-
Notifications
You must be signed in to change notification settings - Fork 0
/
train.py
157 lines (128 loc) · 4.7 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
from __future__ import division, print_function
import torch
import torch.nn as nn
import torch.optim as optim
from torch.autograd import Variable
import os
import shutil
import numpy as np
import argparse
from dataset import load_dataset
from densenet import DenseNet
from tqdm import tqdm
parser = argparse.ArgumentParser(description='Densenet for Classification')
parser.add_argument('--gpu', default="0", type=str)
parser.add_argument('--dataset', default="cifar10", type=str)
parser.add_argument('--batch_size', default=128, type=int)
parser.add_argument('--epoch', default=300, type=int)
parser.add_argument('--seed', default=1, type=int)
parser.add_argument('--checkpoint', default="model", type=str)
parser.add_argument('--resume_model', default=None, type=str)
parser.add_argument('--lr', default=0.1, type=float)
parser.add_argument('--momentum', default=0.9, type=float)
FLAGS = parser.parse_args()
os.environ["CUDA_VISIBLE_DEVICES"]=FLAGS.gpu
def init():
torch.manual_seed(FLAGS.seed)
torch.cuda.manual_seed(FLAGS.seed)
torch.backends.cudnn.benchmark = True
data, loader = load_dataset(FLAGS)
print('==> data loaded')
model = DenseNet(classes=FLAGS.classes)
model.cuda()
print('==> model loaded')
optimizer = optim.SGD(model.parameters(), lr=FLAGS.lr, momentum=FLAGS.momentum, nesterov=True)
criterion = nn.CrossEntropyLoss().cuda()
return model, optimizer, criterion, loader
def exp_lr_scheduler(optimizer, epoch, target_epoch, lr):
lr = lr
if float(epoch) / target_epoch > 0.75:
lr = lr * 0.01
elif float(epoch) / target_epoch > 0.5:
lr = lr * 0.1
for param_group in optimizer.param_groups:
param_group['lr'] = lr
def train(model, optimizer, criterion, loader, epoch):
model.train()
Acc = 0.0
Loss = 0.0
for batch_idx, (input, label) in enumerate(loader):
label = Variable(label.cuda(async=True))
input = Variable(input.cuda())
output = model(input)
loss = criterion(output, label)
prec = accuracy(output.data, label.data, topk=(1,))[0]
optimizer.zero_grad()
loss.backward()
optimizer.step()
if batch_idx % 50 == 0:
print('[Train] Epoch: {}/{} Batch: {}/{} Accuracy: {:.6f} Loss: {:.7f}'.format(epoch, FLAGS.epoch, batch_idx, len(loader), prec, loss.data[0]))
Acc += prec
Loss += loss.data[0]
Acc /= len(loader)
Loss /= len(loader)
print('[Train] Epoch {} Average Accuracy: {:.6f} Average Loss {:.7f}'.format(epoch, Acc, Loss))
def valid(model, criterion, loader, epoch):
model.eval()
Acc = 0.0
Loss = 0.0
for batch_idx, (input, label) in enumerate(loader):
label = Variable(label.cuda(async=True), volatile=True)
input = Variable(input.cuda(), volatile=True)
output = model(input)
loss = criterion(output, label)
prec = accuracy(output.data, label.data, topk=(1,))[0]
Acc += prec
Loss += loss.data[0]
Acc /= len(loader)
Loss /= len(loader)
print('[Valid] Epoch {} Average Accuracy: {:.6f} Average Loss {:.7f}'.format(epoch, Acc, Loss))
return Acc
def save_checkpoint(state, filename='checkpoint.pth.tar'):
torch.save(state, filename)
def resume(filename, optimizer, model):
if os.path.isfile(filename):
print('==> loading checkpoint {}'.format(filename))
checkpoint = torch.load(filename)
start_epoch = checkpoint['epoch'] + 1
model.load_state_dict(checkpoint['state_dict'])
optimizer.load_state_dict(checkpoint['optimizer'])
print("==> loaded checkpoint '{}' (epoch {})".format(filename, start_epoch))
else:
print("==> no checkpoint found at '{}'".format(filename))
return model, optimizer, start_epoch
def main():
if not os.path.isdir(FLAGS.checkpoint):
os.mkdir(FLAGS.checkpoint)
model, optimizer, criterion, loader = init()
start_epoch = 1
best_acc = 0.0
if FLAGS.resume_model is not None:
filename = os.path.join(FLAGS.checkpoint, FLAGS.resume_model)
model, optimizer, start_epoch = resume(filename, optimizer, model)
for epoch in range(start_epoch, FLAGS.epoch):
exp_lr_scheduler(optimizer, epoch, FLAGS.epoch, FLAGS.lr)
train(model, optimizer, criterion, loader['train'], epoch)
acc = valid(model, criterion, loader['test'], epoch)
filename = os.path.join(FLAGS.checkpoint, "densenet_{}.pth".format(epoch))
save_checkpoint({
'epoch': epoch,
'state_dict': model.state_dict(),
'optimizer': optimizer.state_dict(),
}, filename=filename)
if acc > best_acc:
best_acc =acc
shutil.copyfile(filename, os.path.join(FLAGS.checkpoint,"best_model.pth"))
def accuracy(output, target, topk=(1,)):
maxk = max(topk)
batch_size = target.size(0)
value, pred = output.topk(maxk, 1, True, True)
pred = pred.t()
correct = pred.eq(target.view(1, -1).expand_as(pred))
res = []
for k in topk:
correct_k = correct[:k].view(-1).float().sum(0)
res.append(correct_k[0] / float(batch_size))
return res
if __name__ == '__main__':
main()