forked from Yuejiang-li/info-diff
-
Notifications
You must be signed in to change notification settings - Fork 0
/
train.py
146 lines (120 loc) · 5.62 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
import os
import argparse
import torch
import numpy as np
import random
import pickle
import time
from datetime import datetime
from utils import constant
# from utils.metrics import metrics
from utils.metrics import metrics, tune_thres, get_preds, tune_thres_new
from loader import DataLoader, EmbLoader
from model import ModelWrapper
from shutil import copyfile
parser = argparse.ArgumentParser()
parser.add_argument('--cuda', type=bool, default=torch.cuda.is_available())
parser.add_argument('--model', type=str, default='InfoLSTM')
parser.add_argument('--data_dir', type=str, default='./data/final')
parser.add_argument('--emb_file', type=str, default='./data/content_dict.pkl')
parser.add_argument('--idx_dict', type=str, default='./data/final/idx_dict.pkl')
parser.add_argument('--followee_count_file', type=str, default='./data/followee_count.pkl')
parser.add_argument('--model_save_dir', type=str, default='./data/saved_model')
parser.add_argument('--save_epoch', type=int, default=5, help='Save model checkpoints every k epochs.')
parser.add_argument('--hidden_dim', type=int, default=200, help='RNN hidden state size.')
parser.add_argument('--num_layers', type=int, default=2, help='Num of RNN layers.')
parser.add_argument('--dropout', type=float, default=0.5, help='Input and RNN dropout rate.')
parser.add_argument('--penalty_coeff', type=float, default=0.5, help='Coefficient of Penalty term used in Clash model')
parser.add_argument('--num_epoch', type=int, default=30)
parser.add_argument('--log_step', type=int, default=20, help='Print log every k steps.')
parser.add_argument('--window_size', type=int, default=10)
parser.add_argument('--no_extra_linear', dest='use_extra_linear', action='store_false')
parser.add_argument('--use_extra_linear', dest='use_extra_linear', action='store_true')
parser.set_defaults(use_extra_linear=True)
parser.add_argument('--train_emb', dest='fix_emb', action='store_false')
parser.add_argument('--fix_emb', dest='fix_emb', action='store_true')
parser.set_defaults(fix_emb=False)
parser.add_argument('--patience', type=int, default=3)
parser.add_argument('--lr_decay', type=float, default=0.5)
parser.add_argument('--lr', type=float, default=1.0, help='Applies to SGD')
parser.add_argument('--max_grad_norm', type=float, default=5.0, help='Gradient clipping.')
parser.add_argument('--emb_dim', type=int, default=768)
parser.add_argument('--batch_size', type=int, default=256)
parser.add_argument('--seed', type=int, default=99)
# parser.add_argument('--cuda', type=bool, default=torch.cuda.is_available())
args = parser.parse_args()
torch.manual_seed(args.seed)
np.random.seed(args.seed)
random.seed(args.seed)
# if args.cpu:
# args.cuda = False
# elif args.cuda:
# torch.cuda.manual_seed(args.seed)
# make opt
opt = vars(args)
print(opt)
with open(opt['idx_dict'], 'rb') as fin:
weibo2embid = pickle.load(fin)
train_batch = DataLoader(os.path.join(opt['data_dir'], 'train.csv'),
opt['batch_size'],
opt,
weibo2embid=weibo2embid,
evaluation=False)
dev_batch = DataLoader(os.path.join(opt['data_dir'], 'dev.csv'),
opt['batch_size'],
opt,
weibo2embid=weibo2embid,
evaluation=True)
# test_batch = DataLoader(os.path.join(opt['data_dir'], 'test.csv'),
# opt['batch_size'],
# opt,
# weibo2embid=weibo2embid,
# evaluation=True)
if not os.path.exists(opt['model_save_dir']):
os.mkdir(opt['model_save_dir'])
weibo2embid = train_batch.weibo2embid
model = ModelWrapper(opt, weibo2embid, train_batch.retw_prob)
global_step = 0
# current_lr = opt['lr']
max_steps = len(train_batch) * opt['num_epoch']
global_start_time = time.time()
format_str = '{}: step {}/{} (epoch {}/{}), loss = {:.6f} ({:.3f} sec/batch), lr: {:.6f}'
dev_f1_history = []
for epoch in range(1, opt['num_epoch'] + 1):
train_loss = 0
current_lr = model.optimizer.param_groups[0]['lr']
for i, batch in enumerate(train_batch):
start_time = time.time()
global_step += 1
loss = model.update(batch)
train_loss += loss
if global_step % opt['log_step'] == 0:
duration = time.time() - start_time
print(format_str.format(datetime.now(), global_step, max_steps, epoch, \
opt['num_epoch'], loss, duration, current_lr))
print("Evaluating on dev set ...")
predictions = []
probabilities = []
dev_loss = 0
for i, batch in enumerate(dev_batch):
preds, probs, loss = model.predict(batch, thres=0.01)
predictions += preds
probabilities += probs
dev_loss += loss
train_loss = train_loss / train_batch.num_examples * opt['batch_size'] # avg loss per batch
dev_loss = dev_loss / dev_batch.num_examples * opt['batch_size']
model.scheduler.step(dev_loss)
# _, _, _, dev_f1 = metrics(dev_batch.gold(), predictions)
_, _, _, dev_f1, _ = tune_thres_new(dev_batch.gold(), probabilities)
print("epoch {}: train_loss = {:.6f}, dev_loss = {:.6f}, dev_f1 = {:.4f}".format(epoch,\
train_loss, dev_loss, dev_f1))
model_file = os.path.join(opt['model_save_dir'], 'checkpoint_epoch_{}.pt'.format(epoch))
model.save(model_file, epoch)
if epoch == 1 or dev_f1 > max(dev_f1_history):
copyfile(model_file, os.path.join(opt['model_save_dir'], 'best_model.pt'))
print('new best model saved.')
if epoch % opt['save_epoch'] != 0:
os.remove(model_file)
dev_f1_history += [dev_f1]
print("")
# test_batch = DataLoader(opt['data_dir'] + '/test.csv', opt['batch_size'], opt, True)