-
Notifications
You must be signed in to change notification settings - Fork 17
Expand file tree
/
Copy pathtrain.py
More file actions
executable file
·257 lines (200 loc) · 7.83 KB
/
train.py
File metadata and controls
executable file
·257 lines (200 loc) · 7.83 KB
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
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
#!/usr/bin/env python3
"""Train S2VC model."""
import argparse
import datetime
import random
from pathlib import Path
import torch
import torch.nn as nn
from torch.optim import AdamW
from torch.utils.data import DataLoader, random_split
from torch.utils.tensorboard import SummaryWriter
from tqdm import tqdm
import numpy as np
from data import IntraSpeakerDataset, collate_batch, plot_attn
from models import S2VC, get_cosine_schedule_with_warmup
random.seed(42)
torch.manual_seed(42)
torch.cuda.manual_seed(42)
torch.cuda.manual_seed_all(42)
np.random.seed(42)
def parse_args():
"""Parse command-line arguments."""
parser = argparse.ArgumentParser()
parser.add_argument("data_dir", type=str)
parser.add_argument("--save_dir", type=str, default=".")
parser.add_argument("--total_steps", type=int, default=250000)
parser.add_argument("--warmup_steps", type=int, default=100)
parser.add_argument("--valid_steps", type=int, default=1000)
parser.add_argument("--log_steps", type=int, default=100)
parser.add_argument("--save_steps", type=int, default=10000)
parser.add_argument("--n_samples", type=int, default=10)
parser.add_argument("--accu_steps", type=int, default=2)
parser.add_argument("--batch_size", type=int, default=6)
parser.add_argument("--n_workers", type=int, default=8)
parser.add_argument('-s', "--src_feat", type=str, default='cpc')
parser.add_argument('-r', "--ref_feat", type=str, default='cpc')
parser.add_argument("--preload", action="store_true")
parser.add_argument("--lr_reduction", action="store_true")
parser.add_argument("--comment", type=str)
return vars(parser.parse_args())
def model_fn(batch, model, criterion, device):
"""Forward a batch through model."""
srcs, src_masks, tgts, tgt_masks, tgt_mels, overlap_lens = batch
srcs = srcs.to(device)
src_masks = src_masks.to(device)
tgts = tgts.to(device)
tgt_masks = tgt_masks.to(device)
tgt_mels = tgt_mels.to(device)
refs = tgts
ref_masks = tgt_masks
outs, attns = model(srcs, refs, src_masks=src_masks, ref_masks=ref_masks)
losses = []
for out, tgt_mel, attn, overlap_len in zip(outs.unbind(), tgt_mels.unbind(), attns[-1], overlap_lens):
loss = criterion(out[:, :overlap_len], tgt_mel[:, :overlap_len])
losses.append(loss)
try:
attns_plot = []
for i in range(len(attns)):
attns_plot.append(attns[i][0][:overlap_lens[0], :overlap_lens[0]])
except:
pass
return sum(losses) / len(losses), attns_plot
def valid(dataloader, model, criterion, device):
"""Validate on validation set."""
model.eval()
running_loss = 0.0
pbar = tqdm(total=len(dataloader.dataset), ncols=0, desc="Valid", unit=" uttr")
for i, batch in enumerate(dataloader):
with torch.no_grad():
loss, attns = model_fn(batch, model, criterion, device)
running_loss += loss.item()
pbar.update(dataloader.batch_size)
pbar.set_postfix(loss=f"{running_loss / (i+1):.2f}")
pbar.close()
model.train()
return running_loss / len(dataloader), attns
def main(
data_dir,
save_dir,
total_steps,
warmup_steps,
valid_steps,
log_steps,
save_steps,
n_samples,
accu_steps,
batch_size,
n_workers,
src_feat,
ref_feat,
preload,
lr_reduction,
comment,
):
"""Main function."""
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
metadata_path = Path(data_dir) / "metadata.json"
dataset = IntraSpeakerDataset(
data_dir, metadata_path, src_feat, ref_feat, n_samples, preload
)
input_dim, ref_dim, tgt_dim = dataset.get_feat_dim()
lengths = [trainlen := int(0.9 * len(dataset)), len(dataset) - trainlen]
trainset, validset = random_split(dataset, lengths)
print(f'Input dim: {input_dim}, Reference dim: {ref_dim}, Target dim: {tgt_dim}')
model = S2VC(input_dim, ref_dim).to(device)
model = torch.jit.script(model)
train_loader = DataLoader(
trainset,
batch_size=batch_size,
shuffle=True,
drop_last=True,
num_workers=n_workers,
pin_memory=True,
collate_fn=collate_batch,
)
valid_loader = DataLoader(
validset,
batch_size=batch_size * accu_steps,
num_workers=n_workers,
drop_last=True,
pin_memory=True,
# shuffle to make the plot on tensorboard differenct
shuffle=True,
collate_fn=collate_batch,
)
train_iterator = iter(train_loader)
if comment is not None:
log_dir = "logs/"
log_dir += datetime.datetime.now().strftime("%Y-%m-%d_%H:%M:%S")
log_dir += "_" + comment
writer = SummaryWriter(log_dir)
save_dir_path = Path(save_dir)
save_dir_path.mkdir(parents=True, exist_ok=True)
learning_rate = 5e-5
criterion = nn.L1Loss()
optimizer = AdamW(model.parameters(), lr=learning_rate)
scheduler = get_cosine_schedule_with_warmup(optimizer, warmup_steps, total_steps)
best_loss = float("inf")
best_state_dict = None
pbar = tqdm(total=valid_steps, ncols=0, desc="Train", unit=" step")
for step in range(total_steps):
if step == 40002:
file = open('completed.txt', 'a')
print(f'{comment} completed', file=file)
break
batch_loss = 0.0
for _ in range(accu_steps):
try:
batch = next(train_iterator)
except StopIteration:
train_iterator = iter(train_loader)
batch = next(train_iterator)
loss, attns = model_fn(batch, model, criterion, device)
loss = loss / accu_steps
batch_loss += loss.item()
loss.backward()
optimizer.step()
scheduler.step()
optimizer.zero_grad()
pbar.update()
pbar.set_postfix(loss=f"{batch_loss:.2f}", step=step + 1)
if step % log_steps == 0 and comment is not None:
writer.add_scalar("Loss/train", batch_loss, step)
try:
attn = [attns[i].unsqueeze(0) for i in range(len(attns))]
figure = plot_attn(attn, save=False)
writer.add_figure(f"Image/Train-Attentions.png", figure, step + 1)
except:
pass
if (step + 1) % valid_steps == 0:
pbar.close()
valid_loss, attns = valid(valid_loader, model, criterion, device)
if comment is not None:
writer.add_scalar("Loss/valid", valid_loss, step + 1)
try:
attn = [attns[i].unsqueeze(0) for i in range(len(attns))]
figure = plot_attn(attn, save=False)
writer.add_figure(f"Image/Valid-Attentions.png", figure, step + 1)
except:
pass
if valid_loss < best_loss:
best_loss = valid_loss
best_state_dict = model.state_dict()
pbar = tqdm(total=valid_steps, ncols=0, desc="Train", unit=" step")
if (step + 1) % save_steps == 0 and best_state_dict is not None:
loss_str = f"{best_loss:.4f}".replace(".", "dot")
best_ckpt_name = f"retriever-best-loss{loss_str}.pt"
loss_str = f"{valid_loss:.4f}".replace(".", "dot")
curr_ckpt_name = f"retriever-step{step+1}-loss{loss_str}.pt"
current_state_dict = model.state_dict()
model.cpu()
model.load_state_dict(best_state_dict)
model.save(str(save_dir_path / best_ckpt_name))
model.load_state_dict(current_state_dict)
model.save(str(save_dir_path / curr_ckpt_name))
model.to(device)
pbar.write(f"Step {step + 1}, best model saved. (loss={best_loss:.4f})")
pbar.close()
if __name__ == "__main__":
main(**parse_args())