-
Notifications
You must be signed in to change notification settings - Fork 0
/
train_search.py
179 lines (139 loc) · 5.51 KB
/
train_search.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
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 torch.utils
import torch.nn.functional as F
import torchvision.datasets as dset
import torch.backends.cudnn as cudnn
from torchvision import transforms
from torch.autograd import Variable
from model_search import Network
from architect import Architect
from datasets import KMNIST, K49
from settings import get_darts_args
def darts(exp_name, args):
if not torch.cuda.is_available():
logging.info('no gpu device available')
sys.exit(1)
args['save'] = './{}/{}-{}-{}'.format(exp_name, args['save'], time.strftime("%Y%m%d-%H%M%S"), args['seed'])
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)
np.random.seed(args['seed'])
torch.cuda.set_device(args['gpu'])
cudnn.benchmark = True
torch.manual_seed(args['seed'])
cudnn.enabled = True
torch.cuda.manual_seed(args['seed'])
logging.info('gpu device = %s' % args['gpu'])
logging.info("args = %s", args)
data_augmentations = transforms.ToTensor()
train_data = KMNIST(args['data'], True, data_augmentations)
criterion = nn.CrossEntropyLoss()
criterion = criterion.cuda()
model = Network(args['init_channels'], train_data.n_classes, args['layers'], criterion)
model = model.cuda()
logging.info("param size = %fMB", utils.count_parameters_in_MB(model))
optimizer = torch.optim.SGD(
model.parameters(),
args['learning_rate'],
momentum=args['momentum'],
weight_decay=args['weight_decay'])
num_train = len(train_data)
indices = list(range(num_train))
split = int(np.floor(args['train_portion'] * num_train))
train_queue = torch.utils.data.DataLoader(
train_data, batch_size=args['batch_size'],
sampler=torch.utils.data.sampler.SubsetRandomSampler(indices[:split]))
valid_queue = torch.utils.data.DataLoader(
train_data, batch_size=args['batch_size'],
sampler=torch.utils.data.sampler.SubsetRandomSampler(indices[split:num_train]))
scheduler = torch.optim.lr_scheduler.CosineAnnealingLR(
optimizer, float(args['epochs']), eta_min=args['learning_rate_min'])
architect = Architect(model, args)
for epoch in range(args['epochs']):
scheduler.step()
lr = scheduler.get_lr()[0]
logging.info('epoch %d lr %e', epoch, lr)
genotype = model.genotype()
logging.info('genotype = %s', genotype)
print(F.softmax(model.alphas_normal, dim=-1))
print(F.softmax(model.alphas_reduce, dim=-1))
print(F.softmax(model.betas_normal[2:5], dim=-1))
#model.drop_path_prob = args['drop_path_prob * epoch / args['epochs
# training
train_acc, train_obj = train(train_queue, valid_queue, model, architect, criterion, optimizer, lr,epoch)
logging.info('train_acc %f', train_acc)
# validation
if args['epochs']-epoch<=1:
valid_acc, valid_obj = infer(valid_queue, model, criterion)
logging.info('valid_acc %f', valid_acc)
utils.save(model, os.path.join(args['save'], 'weights.pt'))
def train(train_queue, valid_queue, model, architect, criterion, optimizer, lr,epoch):
objs = utils.AvgrageMeter()
top1 = utils.AvgrageMeter()
top5 = utils.AvgrageMeter()
args = get_darts_args()
for step, (input, target) in enumerate(train_queue):
model.train()
n = input.size(0)
input = input.cuda()
target = target.cuda(non_blocking=True)
# get a random minibatch from the search queue with replacement
#input_search, target_search = next(iter(valid_queue))
try:
input_search, target_search = next(valid_queue_iter)
except:
valid_queue_iter = iter(valid_queue)
input_search, target_search = next(valid_queue_iter)
input_search = input_search.cuda()
target_search = target_search.cuda(non_blocking=True)
if epoch >= 15:
architect.step(input, target, input_search, target_search, lr, optimizer, unrolled=args['unrolled'])
optimizer.zero_grad()
logits = model(input)
loss = criterion(logits, target)
loss.backward()
nn.utils.clip_grad_norm(model.parameters(), args['grad_clip'])
optimizer.step()
prec1, prec5 = utils.accuracy(logits, target, topk=(1, 5))
objs.update(loss.data.item(), n)
top1.update(prec1.data.item(), n)
top5.update(prec5.data.item(), n)
if step % args['report_freq'] == 0:
logging.info('train %03d %e %f %f', step, objs.avg, top1.avg, top5.avg)
return top1.avg, objs.avg
def infer(valid_queue, model, criterion):
objs = utils.AvgrageMeter()
top1 = utils.AvgrageMeter()
top5 = utils.AvgrageMeter()
model.eval()
args = get_darts_args()
with torch.no_grad():
for step, (input, target) in enumerate(valid_queue):
input = input.cuda()
target = target.cuda(non_blocking=True)
logits = model(input)
loss = criterion(logits, target)
prec1, prec5 = utils.accuracy(logits, target, topk=(1, 5))
n = input.size(0)
objs.update(loss.data.item(), n)
top1.update(prec1.data.item(), n)
top5.update(prec5.data.item(), n)
if step % args['report_freq'] == 0:
logging.info('valid %03d %e %f %f', step, objs.avg, top1.avg, top5.avg)
return top1.avg, objs.avg
if __name__ == '__main__':
args = get_darts_args()
darts(0, args)