.LongTensor
args, cfg = get_opts()
# colors = [(0, 0, 255), (0, 255, 255), (0, 255, 0), (255, 0, 0), (255, 0, 255),
# (0, 102, 153), (255, 255, 0), (255, 255, 255), (153, 153, 255), (255, 0, 255),
# (128, 0, 255), (0, 0, 255), (0, 128, 255), (0, 255, 255), (0, 255, 0),
# (128, 255, 0), (255, 255, 0), (255, 128, 0), (255, 0, 0), (255, 0, 128)]
def validate_intent(epoch, model, dataloader, args, recorder, writer):
model.eval()
niters = len(dataloader)
for itern, data in enumerate(dataloader):
intent_logit = model.forward(data, training=False)
intent_prob = torch.sigmoid(intent_logit)
# intent_pred: logit output, bs
# traj_pred: logit, bs x ts x 4
# 1. intent loss
if args.intent_type == 'mean' and args.intent_num == 2: # BCEWithLogitsLoss
gt_intent = data['intention_binary'][:, args.observe_length].type(FloatTensor)
gt_intent_prob = data['intention_prob'][:, args.observe_length].type(FloatTensor)
# gt_disagreement = data['disagree_score'][:, args.observe_length]
# gt_consensus = (1 - gt_disagreement).to(device)
recorder.eval_intent_batch_update(itern, data, gt_intent.detach().cpu().numpy(),
intent_prob.detach().cpu().numpy(), gt_intent_prob.detach().cpu().numpy())
if itern % args.print_freq == 0:
print(f"Epoch {epoch}/{args.epochs} | Batch {itern}/{niters}")
recorder.eval_intent_epoch_calculate(writer)
return recorder
def test_intent(epoch, model, dataloader, args, recorder, writer):
model.eval()
niters = len(dataloader)
recorder.eval_epoch_reset(epoch, niters)
for itern, data in enumerate(dataloader):
intent_logit = model.forward(data, training=False)
intent_prob = torch.sigmoid(intent_logit)
# intent_pred: logit output, bs x 1
# traj_pred: logit, bs x ts x 4
# 1. intent loss
if args.intent_type == 'mean' and args.intent_num == 2: # BCEWithLogitsLoss
gt_intent = data['intention_binary'][:, args.observe_length].type(FloatTensor)
gt_intent_prob = data['intention_prob'][:, args.observe_length].type(FloatTensor)
recorder.eval_intent_batch_update(itern, data, gt_intent.detach().cpu().numpy(),
intent_prob.detach().cpu().numpy(), gt_intent_prob.detach().cpu().numpy())
recorder.eval_intent_epoch_calculate(writer)
return recorder
def predict_intent(model, dataloader, args):
model.eval()
dt = {}
for itern, data in enumerate(dataloader):
intent_logit = model.forward(data, training=False)
intent_prob = torch.sigmoid(intent_logit)
for i in range(len(data['frames'])):
vid = data['video_id'][i] # str list, bs x 60
pid = data['ped_id'][i] # str list, bs x 60
fid = (data['frames'][i][-1] + 1).item() # int list, bs x 15, observe 0~14, predict 15th intent
# gt_int = data['intention_binary'][i][args.observe_length].item() # int list, bs x 60
# gt_int_prob = data['intention_prob'][i][args.observe_length].item() # float list, bs x 60
# gt_disgr = data['disagree_score'][i][args.observe_length].item() # float list, bs x 60
int_prob = intent_prob[i].item()
int_pred = round(int_prob) # <0.5 --> 0, >=0.5 --> 1.
if vid not in dt:
dt[vid] = {}
if pid not in dt[vid]:
dt[vid][pid] = {}
if fid not in dt[vid][pid]:
dt[vid][pid][fid] = {}
dt[vid][pid][fid]['intent_pred'] = int_pred
dt[vid][pid][fid]['intent_pred_prob'] = int_prob
with open(os.path.join(args.checkpoint_path, 'results', 'test_intent_prediction.json'), 'w') as f:
json.dump(dt, f)
def validate_traj(model, dataloader, args, recorder, writer):
total_val_loss = 0
model.eval()
niters = len(dataloader)
for itern, data in enumerate(tqdm(dataloader, desc='Validation')):
with torch.no_grad():
result_dict = model(data, training=False)
traj_pred = result_dict['traj_pred']
if args.absolute_bbox_input: # traj_gt is relative to the first frame
traj_gt = data['bboxes'][:,args.observe_length:,:].type(FloatTensor) - data['bboxes'][:,:1,:].type(FloatTensor)
else:
traj_gt = data['bboxes'][:,args.observe_length:,:].type(FloatTensor)
loss_dict = model.get_loss(data['targets'].to(device))
traj_loss = loss_dict['traj_loss']
total_val_loss += traj_loss * args.batch_size
min_bbox = torch.tensor(args.min_bbox).type(FloatTensor).to(device)
max_bbox = torch.tensor(args.max_bbox).type(FloatTensor).to(device)
traj_pred = utils.convert_unnormalize_bboxes(
bboxes=traj_pred,
normalize=args.normalize_bbox,
# bbox_type='ltrb' if args.bbox_type == 'cxcywh' else None,
bbox_type2cvt='ltrb' if args.bbox_type == 'cxcywh' else None,
min_bbox=min_bbox,
max_bbox=max_bbox,
)
traj_gt = utils.convert_unnormalize_bboxes(
bboxes=traj_gt,
normalize=args.normalize_bbox,
# bbox_type='ltrb' if args.bbox_type == 'cxcywh' else None,
bbox_type2cvt='ltrb' if args.bbox_type == 'cxcywh' else None,
min_bbox=min_bbox,
max_bbox=max_bbox,
)
recorder.eval_traj_batch_update(itern, data, traj_gt.detach().cpu().numpy(), traj_pred.detach().cpu().numpy())
val_loss = total_val_loss / len(dataloader)
score = recorder.eval_traj_epoch_calculate(writer)
return recorder, score, val_loss
def predict_traj(model, dataloader, args, dset='test'):
model.eval()
dt = {}
for j, data in enumerate(dataloader):
result_dict = model(data, training=False)
traj_pred = result_dict['traj_pred']
if args.absolute_bbox_input: # traj_gt is relative to the first frame
traj_gt = data['bboxes'][:,args.observe_length:,:].type(FloatTensor) - data['bboxes'][:,:1,:].type(FloatTensor)
else:
traj_gt = data['bboxes'][:,args.observe_length:,:].type(FloatTensor)
min_bbox = torch.tensor(args.min_bbox).type(FloatTensor).to(device)
max_bbox = torch.tensor(args.max_bbox).type(FloatTensor).to(device)
traj_pred = utils.convert_unnormalize_bboxes(
bboxes=traj_pred,
normalize=args.normalize_bbox,
bbox_type2cvt='ltrb' if args.bbox_type == 'cxcywh' else None,
min_bbox=min_bbox,
max_bbox=max_bbox,
)
traj_gt = utils.convert_unnormalize_bboxes(
bboxes=traj_gt,
normalize=args.normalize_bbox,
bbox_type2cvt='ltrb' if args.bbox_type == 'cxcywh' else None,
min_bbox=min_bbox,
max_bbox=max_bbox,
)
if args.visualize:
os.makedirs(f"./psi_dataset/frames_dot/{data['video_id'][0]}", exist_ok=True)
traj_pred_abs = traj_pred + data['original_bboxes'][:,:1,:].type(FloatTensor) if args.absolute_bbox_input else traj_pred
traj_gt_abs = traj_gt + data['original_bboxes'][:,:1,:].type(FloatTensor) if args.absolute_bbox_input else traj_gt
traj_pred_vis = utils.convert_bbox(traj_pred_abs, bbox_type='cxcywh')
traj_gt_vis = utils.convert_bbox(traj_gt_abs, bbox_type='cxcywh')
for i, (single_traj_pred, single_traj_gt) in enumerate(zip(traj_pred_vis, traj_gt_vis)):
single_traj_pred = single_traj_pred.detach().cpu().numpy()
single_traj_gt = single_traj_gt.detach().cpu().numpy()
video_id = data["video_id"][i]
frame_id = int(data["frames"][i][14]) + 1
image = cv2.imread(f"./psi_dataset/frames/{video_id}/{frame_id:03d}.jpg")
for point_pred, point_gt in zip(single_traj_pred, single_traj_gt):
cv2.circle(image, (int(point_pred[0]), int(point_pred[1])), 1, (0,255,0), -1)
cv2.circle(image, (int(point_gt[0]), int(point_gt[1])), 1, (0,0,255), -1)
# draw gt bbox
single_traj_gt_ltrb = utils.convert_bbox(single_traj_gt, bbox_type='ltrb')
first_bbox = single_traj_gt_ltrb[0]
cv2.rectangle(image, (int(first_bbox[0]), int(first_bbox[1])), (int(first_bbox[2]), int(first_bbox[3])), (255,0,0), 1)
cv2.imwrite(f"./psi_dataset/frames_dot/{video_id}/{frame_id:03d}.jpg", image)
for i in range(len(data['frames'])): # for each sample in a batch
vid = data['video_id'][i] # str list, bs x 60
pid = data['ped_id'][i] # str list, bs x 60
fid = (data['frames'][i][-1] + 1).item() # int list, bs x 15, observe 0~14, predict 15th intent
if vid not in dt:
dt[vid] = {}
if pid not in dt[vid]:
dt[vid][pid] = {}
if fid not in dt[vid][pid]:
dt[vid][pid][fid] = {}
dt[vid][pid][fid]['traj'] = traj_pred[i].detach().cpu().numpy().tolist()
# print(len(traj_pred[i].detach().cpu().numpy().tolist()))
# print("saving prediction...")
with open(os.path.join(args.checkpoint_path, 'results', f'{dset}_traj_pred.json'), 'w') as f:
json.dump(dt, f)
def get_test_traj_gt(model, dataloader, args, dset='test'):
model.eval()
gt = {}
for itern, data in enumerate(dataloader):
output = model(data, training=False)
traj_pred = output['traj_pred']
if args.absolute_bbox_input:
traj_gt = data['bboxes'][:,args.observe_length:,:].type(FloatTensor) - data['bboxes'][:,:1,:].type(FloatTensor)
else:
traj_gt = data['bboxes'][:,args.observe_length:,:].type(FloatTensor)
min_bbox = torch.tensor(args.min_bbox).type(FloatTensor).to(device)
max_bbox = torch.tensor(args.max_bbox).type(FloatTensor).to(device)
traj_pred = utils.convert_unnormalize_bboxes(
bboxes=traj_pred,
normalize=args.normalize_bbox,
# bbox_type='ltrb' if args.bbox_type == 'cxcywh' else None,
bbox_type2cvt='ltrb' if args.bbox_type == 'cxcywh' else None,
min_bbox=min_bbox,
max_bbox=max_bbox,
)
traj_gt = utils.convert_unnormalize_bboxes(
bboxes=traj_gt,
normalize=args.normalize_bbox,
# bbox_type='ltrb' if args.bbox_type == 'cxcywh' else None,
bbox_type2cvt='ltrb' if args.bbox_type == 'cxcywh' else None,
min_bbox=min_bbox,
max_bbox=max_bbox,
)
for i in range(len(data['frames'])): # for each sample in a batch
vid = data['video_id'][i] # str list, bs x 60
pid = data['ped_id'][i] # str list, bs x 60
fid = (data['frames'][i][-1] + 1).item() # int list, bs x 15, observe 0~14, predict 15th intent
if vid not in gt:
gt[vid] = {}
if pid not in gt[vid]:
gt[vid][pid] = {}
if fid not in gt[vid][pid]:
gt[vid][pid][fid] = {}
gt[vid][pid][fid]['traj'] = traj_gt[i].detach().cpu().numpy().tolist()
# print(len(traj_pred[i].detach().cpu().numpy().tolist()))
os.makedirs(os.path.join(f'./test_gt'), exist_ok=True)
with open(os.path.join(f'./test_gt/{dset}_traj_gt.json'), 'w') as f:
json.dump(gt, f)