-
Notifications
You must be signed in to change notification settings - Fork 0
/
train.py
214 lines (162 loc) · 7.51 KB
/
train.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
import argparse
import os
import shutil
import numpy as np
import torch
import torch.nn as nn
import torch.optim as optim
import torchvision.transforms as T
from torch.utils.data import DataLoader
from models.PSMnet import PSMNet
from models.smoothloss import SmoothL1Loss
from dataloader.KITTI2015_loader import KITTI2015, RandomCrop, ToTensor, Normalize, Pad
import tensorboardX as tX
import matplotlib
matplotlib.use('agg')
import matplotlib.pyplot as plt
parser = argparse.ArgumentParser(description='PSMNet')
parser.add_argument('--maxdisp', type=int, default=192, help='max diparity')
parser.add_argument('--logdir', default='log/runs', help='log directory')
parser.add_argument('--datadir', default='data', help='data directory')
parser.add_argument('--cuda', type=int, default=0, help='gpu number')
parser.add_argument('--batch_size', type=int, default=1, help='batch size')
parser.add_argument('--validate-batch-size', type=int, default=1, help='batch size')
parser.add_argument('--log-per-step', type=int, default=1, help='log per step')
parser.add_argument('--save-per-epoch', type=int, default=1, help='save model per epoch')
parser.add_argument('--model-dir', default='checkpoint', help='directory where save model checkpoint')
parser.add_argument('--model-path', default=None, help='path of model to load')
# parser.add_argument('--start-step', type=int, default=0, help='number of steps at starting')
parser.add_argument('--lr', type=float, default=0.001, help='learning rate')
parser.add_argument('--num-epochs', type=int, default=300, help='number of training epochs')
parser.add_argument('--num-workers', type=int, default=1, help='num workers in loading data')
# parser.add_argument('--')
args = parser.parse_args()
# imagenet
mean = [0.406, 0.456, 0.485]
std = [0.225, 0.224, 0.229]
device_ids = [0]
writer = tX.SummaryWriter(log_dir=args.logdir, comment='FSMNet')
device = torch.device('cuda')
print(device)
def main(args):
train_transform = T.Compose([RandomCrop([256, 512]), Normalize(mean, std), ToTensor()])
train_dataset = KITTI2015(args.datadir, mode='train', transform=train_transform)
train_loader = DataLoader(train_dataset, batch_size=args.batch_size, shuffle=True, num_workers=args.num_workers)
validate_transform = T.Compose([Normalize(mean, std), ToTensor(), Pad(375, 1242)])
validate_dataset = KITTI2015(args.datadir, mode='validate', transform=validate_transform)
validate_loader = DataLoader(validate_dataset, batch_size=args.validate_batch_size, num_workers=args.num_workers)
step = 0
best_error = 100.0
model = PSMNet(args.maxdisp).to(device) #__init__(self, max_disp)
#model = nn.DataParallel(model, device_ids=device_ids)
criterion = SmoothL1Loss().to(device)
optimizer = optim.Adam(model.parameters(), lr=args.lr)
if args.model_path is not None:
state = torch.load(args.model_path)
model.load_state_dict(state['state_dict'])
optimizer.load_state_dict(state['optimizer'])
step = state['step']
best_error = state['error']
print('load model from {}'.format(args.model_path))
print('Number of model parameters: {}'.format(sum([p.data.nelement() for p in model.parameters()])))
for epoch in range(1, args.num_epochs + 1):
model.train()
step = train(model, train_loader, optimizer, criterion, step)
adjust_lr(optimizer, epoch)
if epoch % args.save_per_epoch == 0:
model.eval()
torch.cuda.empty_cache()
error = validate(model, validate_loader, epoch)
best_error = save(model, optimizer, epoch, step, error, best_error)
def train(model, train_loader, optimizer, criterion, step):
'''
train one epoch
'''
for batch in train_loader:
step += 1
optimizer.zero_grad()
left_img = batch['left'].to(device)
right_img = batch['right'].to(device)
target_disp = batch['disp'].to(device)
mask = (target_disp > 0)
mask = mask.detach_()
disp1, disp2, disp3 = model(left_img, right_img)
loss1, loss2, loss3 = criterion(disp1[mask], disp2[mask], disp3[mask], target_disp[mask])
total_loss = 0.5 * loss1 + 0.7 * loss2 + 1.0 * loss3
total_loss.backward()
optimizer.step()
# print(step)
if step % args.log_per_step == 0:
writer.add_scalar('loss/loss1', loss1, step)
writer.add_scalar('loss/loss2', loss2, step)
writer.add_scalar('loss/loss3', loss3, step)
writer.add_scalar('loss/total_loss', total_loss, step)
print('step: {:05} | total loss: {:.5} | loss1: {:.5} | loss2: {:.5} | loss3: {:.5}'.format(step, total_loss.item(), loss1.item(), loss2.item(), loss3.item()))
return step
def validate(model, validate_loader, epoch):
'''
validate 40 image pairs
'''
num_batches = len(validate_loader)
idx = np.random.randint(num_batches)
avg_error = 0.0
for i, batch in enumerate(validate_loader):
left_img = batch['left'].to(device)
right_img = batch['right'].to(device)
target_disp = batch['disp'].to(device)
mask = (target_disp > 0)
mask = mask.detach_()
with torch.no_grad():
_, _, disp = model(left_img, right_img)
delta = torch.abs(disp[mask] - target_disp[mask])
error_mat = (((delta >= 3.0) + (delta >= 0.05 * (target_disp[mask]))) == 2)
error = torch.sum(error_mat).item() / torch.numel(disp[mask]) * 100
avg_error += error
if i == idx:
left_save = left_img
disp_save = disp
avg_error = avg_error / num_batches
print('epoch: {:03} | 3px-error: {:.5}%'.format(epoch, avg_error))
writer.add_scalar('error/3px', avg_error, epoch)
save_image(left_save[0], disp_save[0], epoch)
return avg_error
def adjust_lr(optimizer, epoch):
if epoch == 200:
lr = 0.0001
for param_group in optimizer.param_groups:
param_group['lr'] = lr
def save(model, optimizer, epoch, step, error, best_error):
path = os.path.join(args.model_dir, '{:03}.pt'.format(epoch))
# torch.save(model.state_dict(), path)
# model.save_state_dict(path)
state = {}
state['state_dict'] = model.state_dict()
state['optimizer'] = optimizer.state_dict()
state['error'] = error
state['epoch'] = epoch
state['step'] = step
torch.save(state, path)
print('save model at epoch{}'.format(epoch))
if error < best_error:
best_error = error
best_path = os.path.join(args.model_dir, 'best_model.pt'.format(epoch))
shutil.copyfile(path, best_path)#clone 'path' named 'best_path'
print('best model in epoch {}'.format(epoch))
return best_error
def save_image(left_image, disp, epoch):
for i in range(3):
left_image[i] = left_image[i] * std[i] + mean[i]
b, r = left_image[0], left_image[2]
left_image[0] = r # BGR --> RGB
left_image[2] = b
# left_image = torch.from_numpy(left_image.cpu().numpy()[::-1])
disp_img = disp.detach().cpu().numpy()
fig = plt.figure(12.84, 3.84)
plt.axis('off') # hide axis
plt.imshow(disp_img)
plt.colorbar()
writer.add_figure('image/disp', fig, global_step=epoch)
writer.add_image('image/left', left_image, global_step=epoch)
if __name__ == '__main__':
main(args)
writer.close()