-
Notifications
You must be signed in to change notification settings - Fork 1
/
train_coral.py
135 lines (111 loc) · 6.55 KB
/
train_coral.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
import os
import torch
from loss_and_metrics import cov
from collections import defaultdict
from ae import AE
from evaluation_utils import model_save_check, evaluate_target_regression_epoch
from loss_and_metrics import masked_simse,masked_mse
from multi_out_mlp import MoMLP
from encoder_decoder import EncoderDecoder
def coral_train_step(model, s_batch, t_batch, device, optimizer, alpha, history, scheduler=None):
model.zero_grad()
model.train()
s_x = s_batch[0].to(device)
s_y = s_batch[1].to(device)
t_x = t_batch[0].to(device)
t_y = t_batch[1].to(device)
s_code = model.encode(s_x)
t_code = model.encode(t_x)
s_cov = cov(s_code)
t_cov = cov(t_code)
coral_loss = (torch.square(torch.norm(s_cov - t_cov, p='fro'))) / (4 * (s_code.size()[-1] ** 2))
# coral_loss = torch.square(torch.norm(s_cov-t_cov, p='fro'))
loss = masked_mse(preds=model(s_x), labels=s_y) + masked_mse(preds=model(t_x), labels=t_y) + alpha * coral_loss
optimizer.zero_grad()
loss.backward()
optimizer.step()
if scheduler is not None:
scheduler.step()
history['loss'].append(loss.cpu().detach().item())
history['coral_loss'].append(coral_loss.cpu().detach().item())
return history
def train_coral(s_dataloaders, t_dataloaders, val_dataloader, test_dataloader, metric_name, seed, **kwargs):
"""
:param s_dataloaders:
:param t_dataloaders:
:param kwargs:
:return:
"""
s_train_dataloader = s_dataloaders
t_train_dataloader = t_dataloaders
autoencoder = AE(input_dim=kwargs['input_dim'],
latent_dim=kwargs['latent_dim'],
hidden_dims=kwargs['encoder_hidden_dims'],
dop=kwargs['dop']).to(kwargs['device'])
encoder = autoencoder.encoder
target_decoder = MoMLP(input_dim=kwargs['latent_dim'],
output_dim=kwargs['output_dim'],
hidden_dims=kwargs['regressor_hidden_dims'],
out_fn=torch.nn.Sigmoid).to(kwargs['device'])
target_regressor = EncoderDecoder(encoder=encoder,
decoder=target_decoder).to(kwargs['device'])
train_history = defaultdict(list)
# ae_eval_train_history = defaultdict(list)
val_history = defaultdict(list)
s_target_regression_eval_train_history = defaultdict(list)
t_target_regression_eval_train_history = defaultdict(list)
target_regression_eval_val_history = defaultdict(list)
target_regression_eval_test_history = defaultdict(list)
model_optimizer = torch.optim.AdamW(target_regressor.parameters(), lr=kwargs['lr'])
for epoch in range(int(kwargs['train_num_epochs'])):
if epoch % 50 == 0:
print(f'Coral training epoch {epoch}')
for step, s_batch in enumerate(s_train_dataloader):
t_batch = next(iter(t_train_dataloader))
train_history = coral_train_step(model=target_regressor,
s_batch=s_batch,
t_batch=t_batch,
device=kwargs['device'],
optimizer=model_optimizer,
alpha=kwargs['alpha'],
history=train_history)
s_target_regression_eval_train_history = evaluate_target_regression_epoch(regressor=target_regressor,
dataloader=s_train_dataloader,
device=kwargs['device'],
history=s_target_regression_eval_train_history)
t_target_regression_eval_train_history = evaluate_target_regression_epoch(regressor=target_regressor,
dataloader=t_train_dataloader,
device=kwargs['device'],
history=t_target_regression_eval_train_history)
target_regression_eval_val_history = evaluate_target_regression_epoch(regressor=target_regressor,
dataloader=val_dataloader,
device=kwargs['device'],
history=target_regression_eval_val_history)
target_regression_eval_test_history = evaluate_target_regression_epoch(regressor=target_regressor,
dataloader=test_dataloader,
device=kwargs['device'],
history=target_regression_eval_test_history)
save_flag, stop_flag = model_save_check(history=target_regression_eval_val_history,
metric_name=metric_name,
tolerance_count=50)
if save_flag:
torch.save(target_regressor.state_dict(), os.path.join(kwargs['model_save_folder'], f'coral_regressor_{seed}.pt'))
if stop_flag:
break
target_regressor.load_state_dict(
torch.load(os.path.join(kwargs['model_save_folder'], f'coral_regressor_{seed}.pt')))
# evaluate_target_regression_epoch(regressor=target_regressor,
# dataloader=val_dataloader,
# device=kwargs['device'],
# history=None,
# seed=seed,
# output_folder=kwargs['model_save_folder'])
evaluate_target_regression_epoch(regressor=target_regressor,
dataloader=test_dataloader,
device=kwargs['device'],
history=None,
seed=seed,
output_folder=kwargs['model_save_folder'])
return target_regressor, (
train_history, s_target_regression_eval_train_history, t_target_regression_eval_train_history,
target_regression_eval_val_history, target_regression_eval_test_history)