-
Notifications
You must be signed in to change notification settings - Fork 10
/
main_recog.py
221 lines (190 loc) · 7.96 KB
/
main_recog.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
from __future__ import print_function
import argparse
import os
import sys
import random
import time
import torch
import torch.nn as nn
import torch.nn.parallel
import torch.backends.cudnn as cudnn
cudnn.benchmark = True
cudnn.fastest = True
import torch.optim as optim
from torch.autograd import Variable
import torchvision.transforms as transforms
import models.DTNet_32 as net
from misc import *
parser = argparse.ArgumentParser()
parser.add_argument('--dataset', required=False,
default='svhn', help='')
parser.add_argument('--dataroot', required=False,
default='')
parser.add_argument('--valDataroot', required=False,
default='')
parser.add_argument('--batchSize', type=int, default=64, help='input batch size')
parser.add_argument('--valBatchSize', type=int, default=64, help='input batch size')
parser.add_argument('--originalSize', type=int,
default=36, help='the height / width of the original input image')
parser.add_argument('--imageSize', type=int,
default=32, help='the height / width of the cropped input image to network')
parser.add_argument('--nClasses', type=int,
default=10, help='size of the input channels')
parser.add_argument('--netE', default='', help="path to netE (to continue training)")
parser.add_argument('--lrE', type=float, default=0.001, help='learning rate, default=0.0002')
parser.add_argument('--beta1', type=float, default=0.9, help='beta1 for adam. default=0.9')
parser.add_argument('--rmsprop', action='store_true', help='Whether to use adam (default is adam)')
parser.add_argument('--wd', type=float, default=0.0000, help='weight decay, default=0.000')
parser.add_argument('--niter', type=int, default=150, help='number of epochs to train for')
parser.add_argument('--workers', type=int, help='number of data loading workers', default=4)
parser.add_argument('--exp', default='recog', help='folder to output images and model checkpoints')
parser.add_argument('--display', type=int, default=50, help='Where to store samples and models')
opt = parser.parse_args()
print(opt)
create_exp_dir(opt.exp)
#opt.manualSeed = 10
opt.manualSeed = random.randint(1, 10000)
torch.manual_seed(opt.manualSeed)
print("Random Seed: ", opt.manualSeed)
# get dataloader and preprocessor
#import pdb; pdb.set_trace()
mean, std = (0.5, 0.5, 0.5), (0.5, 0.5, 0.5)
transform_trn = transforms.Compose([transforms.Scale(opt.originalSize),
transforms.RandomCrop(opt.imageSize),
transforms.RandomHorizontalFlip(),
transforms.ToTensor(),
transforms.Normalize(mean, std),
])
transform_tst = transforms.Compose([transforms.Scale(opt.originalSize),
transforms.CenterCrop(opt.imageSize),
transforms.ToTensor(),
transforms.Normalize(mean, std),
])
dataloader, _ = getLoader(opt.dataset,
opt.dataroot,
opt.originalSize,
opt.imageSize,
opt.batchSize,
opt.workers,
mean=mean, std=std,
#split='extra', #SVHN
split='train', #MNIST
transform_fn=transform_trn)
valDataloader, _ = getLoader(opt.dataset,
opt.valDataroot,
opt.imageSize,
opt.imageSize,
opt.valBatchSize,
opt.workers,
mean=mean, std=std,
split='test',
transform_fn=transform_tst)
ngf = 64
inputChannelSize = 3
netE = net.E(inputChannelSize, ngf, opt.nClasses)
netE.apply(weights_init)
if opt.netE != '':
netE.load_state_dict(torch.load(opt.netE))
print(netE)
criterionCE = nn.CrossEntropyLoss()
netE.cuda()
criterionCE.cuda()
if opt.rmsprop:
optimizerE = optim.RMSprop(netE.parameters(), lr = opt.lrE)
else:
optimizerE = optim.Adam(netE.parameters(), lr = opt.lrE, betas = (opt.beta1, 0.999), weight_decay=opt.wd)
def validate(val_loader, model, criterion, display=100):
batch_time = AverageMeter()
losses = AverageMeter()
top1 = AverageMeter()
top5 = AverageMeter()
model.eval()
end = time.time()
for i, (input_, target) in enumerate(val_loader):
input = torch.FloatTensor(input_.size(0), 3, input_.size(2), input_.size(3))
if opt.dataset == 'mnist':
input[:,0,:,:].copy_(input_)
input[:,1,:,:].copy_(input_)
input[:,2,:,:].copy_(input_)
else:
target -= 1
input.copy_(input_)
input = input.cuda(async=True)
target = torch.LongTensor(target.size(0)).copy_(target).cuda(async=True)
input_var = torch.autograd.Variable(input, volatile=True)
target_var = torch.autograd.Variable(target, volatile=True)
output,_ = model(input_var)
loss = criterion(output.squeeze(3).squeeze(2), target_var)
prec1, prec5 = accuracy(output.data.squeeze(3).squeeze(2), target, topk=(1, 5))
losses.update(loss.data[0], input.size(0))
top1.update(prec1[0], input.size(0))
top5.update(prec5[0], input.size(0))
batch_time.update(time.time() - end)
end = time.time()
if i % display == 0:
print('Test: [{0}/{1}]\t'
'Time {batch_time.val:.3f} ({batch_time.avg:.3f})\t'
'Loss {loss.val:.4f} ({loss.avg:.4f})\t'
'Prec@1 {top1.val:.3f} ({top1.avg:.3f})\t'
'Prec@5 {top5.val:.3f} ({top5.avg:.3f})'.format(
i, len(val_loader), batch_time=batch_time, loss=losses,
top1=top1, top5=top5))
print(' * Prec@1 {top1.avg:.3f} Prec@5 {top5.avg:.3f}'
.format(top1=top1, top5=top5))
sys.stdout.flush()
return top1.avg
def train(train_loader, model, criterion, optimizer, epoch, display=100):
batch_time = AverageMeter()
data_time = AverageMeter()
losses = AverageMeter()
top1 = AverageMeter()
top5 = AverageMeter()
model.train()
end = time.time()
for i, (input_, target) in enumerate(train_loader):
data_time.update(time.time() - end)
input = torch.FloatTensor(input_.size(0), 3, input_.size(2), input_.size(3))
if opt.dataset == 'mnist':
input[:,0,:,:].copy_(input_)
input[:,1,:,:].copy_(input_)
input[:,2,:,:].copy_(input_)
else:
target -= 1
input.copy_(input_)
input = input.cuda(async=True)
target = torch.LongTensor(target.size(0)).copy_(target).cuda(async=True)
input_var = torch.autograd.Variable(input)
target_var = torch.autograd.Variable(target)
output,_ = model(input_var)
loss = criterion(output.squeeze(3).squeeze(2), target_var)
prec1, prec5 = accuracy(output.data.squeeze(3).squeeze(2), target, topk=(1, 5))
losses.update(loss.data[0], input.size(0))
top1.update(prec1[0], input.size(0))
top5.update(prec5[0], input.size(0))
optimizer.zero_grad()
loss.backward()
optimizer.step()
batch_time.update(time.time() - end)
end = time.time()
if i % display == 0:
print('Epoch: [{0}][{1}/{2}]\t'
'Time {batch_time.val:.3f} ({batch_time.avg:.3f})\t'
'Data {data_time.val:.3f} ({data_time.avg:.3f})\t'
'Loss {loss.val:.4f} ({loss.avg:.4f})\t'
'Prec@1 {top1.val:.3f} ({top1.avg:.3f})\t'
'Prec@5 {top5.val:.3f} ({top5.avg:.3f})'.format(
epoch, i, len(train_loader), batch_time=batch_time,
data_time=data_time, loss=losses, top1=top1, top5=top5))
sys.stdout.flush()
best_prec1 = -1
best_epoch = -1
#import pdb; pdb.set_trace()
for epoch in range(opt.niter):
train(dataloader, netE, criterionCE, optimizerE, epoch)
prec1 = validate(valDataloader, netE, criterionCE)
is_best = prec1 > best_prec1
if is_best:
best_prec1 = prec1
best_epoch = epoch
torch.save(netE.state_dict(), '%s/netE_epoch_%d.pth' % (opt.exp, epoch))
print('Best accuracy: %f in epoch %d' % (best_prec1, best_epoch))