-
Notifications
You must be signed in to change notification settings - Fork 1
/
train_multi_pred.py
621 lines (540 loc) · 30 KB
/
train_multi_pred.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
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
"""Multilane and multi prediction experiments.
Exp 1: 1 predictions with one centerline.
Exp 2: 2 predictions with one centerline. (Expectation Maximization should be used here.)
Exp 3: Use image with predictions in centerline frame with one prediction per centerline. (train with prediction error)
Exp 4: Use image with predictions in centerline frame with multiple prediction per centerline. (train with expectation maximization)
(this should work the best).
Design: dataset 1 will give one centerline and one trajectory for one train frame. while test it will provide candidate centerlines.
Dataset 2 will give one centerline with one trajectory and one image. while test multiple candidate trajectory with images.
Model Exp1: train with lstm with the centerline and social features. Loss is prediction loss with lstm.
Exp 2: train the model with expectation maximization.
Exp 3: train with prediction error
Exp 4: train with expectation maximization
"""
from argoverse.data_loading.argoverse_forecasting_loader import ArgoverseForecastingLoader
from argoverse.map_representation.map_api import ArgoverseMap
from data_new import collate_traj_multilane,collate_traj_xy,collate_traj_social_centerline,Argoverse_MultiLane_Data,Argoverse_Social_Data,Argoverse_Social_Centerline_Data
from model_new import LSTMModel,LSTMModel_CenterlineEmbed,Social_Model,Social_Model_Centerline,LSTMModel_XY,Social_Model_Centerline_Refined,Social_Model_Refined,ConstantVelocity_XY
from torch.utils.data import Dataset, DataLoader
from argoverse.evaluation.eval_forecasting import get_ade, get_fde
from argoverse.evaluation.competition_util import generate_forecasting_h5
import glob,warnings
from argoverse.visualization.visualize_sequences import viz_sequence
from visualize import viz_predictions
import torch
import pickle
import pandas as pd
import os
import numpy as np
import pdb
import torch.nn as nn
import argparse
from time import localtime, strftime
class Train():
def __init__(self,model,optimizer,train_loader,val_loader,test_loader,loss_fn,model_dir,pretrained_dir=None):
self.model=model
if pretrained_dir!=None:
self.model.load_state_dict(torch.load(pretrained_dir+'best-model.pt')['model_state_dict'])
self.optimizer=optimizer
self.train_loader=train_loader
self.val_loader=val_loader
self.test_loader=test_loader
self.model_dir=model_dir
self.loss_fn=loss_fn
self.best_1_ade = np.inf
self.best_1_fde = np.inf
self.best_3_ade = np.inf
self.best_3_fde = np.inf
def train_epoch(self):
total_loss=0
self.model.train()
num_batches=len(self.train_loader.batch_sampler)
batch_size=self.train_loader.batch_size
eliminated=0
num_samples=0
for i_batch,traj_dict in enumerate(self.train_loader):
pred_traj=self.model(traj_dict,mode='train')
# pdb.set_trace()
loss=self.loss_fn(pred_traj,traj_dict['gt_traj'].cuda())
num_samples+=pred_traj.shape[0]
total_loss=total_loss+(loss.data*pred_traj.shape[0])
avg_loss = float(total_loss)/(num_samples)
eliminated+=batch_size-pred_traj.shape[0]
print(f"Training Iter {i_batch+1}/{num_batches} Avg Loss {avg_loss:.4f} Batch Loss {loss.data:.4f} Eliminated : {eliminated}/{num_samples}",end="\r")
self.optimizer.zero_grad()
loss.backward()
self.optimizer.step()
print()
def val_epoch(self,epoch):
total_loss=0
num_batches=len(self.val_loader.batch_sampler)
self.model.eval()
ade_one_sec,fde_one_sec,ade_three_sec,fde_three_sec=(0,0,0,0)
ade_one_sec_avg, fde_one_sec_avg ,ade_three_sec_avg, fde_three_sec_avg = (0,0,0,0)
no_samples=0
for i_batch,traj_dict in enumerate(self.val_loader):
pred_traj=self.model(traj_dict,mode='validate')
gt_traj=traj_dict['gt_unnorm_traj'].cuda()
loss=self.loss_fn(pred_traj,gt_traj)
batch_samples=pred_traj.shape[0]
no_samples+=batch_samples
total_loss=total_loss+(loss.data*batch_samples)
avg_loss = float(total_loss)/(no_samples)
# pdb.set_trace()
ade_one_sec+=sum([get_ade(pred_traj[i,:10,:],gt_traj[i,:10,:]) for i in range(batch_samples)])
fde_one_sec+=sum([get_fde(pred_traj[i,:10,:],gt_traj[i,:10,:]) for i in range(batch_samples)])
ade_three_sec+=sum([get_ade(pred_traj[i,:,:],gt_traj[i,:,:]) for i in range(batch_samples)])
fde_three_sec+=sum([get_fde(pred_traj[i,:,:],gt_traj[i,:,:]) for i in range(batch_samples)])
ade_one_sec_avg = float(ade_one_sec)/no_samples
ade_three_sec_avg = float(ade_three_sec)/no_samples
fde_one_sec_avg = float(fde_one_sec)/no_samples
fde_three_sec_avg = float(fde_three_sec)/no_samples
print(f"Validation Iter {i_batch+1}/{num_batches} Avg Loss {avg_loss:.4f} Batch Loss {loss.data:.4f} \
One sec:- ADE:{ade_one_sec/(no_samples):.4f} FDE: {fde_one_sec/(no_samples):.4f}\
Three sec:- ADE:{ade_three_sec/(no_samples):.4f} FDE: {fde_three_sec/(no_samples):.4f}",end="\r")
# print(f"Validation Iter {i_batch+1}/{num_batches} Avg Loss {avg_loss:.4f} \
# One sec:- ADE:{ade_one_sec/(no_samples):.4f} FDE: {fde_one_sec/(no_samples):.4f}\
# Three sec:- ADE:{ade_three_sec/(no_samples):.4f} FDE: {fde_three_sec/(no_samples):.4f}",end="\r")
_filename = self.model_dir + 'best-model.pt'
if ade_three_sec_avg < self.best_3_ade and fde_three_sec_avg < self.best_3_fde:
torch.save({
'epoch': epoch,
'model_state_dict': model.state_dict(),
'opt_state_dict': optimizer.state_dict(),
'loss': total_loss/(i_batch+1)
}, _filename)
self.best_1_ade = ade_one_sec_avg
self.best_1_fde = fde_one_sec_avg
self.best_3_ade = ade_three_sec_avg
self.best_3_fde = fde_three_sec_avg
self.best_model_updated=True
print("\nModel updated")
else:
print()
def run(self,num_epochs):
for i in range(num_epochs):
self.train_epoch()
self.val_epoch(epoch=i)
# self.test_epoch()
class Validate():
def __init__(self,model,val_loader,multi_val_loader,loss_fn,model_dir):
self.model=model
self.val_loader=val_loader
self.multi_val_loader=multi_val_loader
self.loss_fn=loss_fn
self.model_dir=model_dir
def val_epoch(self):
total_loss=0
num_batches=len(self.val_loader.batch_sampler)
# self.model.load_state_dict(torch.load(self.model_dir+'best-model.pt')['model_state_dict'])
self.model.eval()
ade_one_sec,fde_one_sec,ade_three_sec,fde_three_sec=(0,0,0,0)
ade_one_sec_avg, fde_one_sec_avg ,ade_three_sec_avg, fde_three_sec_avg = (0,0,0,0)
no_samples=0
for i_batch,traj_dict in enumerate(self.val_loader):
pred_traj=self.model(traj_dict,mode='validate')
gt_traj=traj_dict['gt_unnorm_traj']
loss=self.loss_fn(traj_dict['gt_unnorm_traj'].cuda(),pred_traj)
# pdb.set_trace()
total_loss=total_loss+loss.data
avg_loss = float(total_loss)/(i_batch+1)
batch_samples=gt_traj.shape[0]
no_samples+=batch_samples
ade_one_sec+=sum([get_ade(pred_traj[i,:10,:],gt_traj[i,:10,:]) for i in range(batch_samples)])
fde_one_sec+=sum([get_fde(pred_traj[i,:10,:],gt_traj[i,:10,:]) for i in range(batch_samples)])
ade_three_sec+=sum([get_ade(pred_traj[i,:,:],gt_traj[i,:,:]) for i in range(batch_samples)])
fde_three_sec+=sum([get_fde(pred_traj[i,:,:],gt_traj[i,:,:]) for i in range(batch_samples)])
ade_one_sec_avg = float(ade_one_sec)/no_samples
ade_three_sec_avg = float(ade_three_sec)/no_samples
fde_one_sec_avg = float(fde_one_sec)/no_samples
fde_three_sec_avg = float(fde_three_sec)/no_samples
print(f"Validation Iter {i_batch+1}/{num_batches} Avg Loss {total_loss/(i_batch+1):.4f} \
One sec:- ADE:{ade_one_sec/(no_samples):.4f} FDE: {fde_one_sec/(no_samples):.4f}\
Three sec:- ADE:{ade_three_sec/(no_samples):.4f} FDE: {fde_three_sec/(no_samples):.4f}",end="\r")
print()
def save_results_single_pred(self):
print("running save results")
afl=ArgoverseForecastingLoader("data/val/data/")
torch.load(self.model_dir+'best-model.pt', map_location=lambda storage, loc: storage)
# self.model.load_state_dict(torch.load(self.model_dir+'best-model.pt')['model_state_dict'])
self.model.load_state_dict(torch.load(self.model_dir+'best-model.pt', map_location=lambda storage, loc: storage)['model_state_dict'])
self.model.eval()
save_results_path=self.model_dir+"/results/"
# pdb.set_trace()
if not os.path.exists(save_results_path):
os.mkdir(save_results_path)
num_batches=len(self.val_loader.batch_sampler)
for i_batch,traj_dict in enumerate(self.val_loader):
print(f"Running {i_batch}/{num_batches}",end="\r")
gt_traj=traj_dict['gt_unnorm_traj'].numpy()
output=self.model(traj_dict,mode='validate')
# output=output.detach().cpu().numpy()
output=output.detach().numpy()
seq_paths=traj_dict['seq_path']
# input_tensor=[]
for index,seq_path in enumerate(seq_paths):
loader=afl.get(seq_path)
input_array=loader.agent_traj[0:20,:]
city=loader.city
del loader
seq_index=int(os.path.basename(seq_path).split('.')[0])
output_dict={'seq_path':seq_path,'seq_index':seq_index,'input':input_array,
'output':output[index],'target':gt_traj[index],'city':city}
with open(f"{save_results_path}/{seq_index}.pkl", 'wb') as f:
pickle.dump(output_dict,f)
# input_tensor=np.array(input_tensor)
def save_top_accuracy(self):
print("running save accuracy")
self.model.load_state_dict(torch.load(self.model_dir+'best-model.pt')['model_state_dict'])
self.model.eval()
min_loss=np.inf
max_loss=0
num_images=10
loss_list_max=[]
input_max_list=[]
pred_max_list=[]
target_max_list=[]
city_name_max=[]
seq_path_list_max=[]
loss_list_min=[]
input_min_list=[]
pred_min_list=[]
target_min_list=[]
city_name_min=[]
seq_path_list_min=[]
num_batches=len(self.multi_val_loader.batch_sampler)
for i_batch,traj_dict in enumerate(self.multi_val_loader):
print(f"Running {i_batch}/{num_batches}",end="\r")
gt_traj=traj_dict['gt_unnorm_traj'].numpy()
pred_traj=self.model(traj_dict,mode='validate_multiple')
loss=[]
# import pdb;pdb.set_trace()
for index in range(len(pred_traj)):
loss_temp=[]
for j in range(pred_traj[index].shape[0]):
loss_temp.append(np.linalg.norm(pred_traj[index][j]- gt_traj[index]))
# import pdb;pdb.set_trace()
loss.append(min(loss_temp))
# import pdb;pdb.set_trace()
loss=torch.Tensor(loss).float()
min_loss,min_index=torch.min(loss,dim=0)
max_loss,max_index=torch.max(loss,dim=0)
input_min_list.append(traj_dict['train_unnorm_traj'][min_index])
pred_min_list.append(pred_traj[min_index])
target_min_list.append(traj_dict['gt_unnorm_traj'][min_index])
input_max_list.append(traj_dict['train_unnorm_traj'][max_index])
pred_max_list.append(pred_traj[max_index])
target_max_list.append(traj_dict['gt_unnorm_traj'][max_index])
city_name_min.append(traj_dict['city'][min_index])
city_name_max.append(traj_dict['city'][max_index])
seq_path_list_max.append(traj_dict['seq_path'][max_index])
seq_path_list_min.append(traj_dict['seq_path'][min_index])
loss_list_max.append(min_loss.data)
loss_list_min.append(max_loss.data)
loss_list_max_array=np.array(loss_list_max)
loss_list_max=list(loss_list_max_array.argsort()[-num_images:][::-1])
loss_list_min_array=np.array(loss_list_min)
loss_list_min=list(loss_list_min_array.argsort()[:num_images])
avm=ArgoverseMap()
high_error_path=self.model_dir+"/visualization/high_errors/"
low_error_path=self.model_dir+"/visualization/low_errors/"
if not os.path.exists(high_error_path):
os.makedirs(high_error_path)
if not os.path.exists(low_error_path):
os.makedirs(low_error_path)
input_max=[]
pred_max=[]
target_max=[]
city_max=[]
centerlines_max=[]
for i,index in enumerate(loss_list_max):
print(f"Max: {i}")
input_max.append(input_max_list[index].numpy())
pred_max.append(pred_max_list[index])
target_max.append(target_max_list[index].numpy())
city_max.append(city_name_max[index])
viz_sequence(df=pd.read_csv(seq_path_list_max[index]) ,save_path=f"{high_error_path}/dataframe_{i}.png",show=True,avm=avm)
centerlines_max.append(avm.get_candidate_centerlines_for_traj(input_max[-1], city_max[-1],viz=False))
print("Created max array")
input_min=[]
pred_min=[]
target_min=[]
city_min=[]
centerlines_min=[]
for i,index in enumerate(loss_list_min):
print(f"Min: {i}")
input_min.append(input_min_list[index].numpy())
pred_min.append(pred_min_list[index])
target_min.append(target_min_list[index].numpy())
city_min.append(city_name_min[index])
viz_sequence(df=pd.read_csv(seq_path_list_min[index]) ,save_path=f"{low_error_path}/dataframe_{i}.png",show=True,avm=avm)
centerlines_min.append(avm.get_candidate_centerlines_for_traj(input_min[-1], city_min[-1],viz=False))
import pdb;pdb.set_trace()
print("Created min array")
print(f"Saving max visualizations at {high_error_path}")
viz_predictions(input_=np.array(input_max), output=pred_max,target=np.array(target_max),centerlines=centerlines_max,city_names=np.array(city_max),avm=avm,save_path=high_error_path)
print(f"Saving min visualizations at {low_error_path}")
viz_predictions(input_=np.array(input_min), output=pred_min,target=np.array(target_min),centerlines=centerlines_min,city_names=np.array(city_min),avm=avm,save_path=low_error_path)
def save_top_errors_accuracy_single_pred(self):
afl=ArgoverseForecastingLoader("data/val/data/")
self.model.load_state_dict(torch.load(self.model_dir+'best-model.pt')['model_state_dict'])
self.model.eval()
min_loss=np.inf
max_loss=0
num_images=10
loss_list_max=[]
input_max_list=[]
pred_max_list=[]
target_max_list=[]
city_name_max=[]
seq_path_list_max=[]
loss_list_min=[]
input_min_list=[]
pred_min_list=[]
target_min_list=[]
city_name_min=[]
seq_path_list_min=[]
num_batches=len(self.val_loader.batch_sampler)
for i_batch,traj_dict in enumerate(self.val_loader):
print(f"Running {i_batch}/{num_batches}",end="\r")
# pdb.set_trace()
# pred_traj=self.model(traj_dict)
# pred_traj=self.val_loader.dataset.inverse_transform(pred_traj,traj_dict)
gt_traj=traj_dict['gt_unnorm_traj']
output=self.model(traj_dict,mode='validate')
output=output.cpu()
loss=torch.norm(output.reshape(output.shape[0],-1)-gt_traj.reshape(gt_traj.shape[0],-1),dim=1)
min_loss,min_index=torch.min(loss,dim=0)
max_loss,max_index=torch.max(loss,dim=0)
print(f"Min loss: {min_loss}, Max loss: {max_loss}" )
seq_path_list_max.append(traj_dict['seq_path'][max_index])
seq_path_list_min.append(traj_dict['seq_path'][min_index])
loader_max=afl.get(traj_dict['seq_path'][max_index])
input_max_list.append(loader_max.agent_traj[0:20,:])
city_name_max.append(loader_max.city)
del loader_max
loader_min=afl.get(traj_dict['seq_path'][min_index])
input_min_list.append(loader_min.agent_traj[0:20,:])
city_name_min.append(loader_min.city)
del loader_min
pred_min_list.append(output[min_index])
target_min_list.append(gt_traj[min_index])
pred_max_list.append(output[max_index])
target_max_list.append(gt_traj[max_index])
# loss_list_max.append(min_loss.data)
# loss_list_min.append(max_loss.data)
loss_list_max.append(max_loss.data)
loss_list_min.append(min_loss.data)
# torch.cuda.empty_cache()
# pdb.set_trace()
loss_list_max_array=np.array(loss_list_max)
loss_list_max=list(loss_list_max_array.argsort()[-num_images:][::-1])
loss_list_min_array=np.array(loss_list_min)
loss_list_min=list(loss_list_min_array.argsort()[:num_images])
# pdb.set_trace()
avm=ArgoverseMap()
high_error_path=model_dir+"/visualization/high_errors/"
low_error_path=model_dir+"/visualization/low_errors/"
if not os.path.exists(high_error_path):
os.makedirs(high_error_path)
if not os.path.exists(low_error_path):
os.makedirs(low_error_path)
# if self.use_cuda:
# input_=input_.cpu()
# output=output.cpu()
input_max=[]
pred_max=[]
target_max=[]
city_max=[]
# import pdb;pdb.set_trace()
# seq_path_max=[]
centerlines_max=[]
for i,index in enumerate(loss_list_max):
print(f"Max: {i}")
# pdb.set_trace()
input_max.append(input_max_list[index])
pred_max.append([pred_max_list[index].detach().numpy()])
print(f"Difference in predicted and input_traj for maximum at {i} is {np.linalg.norm(input_max[-1]-pred_max[-1][0][0:20,:])}")
target_max.append(target_max_list[index].detach().numpy())
city_max.append(city_name_max[index])
viz_sequence(df=pd.read_csv(seq_path_list_max[index]) ,save_path=f"{high_error_path}/dataframe_{i}.png",show=True,avm=avm)
centerlines_max.append(avm.get_candidate_centerlines_for_traj(input_max[-1], city_max[-1],viz=False))
print("Created max array")
input_min=[]
pred_min=[]
target_min=[]
city_min=[]
# seq_path_min=[]
centerlines_min=[]
for i,index in enumerate(loss_list_min):
# pdb.set_trace()
print(f"Min: {i}")
input_min.append(input_min_list[index])
pred_min.append([pred_min_list[index].detach().numpy()])
print(f"Difference in predicted and input_traj for minimum at {i} is {np.linalg.norm(input_min[-1]-pred_min[-1][0][0:20,:])}")
target_min.append(target_min_list[index].detach().numpy())
city_min.append(city_name_min[index])
# seq_path_min.append(seq_path_list_min[index])
viz_sequence(df=pd.read_csv(seq_path_list_min[index]) ,save_path=f"{low_error_path}/dataframe_{i}.png",show=True,avm=avm)
centerlines_min.append(avm.get_candidate_centerlines_for_traj(input_min[-1], city_min[-1],viz=False))
print("Created min array")
print(f"Saving max visualizations at {high_error_path}")
# import pdb;pdb.set_trace()
viz_predictions(input_=np.array(input_max), output=pred_max,target=np.array(target_max),centerlines=centerlines_max,city_names=np.array(city_max),avm=avm,save_path=high_error_path)
print(f"Saving min visualizations at {low_error_path}")
# viz_predictions(input_=pred_min[0], output=pred_min,target=np.array(target_min),centerlines=centerlines_min,city_names=np.array(city_min),avm=avm,save_path=low_error_path)
# viz_predictions(input_=np.expand_dims(input_min[0],axis=0),output=pred_min[0],target=np.expand_dims(target_min[0],axis=0),centerlines=[centerlines_min], city_names=np.expand_dims(city_min[0],axis=0),avm=avm,save_path=low_error_path)
viz_predictions(input_=np.array(input_min), output=pred_min,target=np.array(target_min),centerlines=centerlines_min,city_names=np.array(city_min),avm=avm,save_path=low_error_path)
def run(self):
print("in validation run")
# self.val_epoch()
# self.save_top_errors_accuracy_single_pred()
self.save_results_single_pred()
class Test():
def __init__(self,model,test_loader,model_dir):
self.test_model=model
self.model_dir=model_dir
self.test_loader=test_loader
def save_trajectory(self,output_dict,save_path):
generate_forecasting_h5(output_dict, save_path)
print("done")
def test(self):
num_batches=len(self.test_loader.batch_sampler)
batch_size=self.test_loader.batch_size
self.test_model.load_state_dict(torch.load(model_dir+'best-model.pt')['model_state_dict'])
self.test_model.eval()
no_samples=0
output_all = {}
for i_batch,traj_dict in enumerate(self.test_loader):
seq_paths=traj_dict['seq_path']
seq_index=[int(os.path.basename(seq_path).split('.')[0]) for seq_path in seq_paths]
pred_traj=self.test_model(traj_dict,mode='test')
for index in range(len(pred_traj)):
pred_traj_index=pred_traj[index]
if pred_traj_index.shape[0]>6:
pred_traj_index=pred_traj_index[:6]
else:
while pred_traj_index.shape[0]<6:
pred_traj_index=np.vstack((pred_traj_index,pred_traj_index[0]))
output_all.update({seq_index[index]:pred_traj_index})
print(f"Test Iter {i_batch+1}/{num_batches}",end="\r")
print()
print("Saving the test data results in dir",self.model_dir)
self.save_trajectory(output_all,self.model_dir)
if __name__ == "__main__":
warnings.filterwarnings('ignore')
parser = argparse.ArgumentParser(description='Sequence Modeling - Argoverse Forecasting Task')
parser.add_argument('--batch_size', type=int, default=128, metavar='N',
help='batch size (default: 128)')
parser.add_argument('--lr', type=float, default=0.001,
help='learning rate for optimizer (default: 0.001)')
parser.add_argument('--epochs', type=int, default=10,
help='upper epoch limit (default: 10)')
parser.add_argument('--model', type=str, default='LSTM',
help='model type to execute (LSTM, Social.default: LSTM)')
parser.add_argument('--data', type=str, default='MultiLane',
help='type of data to use for training (default: XY, options: XY,LaneCentre,')
parser.add_argument('--mode',type=str,default='train',help='mode: train, test ,validate')
parser.add_argument('--model_dir',type=str,default=None,help='model path for test or validate')
parser.add_argument('--pretrained_dir',type=str,default=None,help='model path to use as pretrained model')
args = parser.parse_args()
curr_time = strftime("%Y%m%d%H%M%S", localtime())
# curr_time="20191129181134" #i guess lstm with centerline embed
# curr_time="20191201222432" #i guess social"
# curr_time="20191203132341" #social centerline"
if args.mode == 'train':
model_dir = './models/' + args.model + '/' + curr_time + '/'
if not os.path.exists(model_dir):
os.makedirs(model_dir)
else:
model_dir=args.model_dir
# model=LSTMModel()
if args.model=="LSTM_XY":
model=LSTMModel_XY()
if args.model=="ConstantVelocity_XY":
model=ConstantVelocity_XY()
elif args.model=="LSTM":
model=LSTMModel_CenterlineEmbed()
elif args.model=="Social":
model=Social_Model()
elif args.model=="Social_Model_Refined":
model=Social_Model_Refined()
elif args.model=="Social_Centerline":
model=Social_Model_Centerline()
elif args.model=="Social_Model_Centerline_Refined":
model=Social_Model_Centerline_Refined()
# model=LSTMModel_CenterlineEmbed()
loss_fn=nn.MSELoss()
optimizer = torch.optim.Adam(model.parameters(), lr = args.lr)
print("Mode is ",args.mode)
print("Data is", args.data)
print("Model dir is",model_dir)
print(f"Training for {args.epochs} epochs")
if args.data=="MultiLane":
argoverse_map=ArgoverseMap()
if args.mode=="train" or args.mode=="validate":
argoverse_val=Argoverse_MultiLane_Data('data/val/data/',avm=argoverse_map,train_seq_size=20,mode="validate",load_saved=True)
val_loader = DataLoader(argoverse_val, batch_size=16,
shuffle=True, num_workers=4,collate_fn=collate_traj_multilane)
if args.mode=="train":
argoverse_train=Argoverse_MultiLane_Data('data/train/data/',avm=argoverse_map,train_seq_size=20,mode="train",load_saved=True)
train_loader = DataLoader(argoverse_train, batch_size=args.batch_size,
shuffle=True, num_workers=8,collate_fn=collate_traj_multilane)
if args.mode=="validate":
argoverse_val_multiple=Argoverse_MultiLane_Data('data/val/data/',avm=argoverse_map,train_seq_size=20,mode="validate_multiple")
val_multi_loader=DataLoader(argoverse_val_multiple, batch_size=args.batch_size,
shuffle=False, num_workers=4,collate_fn=collate_traj_multilane)
if args.mode=="test" or args.mode=="train" or args.mode=="validate":
argoverse_test = Argoverse_MultiLane_Data('data/test_obs/data',avm=argoverse_map,train_seq_size=20,mode="test")
test_loader = DataLoader(argoverse_test, batch_size=args.batch_size,
shuffle=False, num_workers=16,collate_fn=collate_traj_multilane)
elif args.data=="Social" or args.data=="XY":
argoverse_map=ArgoverseMap()
if args.mode=="train" or args.mode=="validate":
argoverse_val=Argoverse_Social_Data('data/val/data/',avm=argoverse_map,train_seq_size=20,mode="validate",load_saved=False)
val_loader = DataLoader(argoverse_val, batch_size=args.batch_size,
shuffle=True, num_workers=1,collate_fn=collate_traj_xy)
if args.mode=="train":
argoverse_train=Argoverse_Social_Data('data/train/data/',avm=argoverse_map,train_seq_size=20,mode="train",load_saved=True)
train_loader = DataLoader(argoverse_train, batch_size=args.batch_size,
shuffle=True, num_workers=16,collate_fn=collate_traj_xy)
if args.mode=="validate":
argoverse_val_multiple=Argoverse_Social_Data('data/val/data/',avm=argoverse_map,train_seq_size=20,mode="validate_multiple")
val_multi_loader=DataLoader(argoverse_val_multiple, batch_size=args.batch_size,
shuffle=False, num_workers=16,collate_fn=collate_traj_xy)
if args.mode=="test" or args.mode=="train" or args.mode=="validate":
argoverse_test = Argoverse_Social_Data('data/test_obs/data',avm=argoverse_map,train_seq_size=20,mode="test")
test_loader = DataLoader(argoverse_test, batch_size=args.batch_size,
shuffle=False, num_workers=16,collate_fn=collate_traj_xy)
elif args.data=="Social_Centerline":
argoverse_map=ArgoverseMap()
if args.mode=="train" or args.mode=="validate":
argoverse_val=Argoverse_Social_Centerline_Data('data/val/data/',avm=argoverse_map,train_seq_size=20,mode="validate",load_saved=True)
val_loader = DataLoader(argoverse_val, batch_size=args.batch_size,
shuffle=True, num_workers=8,collate_fn=collate_traj_social_centerline)
if args.mode=="train":
argoverse_train=Argoverse_Social_Centerline_Data('data/train/data/',avm=argoverse_map,train_seq_size=20,mode="train",load_saved=True)
train_loader = DataLoader(argoverse_train, batch_size=args.batch_size,
shuffle=True, num_workers=8,collate_fn=collate_traj_social_centerline)
if args.mode=="validate":
argoverse_val_multiple=Argoverse_Social_Centerline_Data('data/val/data/',avm=argoverse_map,train_seq_size=20,mode="validate_multiple")
val_multi_loader=DataLoader(argoverse_val_multiple, batch_size=args.batch_size,
shuffle=False, num_workers=8,collate_fn=collate_traj_social_centerline)
if args.mode=="test" or args.mode=="train" or args.mode=="validate":
argoverse_test = Argoverse_Social_Centerline_Data('data/test_obs/data',avm=argoverse_map,train_seq_size=20,mode="test")
test_loader = DataLoader(argoverse_test, batch_size=args.batch_size,
shuffle=False, num_workers=8,collate_fn=collate_traj_social_centerline)
else:
print(f"No dataset: {args.data}. What are you doing")
# pdb.set_trace()
if args.mode=="train":
model=model.cuda()
trainer=Train(model=model,optimizer=optimizer,train_loader=train_loader,val_loader=val_loader,test_loader=test_loader,loss_fn=loss_fn,model_dir=model_dir,pretrained_dir=args.pretrained_dir)
trainer.run(args.epochs)
elif args.mode=="validate":
print("In validate")
# model=model.cuda()
validater=Validate(model=model,val_loader=val_loader,multi_val_loader=val_multi_loader,loss_fn=loss_fn,model_dir=model_dir)
validater.run()
elif args.mode=="test":
pass