-
Notifications
You must be signed in to change notification settings - Fork 70
/
train.py
148 lines (124 loc) · 5.21 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
from __future__ import print_function
import os
import random
import argparse
import torch
import torch.nn as nn
import torch.optim as optim
import torch.nn.functional as F
import torch.backends.cudnn as cudnn
import torchvision
import torchvision.transforms as transforms
from PIL import Image
from torch.autograd import Variable
from torchcv.models.fpnssd import FPNSSD512
from torchcv.models.ssd import SSD300, SSD512, SSDBoxCoder
from torchcv.loss import SSDLoss
from torchcv.datasets import ListDataset
from torchcv.transforms import resize, random_flip, random_paste, random_crop, random_distort
parser = argparse.ArgumentParser(description='PyTorch SSD Training')
parser.add_argument('--lr', default=1e-3, type=float, help='learning rate')
parser.add_argument('--resume', '-r', action='store_true', help='resume from checkpoint')
parser.add_argument('--model', default='./examples/ssd/model/ssd512_vgg16.pth', type=str, help='initialized model path')
parser.add_argument('--checkpoint', default='./examples/ssd/checkpoint/ckpt.pth', type=str, help='checkpoint path')
args = parser.parse_args()
# Model
print('==> Building model..')
# net = SSD512(num_classes=21)
net = FPNSSD512(num_classes=21)
net.load_state_dict(torch.load(args.model))
best_loss = float('inf') # best test loss
start_epoch = 0 # start from epoch 0 or last epoch
if args.resume:
print('==> Resuming from checkpoint..')
checkpoint = torch.load(args.checkpoint)
net.load_state_dict(checkpoint['net'])
best_loss = checkpoint['loss']
start_epoch = checkpoint['epoch']
# Dataset
print('==> Preparing dataset..')
box_coder = SSDBoxCoder(net)
img_size = 512
def transform_train(img, boxes, labels):
img = random_distort(img)
if random.random() < 0.5:
img, boxes = random_paste(img, boxes, max_ratio=4, fill=(123,116,103))
img, boxes, labels = random_crop(img, boxes, labels)
img, boxes = resize(img, boxes, size=(img_size,img_size), random_interpolation=True)
img, boxes = random_flip(img, boxes)
img = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.485,0.456,0.406),(0.229,0.224,0.225))
])(img)
boxes, labels = box_coder.encode(boxes, labels)
return img, boxes, labels
trainset = ListDataset(root='/search/odin/liukuang/data/voc_all_images',
list_file=['torchcv/datasets/voc/voc07_trainval.txt',
'torchcv/datasets/voc/voc12_trainval.txt'],
transform=transform_train)
def transform_test(img, boxes, labels):
img, boxes = resize(img, boxes, size=(img_size,img_size))
img = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.485,0.456,0.406),(0.229,0.224,0.225))
])(img)
boxes, labels = box_coder.encode(boxes, labels)
return img, boxes, labels
testset = ListDataset(root='/search/odin/liukuang/data/voc_all_images',
list_file='torchcv/datasets/voc/voc07_test.txt',
transform=transform_test)
trainloader = torch.utils.data.DataLoader(trainset, batch_size=32, shuffle=True, num_workers=8)
testloader = torch.utils.data.DataLoader(testset, batch_size=32, shuffle=False, num_workers=8)
net.cuda()
net = torch.nn.DataParallel(net, device_ids=range(torch.cuda.device_count()))
cudnn.benchmark = True
criterion = SSDLoss(num_classes=21)
optimizer = optim.SGD(net.parameters(), lr=args.lr, momentum=0.9, weight_decay=1e-4)
# Training
def train(epoch):
print('\nEpoch: %d' % epoch)
net.train()
train_loss = 0
for batch_idx, (inputs, loc_targets, cls_targets) in enumerate(trainloader):
inputs = Variable(inputs.cuda())
loc_targets = Variable(loc_targets.cuda())
cls_targets = Variable(cls_targets.cuda())
optimizer.zero_grad()
loc_preds, cls_preds = net(inputs)
loss = criterion(loc_preds, loc_targets, cls_preds, cls_targets)
loss.backward()
optimizer.step()
train_loss += loss.data[0]
print('train_loss: %.3f | avg_loss: %.3f [%d/%d]'
% (loss.data[0], train_loss/(batch_idx+1), batch_idx+1, len(trainloader)))
# Test
def test(epoch):
print('\nTest')
net.eval()
test_loss = 0
for batch_idx, (inputs, loc_targets, cls_targets) in enumerate(testloader):
inputs = Variable(inputs.cuda(), volatile=True)
loc_targets = Variable(loc_targets.cuda())
cls_targets = Variable(cls_targets.cuda())
loc_preds, cls_preds = net(inputs)
loss = criterion(loc_preds, loc_targets, cls_preds, cls_targets)
test_loss += loss.data[0]
print('test_loss: %.3f | avg_loss: %.3f [%d/%d]'
% (loss.data[0], test_loss/(batch_idx+1), batch_idx+1, len(testloader)))
# Save checkpoint
global best_loss
test_loss /= len(testloader)
if test_loss < best_loss:
print('Saving..')
state = {
'net': net.module.state_dict(),
'loss': test_loss,
'epoch': epoch,
}
if not os.path.isdir(os.path.dirname(args.checkpoint)):
os.mkdir(os.path.dirname(args.checkpoint))
torch.save(state, args.checkpoint)
best_loss = test_loss
for epoch in range(start_epoch, start_epoch+200):
train(epoch)
test(epoch)