update
Browse files
README.md
CHANGED
@@ -61,7 +61,7 @@ docker run -itd \
|
|
61 |
--ipc=host \
|
62 |
-v /data/tianxing/HuggingDatasets/nx_noise/data:/data/tianxing/HuggingDatasets/nx_noise/data \
|
63 |
-v /data/tianxing/PycharmProjects/cc_vad:/data/tianxing/PycharmProjects/cc_vad \
|
64 |
-
python:3.12
|
65 |
|
66 |
|
67 |
查看GPU
|
|
|
61 |
--ipc=host \
|
62 |
-v /data/tianxing/HuggingDatasets/nx_noise/data:/data/tianxing/HuggingDatasets/nx_noise/data \
|
63 |
-v /data/tianxing/PycharmProjects/cc_vad:/data/tianxing/PycharmProjects/cc_vad \
|
64 |
+
python:3.12 /bin/bash
|
65 |
|
66 |
|
67 |
查看GPU
|
examples/silero_vad_by_webrtcvad/run.sh
CHANGED
@@ -2,19 +2,9 @@
|
|
2 |
|
3 |
: <<'END'
|
4 |
|
5 |
-
|
6 |
-
--
|
7 |
-
--
|
8 |
-
|
9 |
-
sh run.sh --stage 2 --stop_stage 2 --system_version centos --file_folder_name file_dir --final_model_name dfnet2-nx-dns3 \
|
10 |
-
--noise_dir "/data/tianxing/HuggingDatasets/nx_noise/data/noise" \
|
11 |
-
--speech_dir "/data/tianxing/HuggingDatasets/nx_noise/data/speech/dns3-speech"
|
12 |
-
|
13 |
-
sh run.sh --stage 2 --stop_stage 2 --system_version centos --file_folder_name file_dir --final_model_name dfnet2-nx2 \
|
14 |
-
--noise_dir "/data/tianxing/HuggingDatasets/nx_noise/data/noise/nx-noise" \
|
15 |
-
--speech_dir "/data/tianxing/HuggingDatasets/nx_noise/data/speech/nx-speech2"
|
16 |
-
|
17 |
-
sh run.sh --stage 2 --stop_stage 2 --system_version centos --file_folder_name dfnet2-nx2-dns3 --final_model_name dfnet2-nx2-dns3 \
|
18 |
--noise_dir "/data/tianxing/HuggingDatasets/nx_noise/data/noise/" \
|
19 |
--speech_dir "/data/tianxing/HuggingDatasets/nx_noise/data/speech/"
|
20 |
|
|
|
2 |
|
3 |
: <<'END'
|
4 |
|
5 |
+
bash run.sh --stage 1 --stop_stage 1 --system_version centos \
|
6 |
+
--file_folder_name silero-vad-by-webrtcvad-nx2-dns3 \
|
7 |
+
--final_model_name silero-vad-by-webrtcvad-nx2-dns3 \
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
8 |
--noise_dir "/data/tianxing/HuggingDatasets/nx_noise/data/noise/" \
|
9 |
--speech_dir "/data/tianxing/HuggingDatasets/nx_noise/data/speech/"
|
10 |
|
examples/silero_vad_by_webrtcvad/step_2_train_model.py
CHANGED
@@ -1,8 +1,5 @@
|
|
1 |
#!/usr/bin/python3
|
2 |
# -*- coding: utf-8 -*-
|
3 |
-
"""
|
4 |
-
https://github.com/Rikorose/DeepFilterNet
|
5 |
-
"""
|
6 |
import argparse
|
7 |
import json
|
8 |
import logging
|
@@ -13,9 +10,7 @@ from pathlib import Path
|
|
13 |
import random
|
14 |
import sys
|
15 |
import shutil
|
16 |
-
from typing import List
|
17 |
-
|
18 |
-
from fontTools.varLib.plot import stops
|
19 |
|
20 |
pwd = os.path.abspath(os.path.dirname(__file__))
|
21 |
sys.path.append(os.path.join(pwd, "../../"))
|
@@ -27,12 +22,13 @@ from torch.nn import functional as F
|
|
27 |
from torch.utils.data.dataloader import DataLoader
|
28 |
from tqdm import tqdm
|
29 |
|
30 |
-
from toolbox.torch.utils.data.dataset.
|
31 |
-
from toolbox.torchaudio.
|
32 |
-
from toolbox.torchaudio.
|
33 |
-
from toolbox.torchaudio.
|
34 |
-
from toolbox.torchaudio.
|
35 |
-
from toolbox.torchaudio.
|
|
|
36 |
|
37 |
|
38 |
def get_args():
|
@@ -76,28 +72,23 @@ class CollateFunction(object):
|
|
76 |
pass
|
77 |
|
78 |
def __call__(self, batch: List[dict]):
|
79 |
-
clean_audios = list()
|
80 |
noisy_audios = list()
|
81 |
-
|
82 |
|
83 |
for sample in batch:
|
84 |
-
|
85 |
-
|
86 |
-
noisy_audio: torch.Tensor = sample["mix_wave"]
|
87 |
-
# snr_db: float = sample["snr_db"]
|
88 |
|
89 |
-
|
90 |
-
|
91 |
|
92 |
-
clean_audios = torch.stack(clean_audios)
|
93 |
noisy_audios = torch.stack(noisy_audios)
|
94 |
|
95 |
# assert
|
96 |
-
if torch.any(torch.isnan(clean_audios)) or torch.any(torch.isinf(clean_audios)):
|
97 |
-
raise AssertionError("nan or inf in clean_audios")
|
98 |
if torch.any(torch.isnan(noisy_audios)) or torch.any(torch.isinf(noisy_audios)):
|
99 |
raise AssertionError("nan or inf in noisy_audios")
|
100 |
-
|
|
|
101 |
|
102 |
|
103 |
collate_fn = CollateFunction()
|
@@ -106,7 +97,7 @@ collate_fn = CollateFunction()
|
|
106 |
def main():
|
107 |
args = get_args()
|
108 |
|
109 |
-
config =
|
110 |
pretrained_model_name_or_path=args.config_file,
|
111 |
)
|
112 |
|
@@ -125,7 +116,7 @@ def main():
|
|
125 |
logger.info(f"GPU available count: {n_gpu}; device: {device}")
|
126 |
|
127 |
# datasets
|
128 |
-
train_dataset =
|
129 |
jsonl_file=args.train_dataset,
|
130 |
expected_sample_rate=config.sample_rate,
|
131 |
max_wave_value=32768.0,
|
@@ -133,7 +124,7 @@ def main():
|
|
133 |
max_snr_db=config.max_snr_db,
|
134 |
# skip=225000,
|
135 |
)
|
136 |
-
valid_dataset =
|
137 |
jsonl_file=args.valid_dataset,
|
138 |
expected_sample_rate=config.sample_rate,
|
139 |
max_wave_value=32768.0,
|
@@ -165,7 +156,7 @@ def main():
|
|
165 |
|
166 |
# models
|
167 |
logger.info(f"prepare models. config_file: {args.config_file}")
|
168 |
-
model =
|
169 |
model.to(device)
|
170 |
model.train()
|
171 |
|
@@ -210,25 +201,17 @@ def main():
|
|
210 |
else:
|
211 |
raise AssertionError(f"invalid lr_scheduler: {config.lr_scheduler}")
|
212 |
|
213 |
-
|
214 |
-
|
215 |
-
|
216 |
-
|
217 |
-
hop_size_list=[128, 256, 512],
|
218 |
-
factor_sc=1.5,
|
219 |
-
factor_mag=1.0,
|
220 |
-
reduction="mean"
|
221 |
-
).to(device)
|
222 |
|
223 |
# training loop
|
224 |
|
225 |
# state
|
226 |
-
average_pesq_score = 1000000000
|
227 |
average_loss = 1000000000
|
228 |
-
|
229 |
-
|
230 |
-
average_mask_loss = 1000000000
|
231 |
-
average_lsnr_loss = 1000000000
|
232 |
|
233 |
model_list = list()
|
234 |
best_epoch_idx = None
|
@@ -246,13 +229,11 @@ def main():
|
|
246 |
|
247 |
# train
|
248 |
model.train()
|
|
|
249 |
|
250 |
-
total_pesq_score = 0.
|
251 |
total_loss = 0.
|
252 |
-
|
253 |
-
|
254 |
-
total_mask_loss = 0.
|
255 |
-
total_lsnr_loss = 0.
|
256 |
total_batches = 0.
|
257 |
|
258 |
progress_bar_train = tqdm(
|
@@ -260,28 +241,24 @@ def main():
|
|
260 |
desc="Training; epoch-{}".format(epoch_idx),
|
261 |
)
|
262 |
for train_batch in train_data_loader:
|
263 |
-
|
264 |
-
clean_audios: torch.Tensor = clean_audios.to(device)
|
265 |
noisy_audios: torch.Tensor = noisy_audios.to(device)
|
|
|
|
|
|
|
|
|
266 |
|
267 |
-
|
268 |
-
# est_wav shape: [b, 1, n_samples]
|
269 |
-
est_wav = torch.squeeze(est_wav, dim=1)
|
270 |
-
# est_wav shape: [b, n_samples]
|
271 |
|
272 |
-
|
273 |
-
|
274 |
-
mask_loss = model.mask_loss_fn(est_mask, clean_audios, noisy_audios)
|
275 |
-
lsnr_loss = model.lsnr_loss_fn(lsnr, clean_audios, noisy_audios)
|
276 |
|
277 |
-
loss = 1.0 *
|
278 |
if torch.any(torch.isnan(loss)) or torch.any(torch.isinf(loss)):
|
279 |
logger.info(f"find nan or inf in loss. continue.")
|
280 |
continue
|
281 |
|
282 |
-
|
283 |
-
clean_audios_list_r = list(clean_audios.detach().cpu().numpy())
|
284 |
-
pesq_score = run_pesq_score(clean_audios_list_r, denoise_audios_list_r, sample_rate=config.sample_rate, mode="nb")
|
285 |
|
286 |
optimizer.zero_grad()
|
287 |
loss.backward()
|
@@ -289,30 +266,25 @@ def main():
|
|
289 |
optimizer.step()
|
290 |
lr_scheduler.step()
|
291 |
|
292 |
-
total_pesq_score += pesq_score
|
293 |
total_loss += loss.item()
|
294 |
-
|
295 |
-
|
296 |
-
total_mask_loss += mask_loss.item()
|
297 |
-
total_lsnr_loss += lsnr_loss.item()
|
298 |
total_batches += 1
|
299 |
|
300 |
-
average_pesq_score = round(total_pesq_score / total_batches, 4)
|
301 |
average_loss = round(total_loss / total_batches, 4)
|
302 |
-
|
303 |
-
|
304 |
-
|
305 |
-
|
|
|
306 |
|
307 |
progress_bar_train.update(1)
|
308 |
progress_bar_train.set_postfix({
|
309 |
"lr": lr_scheduler.get_last_lr()[0],
|
310 |
-
"pesq_score": average_pesq_score,
|
311 |
"loss": average_loss,
|
312 |
-
"
|
313 |
-
"
|
314 |
-
"
|
315 |
-
"lsnr_loss": average_lsnr_loss,
|
316 |
})
|
317 |
|
318 |
# evaluation
|
@@ -322,13 +294,11 @@ def main():
|
|
322 |
torch.cuda.empty_cache()
|
323 |
|
324 |
model.eval()
|
|
|
325 |
|
326 |
-
total_pesq_score = 0.
|
327 |
total_loss = 0.
|
328 |
-
|
329 |
-
|
330 |
-
total_mask_loss = 0.
|
331 |
-
total_lsnr_loss = 0.
|
332 |
total_batches = 0.
|
333 |
|
334 |
progress_bar_train.close()
|
@@ -336,63 +306,52 @@ def main():
|
|
336 |
desc="Evaluation; steps-{}k".format(int(step_idx/1000)),
|
337 |
)
|
338 |
for eval_batch in valid_data_loader:
|
339 |
-
|
340 |
-
clean_audios: torch.Tensor = clean_audios.to(device)
|
341 |
noisy_audios: torch.Tensor = noisy_audios.to(device)
|
|
|
|
|
|
|
|
|
342 |
|
343 |
-
|
344 |
-
# est_wav shape: [b, 1, n_samples]
|
345 |
-
est_wav = torch.squeeze(est_wav, dim=1)
|
346 |
-
# est_wav shape: [b, n_samples]
|
347 |
|
348 |
-
|
349 |
-
|
350 |
-
mask_loss = model.mask_loss_fn(est_mask, clean_audios, noisy_audios)
|
351 |
-
lsnr_loss = model.lsnr_loss_fn(lsnr, clean_audios, noisy_audios)
|
352 |
|
353 |
-
loss = 1.0 *
|
354 |
if torch.any(torch.isnan(loss)) or torch.any(torch.isinf(loss)):
|
355 |
logger.info(f"find nan or inf in loss. continue.")
|
356 |
continue
|
357 |
|
358 |
-
|
359 |
-
clean_audios_list_r = list(clean_audios.detach().cpu().numpy())
|
360 |
-
pesq_score = run_pesq_score(clean_audios_list_r, denoise_audios_list_r, sample_rate=config.sample_rate, mode="nb")
|
361 |
|
362 |
-
total_pesq_score += pesq_score
|
363 |
total_loss += loss.item()
|
364 |
-
|
365 |
-
|
366 |
-
total_mask_loss += mask_loss.item()
|
367 |
-
total_lsnr_loss += lsnr_loss.item()
|
368 |
total_batches += 1
|
369 |
|
370 |
-
average_pesq_score = round(total_pesq_score / total_batches, 4)
|
371 |
average_loss = round(total_loss / total_batches, 4)
|
372 |
-
|
373 |
-
|
374 |
-
|
375 |
-
|
|
|
376 |
|
377 |
progress_bar_eval.update(1)
|
378 |
progress_bar_eval.set_postfix({
|
379 |
"lr": lr_scheduler.get_last_lr()[0],
|
380 |
-
"pesq_score": average_pesq_score,
|
381 |
"loss": average_loss,
|
382 |
-
"
|
383 |
-
"
|
384 |
-
"
|
385 |
-
"lsnr_loss": average_lsnr_loss,
|
386 |
})
|
387 |
|
388 |
model.train()
|
|
|
389 |
|
390 |
-
total_pesq_score = 0.
|
391 |
total_loss = 0.
|
392 |
-
|
393 |
-
|
394 |
-
total_mask_loss = 0.
|
395 |
-
total_lsnr_loss = 0.
|
396 |
total_batches = 0.
|
397 |
|
398 |
progress_bar_eval.close()
|
@@ -418,12 +377,12 @@ def main():
|
|
418 |
if best_metric is None:
|
419 |
best_epoch_idx = epoch_idx
|
420 |
best_step_idx = step_idx
|
421 |
-
best_metric =
|
422 |
-
elif
|
423 |
# great is better.
|
424 |
best_epoch_idx = epoch_idx
|
425 |
best_step_idx = step_idx
|
426 |
-
best_metric =
|
427 |
else:
|
428 |
pass
|
429 |
|
@@ -431,12 +390,11 @@ def main():
|
|
431 |
"epoch_idx": epoch_idx,
|
432 |
"best_epoch_idx": best_epoch_idx,
|
433 |
"best_step_idx": best_step_idx,
|
434 |
-
"pesq_score": average_pesq_score,
|
435 |
"loss": average_loss,
|
436 |
-
"
|
437 |
-
"
|
438 |
-
|
439 |
-
"
|
440 |
}
|
441 |
metrics_filename = save_dir / "metrics_epoch.json"
|
442 |
with open(metrics_filename, "w", encoding="utf-8") as f:
|
|
|
1 |
#!/usr/bin/python3
|
2 |
# -*- coding: utf-8 -*-
|
|
|
|
|
|
|
3 |
import argparse
|
4 |
import json
|
5 |
import logging
|
|
|
10 |
import random
|
11 |
import sys
|
12 |
import shutil
|
13 |
+
from typing import List, Tuple
|
|
|
|
|
14 |
|
15 |
pwd = os.path.abspath(os.path.dirname(__file__))
|
16 |
sys.path.append(os.path.join(pwd, "../../"))
|
|
|
22 |
from torch.utils.data.dataloader import DataLoader
|
23 |
from tqdm import tqdm
|
24 |
|
25 |
+
from toolbox.torch.utils.data.dataset.vad_jsonl_dataset import VadJsonlDataset
|
26 |
+
from toolbox.torchaudio.models.vad.silero_vad.configuration_silero_vad import SileroVadConfig
|
27 |
+
from toolbox.torchaudio.models.vad.silero_vad.modeling_silero_vad import SileroVadModel, SileroVadPretrainedModel
|
28 |
+
from toolbox.torchaudio.losses.vad_loss.base_vad_loss import BaseVadLoss
|
29 |
+
from toolbox.torchaudio.losses.bce_loss import BCELoss
|
30 |
+
from toolbox.torchaudio.losses.dice_loss import DiceLoss
|
31 |
+
from toolbox.torchaudio.metrics.vad_metrics.vad_accuracy import VadAccuracy
|
32 |
|
33 |
|
34 |
def get_args():
|
|
|
72 |
pass
|
73 |
|
74 |
def __call__(self, batch: List[dict]):
|
|
|
75 |
noisy_audios = list()
|
76 |
+
batch_vad_segments = list()
|
77 |
|
78 |
for sample in batch:
|
79 |
+
noisy_wave: torch.Tensor = sample["noisy_wave"]
|
80 |
+
vad_segments: List[Tuple[float, float]] = sample["vad_segments"]
|
|
|
|
|
81 |
|
82 |
+
noisy_audios.append(noisy_wave)
|
83 |
+
batch_vad_segments.append(vad_segments)
|
84 |
|
|
|
85 |
noisy_audios = torch.stack(noisy_audios)
|
86 |
|
87 |
# assert
|
|
|
|
|
88 |
if torch.any(torch.isnan(noisy_audios)) or torch.any(torch.isinf(noisy_audios)):
|
89 |
raise AssertionError("nan or inf in noisy_audios")
|
90 |
+
|
91 |
+
return noisy_audios, batch_vad_segments
|
92 |
|
93 |
|
94 |
collate_fn = CollateFunction()
|
|
|
97 |
def main():
|
98 |
args = get_args()
|
99 |
|
100 |
+
config = SileroVadConfig.from_pretrained(
|
101 |
pretrained_model_name_or_path=args.config_file,
|
102 |
)
|
103 |
|
|
|
116 |
logger.info(f"GPU available count: {n_gpu}; device: {device}")
|
117 |
|
118 |
# datasets
|
119 |
+
train_dataset = VadJsonlDataset(
|
120 |
jsonl_file=args.train_dataset,
|
121 |
expected_sample_rate=config.sample_rate,
|
122 |
max_wave_value=32768.0,
|
|
|
124 |
max_snr_db=config.max_snr_db,
|
125 |
# skip=225000,
|
126 |
)
|
127 |
+
valid_dataset = VadJsonlDataset(
|
128 |
jsonl_file=args.valid_dataset,
|
129 |
expected_sample_rate=config.sample_rate,
|
130 |
max_wave_value=32768.0,
|
|
|
156 |
|
157 |
# models
|
158 |
logger.info(f"prepare models. config_file: {args.config_file}")
|
159 |
+
model = SileroVadPretrainedModel(config).to(device)
|
160 |
model.to(device)
|
161 |
model.train()
|
162 |
|
|
|
201 |
else:
|
202 |
raise AssertionError(f"invalid lr_scheduler: {config.lr_scheduler}")
|
203 |
|
204 |
+
bce_loss_fn = BCELoss(reduction="mean").to(device)
|
205 |
+
dice_loss_fn = DiceLoss(reduction="mean").to(device)
|
206 |
+
|
207 |
+
vad_accuracy_metrics_fn = VadAccuracy(threshold=0.5)
|
|
|
|
|
|
|
|
|
|
|
208 |
|
209 |
# training loop
|
210 |
|
211 |
# state
|
|
|
212 |
average_loss = 1000000000
|
213 |
+
average_bce_loss = 1000000000
|
214 |
+
average_dice_loss = 1000000000
|
|
|
|
|
215 |
|
216 |
model_list = list()
|
217 |
best_epoch_idx = None
|
|
|
229 |
|
230 |
# train
|
231 |
model.train()
|
232 |
+
vad_accuracy_metrics_fn.reset()
|
233 |
|
|
|
234 |
total_loss = 0.
|
235 |
+
total_bce_loss = 0.
|
236 |
+
total_dice_loss = 0.
|
|
|
|
|
237 |
total_batches = 0.
|
238 |
|
239 |
progress_bar_train = tqdm(
|
|
|
241 |
desc="Training; epoch-{}".format(epoch_idx),
|
242 |
)
|
243 |
for train_batch in train_data_loader:
|
244 |
+
noisy_audios, batch_vad_segments = train_batch
|
|
|
245 |
noisy_audios: torch.Tensor = noisy_audios.to(device)
|
246 |
+
# noisy_audios shape: [b, num_samples]
|
247 |
+
num_samples = noisy_audios.shape[-1]
|
248 |
+
|
249 |
+
predictions = model.forward(noisy_audios)
|
250 |
|
251 |
+
targets = BaseVadLoss.get_targets(predictions, batch_vad_segments, duration=num_samples / config.sample_rate)
|
|
|
|
|
|
|
252 |
|
253 |
+
bce_loss = bce_loss_fn.forward(predictions, targets)
|
254 |
+
dice_loss = dice_loss_fn.forward(predictions, targets)
|
|
|
|
|
255 |
|
256 |
+
loss = 1.0 * bce_loss + 1.0 * dice_loss
|
257 |
if torch.any(torch.isnan(loss)) or torch.any(torch.isinf(loss)):
|
258 |
logger.info(f"find nan or inf in loss. continue.")
|
259 |
continue
|
260 |
|
261 |
+
vad_accuracy_metrics_fn.__call__(predictions, targets)
|
|
|
|
|
262 |
|
263 |
optimizer.zero_grad()
|
264 |
loss.backward()
|
|
|
266 |
optimizer.step()
|
267 |
lr_scheduler.step()
|
268 |
|
|
|
269 |
total_loss += loss.item()
|
270 |
+
total_bce_loss += bce_loss.item()
|
271 |
+
total_dice_loss += dice_loss.item()
|
|
|
|
|
272 |
total_batches += 1
|
273 |
|
|
|
274 |
average_loss = round(total_loss / total_batches, 4)
|
275 |
+
average_bce_loss = round(total_bce_loss / total_batches, 4)
|
276 |
+
average_dice_loss = round(total_dice_loss / total_batches, 4)
|
277 |
+
|
278 |
+
metrics = vad_accuracy_metrics_fn.get_metric()
|
279 |
+
accuracy = metrics["accuracy"]
|
280 |
|
281 |
progress_bar_train.update(1)
|
282 |
progress_bar_train.set_postfix({
|
283 |
"lr": lr_scheduler.get_last_lr()[0],
|
|
|
284 |
"loss": average_loss,
|
285 |
+
"average_bce_loss": average_bce_loss,
|
286 |
+
"average_dice_loss": average_dice_loss,
|
287 |
+
"accuracy": accuracy,
|
|
|
288 |
})
|
289 |
|
290 |
# evaluation
|
|
|
294 |
torch.cuda.empty_cache()
|
295 |
|
296 |
model.eval()
|
297 |
+
vad_accuracy_metrics_fn.reset()
|
298 |
|
|
|
299 |
total_loss = 0.
|
300 |
+
total_bce_loss = 0.
|
301 |
+
total_dice_loss = 0.
|
|
|
|
|
302 |
total_batches = 0.
|
303 |
|
304 |
progress_bar_train.close()
|
|
|
306 |
desc="Evaluation; steps-{}k".format(int(step_idx/1000)),
|
307 |
)
|
308 |
for eval_batch in valid_data_loader:
|
309 |
+
noisy_audios, batch_vad_segments = train_batch
|
|
|
310 |
noisy_audios: torch.Tensor = noisy_audios.to(device)
|
311 |
+
# noisy_audios shape: [b, num_samples]
|
312 |
+
num_samples = noisy_audios.shape[-1]
|
313 |
+
|
314 |
+
predictions = model.forward(noisy_audios)
|
315 |
|
316 |
+
targets = BaseVadLoss.get_targets(predictions, batch_vad_segments, duration=num_samples / config.sample_rate)
|
|
|
|
|
|
|
317 |
|
318 |
+
bce_loss = bce_loss_fn.forward(predictions, targets)
|
319 |
+
dice_loss = dice_loss_fn.forward(predictions, targets)
|
|
|
|
|
320 |
|
321 |
+
loss = 1.0 * bce_loss + 1.0 * dice_loss
|
322 |
if torch.any(torch.isnan(loss)) or torch.any(torch.isinf(loss)):
|
323 |
logger.info(f"find nan or inf in loss. continue.")
|
324 |
continue
|
325 |
|
326 |
+
vad_accuracy_metrics_fn.__call__(predictions, targets)
|
|
|
|
|
327 |
|
|
|
328 |
total_loss += loss.item()
|
329 |
+
total_bce_loss += bce_loss.item()
|
330 |
+
total_dice_loss += dice_loss.item()
|
|
|
|
|
331 |
total_batches += 1
|
332 |
|
|
|
333 |
average_loss = round(total_loss / total_batches, 4)
|
334 |
+
average_bce_loss = round(total_bce_loss / total_batches, 4)
|
335 |
+
average_dice_loss = round(total_dice_loss / total_batches, 4)
|
336 |
+
|
337 |
+
metrics = vad_accuracy_metrics_fn.get_metric()
|
338 |
+
accuracy = metrics["accuracy"]
|
339 |
|
340 |
progress_bar_eval.update(1)
|
341 |
progress_bar_eval.set_postfix({
|
342 |
"lr": lr_scheduler.get_last_lr()[0],
|
|
|
343 |
"loss": average_loss,
|
344 |
+
"average_bce_loss": average_bce_loss,
|
345 |
+
"average_dice_loss": average_dice_loss,
|
346 |
+
"accuracy": accuracy,
|
|
|
347 |
})
|
348 |
|
349 |
model.train()
|
350 |
+
vad_accuracy_metrics_fn.reset()
|
351 |
|
|
|
352 |
total_loss = 0.
|
353 |
+
total_bce_loss = 0.
|
354 |
+
total_dice_loss = 0.
|
|
|
|
|
355 |
total_batches = 0.
|
356 |
|
357 |
progress_bar_eval.close()
|
|
|
377 |
if best_metric is None:
|
378 |
best_epoch_idx = epoch_idx
|
379 |
best_step_idx = step_idx
|
380 |
+
best_metric = accuracy
|
381 |
+
elif accuracy >= best_metric:
|
382 |
# great is better.
|
383 |
best_epoch_idx = epoch_idx
|
384 |
best_step_idx = step_idx
|
385 |
+
best_metric = accuracy
|
386 |
else:
|
387 |
pass
|
388 |
|
|
|
390 |
"epoch_idx": epoch_idx,
|
391 |
"best_epoch_idx": best_epoch_idx,
|
392 |
"best_step_idx": best_step_idx,
|
|
|
393 |
"loss": average_loss,
|
394 |
+
"bce_loss": average_bce_loss,
|
395 |
+
"dice_loss": average_dice_loss,
|
396 |
+
|
397 |
+
"accuracy": accuracy,
|
398 |
}
|
399 |
metrics_filename = save_dir / "metrics_epoch.json"
|
400 |
with open(metrics_filename, "w", encoding="utf-8") as f:
|
toolbox/torchaudio/losses/bce_loss.py
ADDED
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
#!/usr/bin/python3
|
2 |
+
# -*- coding: utf-8 -*-
|
3 |
+
from typing import List, Tuple
|
4 |
+
|
5 |
+
import torch
|
6 |
+
import torch.nn as nn
|
7 |
+
|
8 |
+
from toolbox.torchaudio.losses.vad_loss.base_vad_loss import BaseVadLoss
|
9 |
+
|
10 |
+
|
11 |
+
class BCELoss(BaseVadLoss):
|
12 |
+
"""
|
13 |
+
Binary Cross-Entropy Loss, BCE Loss
|
14 |
+
"""
|
15 |
+
def __init__(self,
|
16 |
+
reduction: str = "mean",
|
17 |
+
):
|
18 |
+
super(BCELoss, self).__init__()
|
19 |
+
self.reduction = reduction
|
20 |
+
|
21 |
+
self.bce_loss_fn = nn.BCELoss(reduction=reduction)
|
22 |
+
|
23 |
+
def forward(self, inputs: torch.Tensor, targets: torch.Tensor):
|
24 |
+
"""
|
25 |
+
:param inputs: torch.Tensor, shape: [b, t, 1]. vad prob, after sigmoid activation.
|
26 |
+
:param targets: shape as `inputs`.
|
27 |
+
:return:
|
28 |
+
"""
|
29 |
+
loss = self.bce_loss_fn.forward(inputs, targets)
|
30 |
+
return loss
|
31 |
+
|
32 |
+
|
33 |
+
def main():
|
34 |
+
inputs = torch.zeros(size=(1, 198, 1), dtype=torch.float32)
|
35 |
+
|
36 |
+
loss_fn = BCELoss()
|
37 |
+
|
38 |
+
loss = loss_fn.forward(inputs, inputs)
|
39 |
+
print(loss)
|
40 |
+
return
|
41 |
+
|
42 |
+
|
43 |
+
if __name__ == "__main__":
|
44 |
+
main()
|
toolbox/torchaudio/losses/dice_loss.py
ADDED
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
#!/usr/bin/python3
|
2 |
+
# -*- coding: utf-8 -*-
|
3 |
+
from typing import List, Tuple
|
4 |
+
|
5 |
+
import torch
|
6 |
+
import torch.nn as nn
|
7 |
+
|
8 |
+
|
9 |
+
class DiceLoss(nn.Module):
|
10 |
+
def __init__(self,
|
11 |
+
reduction: str = "mean",
|
12 |
+
eps: float = 1e-6,
|
13 |
+
):
|
14 |
+
super(DiceLoss, self).__init__()
|
15 |
+
self.reduction = reduction
|
16 |
+
self.eps = eps
|
17 |
+
|
18 |
+
if reduction not in ("sum", "mean"):
|
19 |
+
raise AssertionError(f"param reduction must be sum or mean.")
|
20 |
+
|
21 |
+
def forward(self, inputs: torch.Tensor, targets: torch.Tensor):
|
22 |
+
"""
|
23 |
+
:param inputs: torch.Tensor, shape: [b, t, 1]. vad prob, after sigmoid activation.
|
24 |
+
:param targets: shape as `inputs`.
|
25 |
+
:return:
|
26 |
+
"""
|
27 |
+
inputs_ = torch.squeeze(inputs, dim=-1)
|
28 |
+
targets_ = torch.squeeze(targets, dim=-1)
|
29 |
+
# shape: [b, t]
|
30 |
+
|
31 |
+
intersection = (inputs_ * targets_).sum(dim=-1)
|
32 |
+
union = (inputs_ + targets_).sum(dim=-1)
|
33 |
+
# shape: [b,]
|
34 |
+
|
35 |
+
dice = (2. * intersection + self.eps) / (union + self.eps)
|
36 |
+
# shape: [b,]
|
37 |
+
|
38 |
+
loss = 1. - dice
|
39 |
+
# shape: [b,]
|
40 |
+
|
41 |
+
if self.reduction == "mean":
|
42 |
+
loss = torch.mean(loss)
|
43 |
+
elif self.reduction == "sum":
|
44 |
+
loss = torch.sum(loss)
|
45 |
+
else:
|
46 |
+
raise AssertionError
|
47 |
+
return loss
|
48 |
+
|
49 |
+
|
50 |
+
def main():
|
51 |
+
inputs = torch.zeros(size=(1, 198, 1), dtype=torch.float32)
|
52 |
+
|
53 |
+
loss_fn = DiceLoss()
|
54 |
+
|
55 |
+
loss = loss_fn.forward(inputs, inputs)
|
56 |
+
print(loss)
|
57 |
+
return
|
58 |
+
|
59 |
+
|
60 |
+
if __name__ == "__main__":
|
61 |
+
main()
|
toolbox/torchaudio/models/vad/silero_vad/modeling_silero_vad.py
CHANGED
@@ -8,9 +8,13 @@ https://github.com/snakers4/silero-vad
|
|
8 |
|
9 |
https://github.com/snakers4/silero-vad/blob/master/src/silero_vad/data/silero_vad.jit
|
10 |
"""
|
|
|
|
|
|
|
11 |
import torch
|
12 |
import torch.nn as nn
|
13 |
|
|
|
14 |
from toolbox.torchaudio.models.vad.silero_vad.configuration_silero_vad import SileroVadConfig
|
15 |
from toolbox.torchaudio.modules.conv_stft import ConvSTFT
|
16 |
|
@@ -134,6 +138,52 @@ class SileroVadModel(nn.Module):
|
|
134 |
return x
|
135 |
|
136 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
137 |
def main():
|
138 |
config = SileroVadConfig()
|
139 |
model = SileroVadModel(config=config)
|
|
|
8 |
|
9 |
https://github.com/snakers4/silero-vad/blob/master/src/silero_vad/data/silero_vad.jit
|
10 |
"""
|
11 |
+
import os
|
12 |
+
from typing import Optional, Union
|
13 |
+
|
14 |
import torch
|
15 |
import torch.nn as nn
|
16 |
|
17 |
+
from toolbox.torchaudio.configuration_utils import CONFIG_FILE
|
18 |
from toolbox.torchaudio.models.vad.silero_vad.configuration_silero_vad import SileroVadConfig
|
19 |
from toolbox.torchaudio.modules.conv_stft import ConvSTFT
|
20 |
|
|
|
138 |
return x
|
139 |
|
140 |
|
141 |
+
class SileroVadPretrainedModel(SileroVadModel):
|
142 |
+
def __init__(self,
|
143 |
+
config: SileroVadConfig,
|
144 |
+
):
|
145 |
+
super(SileroVadPretrainedModel, self).__init__(
|
146 |
+
config=config,
|
147 |
+
)
|
148 |
+
|
149 |
+
@classmethod
|
150 |
+
def from_pretrained(cls, pretrained_model_name_or_path, **kwargs):
|
151 |
+
config = SileroVadConfig.from_pretrained(pretrained_model_name_or_path, **kwargs)
|
152 |
+
|
153 |
+
model = cls(config)
|
154 |
+
|
155 |
+
if os.path.isdir(pretrained_model_name_or_path):
|
156 |
+
ckpt_file = os.path.join(pretrained_model_name_or_path, MODEL_FILE)
|
157 |
+
else:
|
158 |
+
ckpt_file = pretrained_model_name_or_path
|
159 |
+
|
160 |
+
with open(ckpt_file, "rb") as f:
|
161 |
+
state_dict = torch.load(f, map_location="cpu", weights_only=True)
|
162 |
+
model.load_state_dict(state_dict, strict=True)
|
163 |
+
return model
|
164 |
+
|
165 |
+
def save_pretrained(self,
|
166 |
+
save_directory: Union[str, os.PathLike],
|
167 |
+
state_dict: Optional[dict] = None,
|
168 |
+
):
|
169 |
+
|
170 |
+
model = self
|
171 |
+
|
172 |
+
if state_dict is None:
|
173 |
+
state_dict = model.state_dict()
|
174 |
+
|
175 |
+
os.makedirs(save_directory, exist_ok=True)
|
176 |
+
|
177 |
+
# save state dict
|
178 |
+
model_file = os.path.join(save_directory, MODEL_FILE)
|
179 |
+
torch.save(state_dict, model_file)
|
180 |
+
|
181 |
+
# save config
|
182 |
+
config_file = os.path.join(save_directory, CONFIG_FILE)
|
183 |
+
self.config.to_yaml_file(config_file)
|
184 |
+
return save_directory
|
185 |
+
|
186 |
+
|
187 |
def main():
|
188 |
config = SileroVadConfig()
|
189 |
model = SileroVadModel(config=config)
|