-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathtrain.py
executable file
·180 lines (164 loc) · 7.64 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
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
import os,sys
sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__)))) # to import shared utils
import torch
import torch.optim as optim
import torch.nn as nn
import numpy as np
import argparse
import time
import logging
logging.basicConfig(level=logging.INFO, format='%(asctime)s %(levelname)-8s %(message)s')
logFormatter = logging.Formatter('%(asctime)s %(levelname)-8s %(message)s')
rootLogger = logging.getLogger()
from DataLoader import VQADataLoader
# from model.net import XNMNet
from model.net import NKM
from utils.misc import todevice
from validate import validate
def train(args):
logging.info("Create train_loader and val_loader.........")
train_loader_kwargs = {
'question_pt': args.train_question_pt,
'vocab_json': args.vocab_json,
'feature_h5': args.feature_h5,
'batch_size': args.batch_size,
'spatial': args.spatial,
'num_workers': 4,
'shuffle': True
}
train_loader = VQADataLoader(**train_loader_kwargs)
if args.val:
val_loader_kwargs = {
'question_pt': args.val_question_pt,
'vocab_json': args.vocab_json,
'feature_h5': args.feature_h5,
'batch_size': args.batch_size,
'spatial': args.spatial,
'num_workers': 4,
'shuffle': False
}
val_loader = VQADataLoader(**val_loader_kwargs)
logging.info("Create model.........")
device = 'cuda' if torch.cuda.is_available() else 'cpu'
model_kwargs = {
'vocab': train_loader.vocab,
'dim_v': args.dim_v,
'dim_word': args.dim_word,
'dim_hidden': args.dim_hidden,
'dim_vision': args.dim_vision,
'dim_edge': args.dim_edge,
'cls_fc_dim': args.cls_fc_dim,
'dropout_prob': args.dropout,
'T_ctrl': args.T_ctrl,
'glimpses': args.glimpses,
'stack_len': args.stack_len,
'device': device,
'spatial': args.spatial,
'use_gumbel': args.module_prob_use_gumbel==1,
'use_validity': args.module_prob_use_validity==1,
}
model_kwargs_tosave = { k:v for k,v in model_kwargs.items() if k != 'vocab' }
model = NKM(**model_kwargs).to(device)
logging.info(model)
logging.info('load glove vectors')
train_loader.glove_matrix = torch.FloatTensor(train_loader.glove_matrix).to(device)
model.token_embedding.weight.data.set_(train_loader.glove_matrix)
################################################################
parameters = [p for p in model.parameters() if p.requires_grad]
optimizer = optim.Adam(parameters, args.lr, weight_decay=1e-4)
start_epoch = 0
if args.restore:
print("Restore checkpoint and optimizer...")
ckpt = os.path.join(args.save_dir, 'model.pt')
ckpt = torch.load(ckpt, map_location={'cuda:0': 'cpu'})
start_epoch = ckpt['epoch'] + 1
model.load_state_dict(ckpt['state_dict'])
optimizer.load_state_dict(ckpt['optimizer'])
scheduler = optim.lr_scheduler.ExponentialLR(optimizer, 0.5**(1 / args.lr_halflife))
logging.info("Start training........")
for epoch in range(start_epoch, args.num_epoch):
model.train()
for i, batch in enumerate(train_loader):
progress = epoch+i/len(train_loader)
coco_ids, answers, *batch_input = [todevice(x, device) for x in batch]
logits, others = model(*batch_input)
##################### loss #####################
nll = -nn.functional.log_softmax(logits, dim=1)
loss = (nll * answers / 10).sum(dim=1).mean()
#################################################
scheduler.step()
optimizer.zero_grad()
loss.backward()
nn.utils.clip_grad_value_(parameters, clip_value=0.5)
optimizer.step()
if (i+1) % (len(train_loader) // 50) == 0:
logging.info("Progress %.3f ce_loss = %.3f" % (progress, loss.item()))
save_checkpoint(epoch, model, optimizer, model_kwargs_tosave, os.path.join(args.save_dir, 'model.pt'))
logging.info(' >>>>>> save to %s <<<<<<' % (args.save_dir))
if args.val:
valid_acc = validate(model, val_loader, device)
logging.info('\n ~~~~~~ Valid Accuracy: %.4f ~~~~~~~\n' % valid_acc)
def save_checkpoint(epoch, model, optimizer, model_kwargs, filename):
state = {
'epoch': epoch,
'state_dict': model.state_dict(),
'optimizer': optimizer.state_dict(),
'model_kwargs': model_kwargs,
}
torch.save(state, filename)
def main():
parser = argparse.ArgumentParser()
# input and output
parser.add_argument('--save_dir', type=str, required=True, help='path to save checkpoints and logs')
parser.add_argument('--input_dir', required=True)
parser.add_argument('--train_question_pt', default='train_questions.pt')
parser.add_argument('--val_question_pt', default='val_questions.pt')
parser.add_argument('--vocab_json', default='vocab.json')
parser.add_argument('--feature_h5', default='trainval_feature.h5')
parser.add_argument('--restore', action='store_true')
# training parameters
parser.add_argument('--lr', default=8e-4, type=float)
parser.add_argument('--lr_halflife', default=50000, type=int)
parser.add_argument('--num_epoch', default=100, type=int)
parser.add_argument('--batch_size', default=256, type=int)
parser.add_argument('--seed', type=int, default=666, help='random seed')
parser.add_argument('--val', action='store_true', help='whether validate after each training epoch')
# model hyperparameters
parser.add_argument('--dim_word', default=300, type=int, help='word embedding')
parser.add_argument('--dim_hidden', default=1024, type=int, help='hidden state of seq2seq parser')
parser.add_argument('--dim_v', default=512, type=int, help='node embedding')
parser.add_argument('--dim_edge', default=256, type=int, help='edge embedding')
parser.add_argument('--dim_vision', default=2048, type=int)
parser.add_argument('--cls_fc_dim', default=1024, type=int, help='classifier fc dim')
parser.add_argument('--glimpses', default=2, type=int)
parser.add_argument('--dropout', default=0.5, type=float)
parser.add_argument('--T_ctrl', default=3, type=int, help='controller decode length')
parser.add_argument('--stack_len', default=4, type=int, help='stack length')
parser.add_argument('--spatial', action='store_true')
parser.add_argument('--module_prob_use_gumbel', default=0, choices=[0, 1], type=int, help='whether use gumbel softmax for module prob. 0 not use, 1 use')
parser.add_argument('--module_prob_use_validity', default=1, choices=[0, 1], type=int, help='whether validate module prob.')
args = parser.parse_args()
# make logging.info display into both shell and file
if not args.restore:
os.mkdir(args.save_dir)
else:
assert os.path.isdir(args.save_dir)
fileHandler = logging.FileHandler(os.path.join(args.save_dir, 'stdout.log'))
fileHandler.setFormatter(logFormatter)
rootLogger.addHandler(fileHandler)
# args display
for k, v in vars(args).items():
logging.info(k+':'+str(v))
# concat obsolute path of input files
args.train_question_pt = os.path.join(args.input_dir, args.train_question_pt)
args.vocab_json = os.path.join(args.input_dir, args.vocab_json)
args.val_question_pt = os.path.join(args.input_dir, args.val_question_pt)
args.feature_h5 = os.path.join(args.input_dir, args.feature_h5)
# set random seed
torch.manual_seed(args.seed)
np.random.seed(args.seed)
if args.spatial:
args.dim_vision += 5
train(args)
if __name__ == '__main__':
main()