forked from chenxin061/pdarts
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathtrain_cifar.py
188 lines (159 loc) · 7.51 KB
/
train_cifar.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
import os
import sys
import time
import glob
import numpy as np
import torch
import utils
import logging
import argparse
import torch.nn as nn
import genotypes
import torch.utils
import torchvision.datasets as dset
import torch.backends.cudnn as cudnn
from torch.autograd import Variable
from model import NetworkCIFAR as Network
parser = argparse.ArgumentParser("cifar")
parser.add_argument('--workers', type=int, default=4, help='number of workers')
parser.add_argument('--batch_size', type=int, default=128, help='batch size')
parser.add_argument('--learning_rate', type=float, default=0.025, help='init learning rate')
parser.add_argument('--momentum', type=float, default=0.9, help='momentum')
parser.add_argument('--weight_decay', type=float, default=3e-4, help='weight decay')
parser.add_argument('--report_freq', type=float, default=50, help='report frequency')
parser.add_argument('--epochs', type=int, default=600, help='num of training epochs')
parser.add_argument('--init_channels', type=int, default=36, help='num of init channels')
parser.add_argument('--layers', type=int, default=20, help='total number of layers')
parser.add_argument('--auxiliary', action='store_true', default=False, help='use auxiliary tower')
parser.add_argument('--auxiliary_weight', type=float, default=0.4, help='weight for auxiliary loss')
parser.add_argument('--cutout', action='store_true', default=False, help='use cutout')
parser.add_argument('--cutout_length', type=int, default=16, help='cutout length')
parser.add_argument('--drop_path_prob', type=float, default=0.3, help='drop path probability')
parser.add_argument('--save', type=str, default='/tmp/checkpoints/', help='experiment name')
parser.add_argument('--seed', type=int, default=0, help='random seed')
parser.add_argument('--arch', type=str, default='PDARTS', help='which architecture to use')
parser.add_argument('--grad_clip', type=float, default=5, help='gradient clipping')
parser.add_argument('--tmp_data_dir', type=str, default='/tmp/cache/', help='temp data dir')
parser.add_argument('--note', type=str, default='try', help='note for this run')
parser.add_argument('--cifar100', action='store_true', default=False, help='if use cifar100')
args, unparsed = parser.parse_known_args()
args.save = '{}eval-{}-{}'.format(args.save, args.note, time.strftime("%Y%m%d-%H%M%S"))
utils.create_exp_dir(args.save, scripts_to_save=glob.glob('*.py'))
log_format = '%(asctime)s %(message)s'
logging.basicConfig(stream=sys.stdout, level=logging.INFO,
format=log_format, datefmt='%m/%d %I:%M:%S %p')
fh = logging.FileHandler(os.path.join(args.save, 'log.txt'))
fh.setFormatter(logging.Formatter(log_format))
logging.getLogger().addHandler(fh)
if args.cifar100:
CIFAR_CLASSES = 100
data_folder = 'cifar-100-python'
else:
CIFAR_CLASSES = 10
data_folder = 'cifar-10-batches-py'
def main():
if not torch.cuda.is_available():
logging.info('No GPU device available')
sys.exit(1)
np.random.seed(args.seed)
cudnn.benchmark = True
torch.manual_seed(args.seed)
cudnn.enabled=True
torch.cuda.manual_seed(args.seed)
logging.info("args = %s", args)
logging.info("unparsed args = %s", unparsed)
num_gpus = torch.cuda.device_count()
genotype = eval("genotypes.%s" % args.arch)
print('---------Genotype---------')
logging.info(genotype)
print('--------------------------')
model = Network(args.init_channels, CIFAR_CLASSES, args.layers, args.auxiliary, genotype)
model = torch.nn.DataParallel(model)
model = model.cuda()
logging.info("param size = %fMB", utils.count_parameters_in_MB(model))
criterion = nn.CrossEntropyLoss()
criterion = criterion.cuda()
optimizer = torch.optim.SGD(
model.parameters(),
args.learning_rate,
momentum=args.momentum,
weight_decay=args.weight_decay
)
if args.cifar100:
train_transform, valid_transform = utils._data_transforms_cifar100(args)
else:
train_transform, valid_transform = utils._data_transforms_cifar10(args)
if args.cifar100:
train_data = dset.CIFAR100(root=args.tmp_data_dir, train=True, download=True, transform=train_transform)
valid_data = dset.CIFAR100(root=args.tmp_data_dir, train=False, download=True, transform=valid_transform)
else:
train_data = dset.CIFAR10(root=args.tmp_data_dir, train=True, download=True, transform=train_transform)
valid_data = dset.CIFAR10(root=args.tmp_data_dir, train=False, download=True, transform=valid_transform)
train_queue = torch.utils.data.DataLoader(
train_data, batch_size=args.batch_size, shuffle=True, pin_memory=True, num_workers=args.workers)
valid_queue = torch.utils.data.DataLoader(
valid_data, batch_size=args.batch_size, shuffle=False, pin_memory=True, num_workers=args.workers)
scheduler = torch.optim.lr_scheduler.CosineAnnealingLR(optimizer, float(args.epochs))
best_acc = 0.0
for epoch in range(args.epochs):
scheduler.step()
logging.info('Epoch: %d lr %e', epoch, scheduler.get_lr()[0])
model.module.drop_path_prob = args.drop_path_prob * epoch / args.epochs
model.drop_path_prob = args.drop_path_prob * epoch / args.epochs
start_time = time.time()
train_acc, train_obj = train(train_queue, model, criterion, optimizer)
logging.info('Train_acc: %f', train_acc)
valid_acc, valid_obj = infer(valid_queue, model, criterion)
if valid_acc > best_acc:
best_acc = valid_acc
logging.info('Valid_acc: %f', valid_acc)
end_time = time.time()
duration = end_time - start_time
print('Epoch time: %ds.' % duration )
utils.save(model.module, os.path.join(args.save, 'weights.pt'))
def train(train_queue, model, criterion, optimizer):
objs = utils.AvgrageMeter()
top1 = utils.AvgrageMeter()
model.train()
for step, (input, target) in enumerate(train_queue):
input = input.cuda(non_blocking=True)
target = target.cuda(non_blocking=True)
optimizer.zero_grad()
logits, logits_aux = model(input)
loss = criterion(logits, target)
if args.auxiliary:
loss_aux = criterion(logits_aux, target)
loss += args.auxiliary_weight*loss_aux
loss.backward()
nn.utils.clip_grad_norm_(model.parameters(), args.grad_clip)
optimizer.step()
prec1, _ = utils.accuracy(logits, target, topk=(1,5))
n = input.size(0)
objs.update(loss.data.item(), n)
top1.update(prec1.data.item(), n)
if step % args.report_freq == 0:
logging.info('Train Step: %03d Objs: %e Acc: %f', step, objs.avg, top1.avg)
return top1.avg, objs.avg
def infer(valid_queue, model, criterion):
objs = utils.AvgrageMeter()
top1 = utils.AvgrageMeter()
model.eval()
for step, (input, target) in enumerate(valid_queue):
input = input.cuda(non_blocking=True)
target = target.cuda(non_blocking=True)
with torch.no_grad():
logits, _ = model(input)
loss = criterion(logits, target)
prec1, _ = utils.accuracy(logits, target, topk=(1,5))
n = input.size(0)
objs.update(loss.data.item(), n)
top1.update(prec1.data.item(), n)
if step % args.report_freq == 0:
logging.info('Valid Step: %03d Objs: %e Acc: %f', step, objs.avg, top1.avg)
return top1.avg, objs.avg
if __name__ == '__main__':
start_time = time.time()
main()
end_time = time.time()
duration = end_time - start_time
logging.info('Eval time: %ds.', duration)