GPT-SoVITS-V2-NIIMI_SORA / GPT_SoVITS /AR /models /t2s_lightning_module_onnx.py
AkitoP's picture
Upload 182 files
e82212c verified
# modified from https://github.com/yangdongchao/SoundStorm/blob/master/soundstorm/s1/AR/models/t2s_lightning_module.py
# reference: https://github.com/lifeiteng/vall-e
import os, sys
now_dir = os.getcwd()
sys.path.append(now_dir)
from typing import Dict
import torch
from pytorch_lightning import LightningModule
from AR.models.t2s_model_onnx import Text2SemanticDecoder
from AR.modules.lr_schedulers import WarmupCosineLRSchedule
from AR.modules.optim import ScaledAdam
class Text2SemanticLightningModule(LightningModule):
def __init__(self, config, output_dir, is_train=True):
super().__init__()
self.config = config
self.top_k = 3
self.model = Text2SemanticDecoder(config=config, top_k=self.top_k)
pretrained_s1 = config.get("pretrained_s1")
if pretrained_s1 and is_train:
# print(self.load_state_dict(torch.load(pretrained_s1,map_location="cpu")["state_dict"]))
print(
self.load_state_dict(
torch.load(pretrained_s1, map_location="cpu")["weight"]
)
)
if is_train:
self.automatic_optimization = False
self.save_hyperparameters()
self.eval_dir = output_dir / "eval"
self.eval_dir.mkdir(parents=True, exist_ok=True)
def training_step(self, batch: Dict, batch_idx: int):
opt = self.optimizers()
scheduler = self.lr_schedulers()
loss, acc = self.model.forward(
batch["phoneme_ids"],
batch["phoneme_ids_len"],
batch["semantic_ids"],
batch["semantic_ids_len"],
batch["bert_feature"],
)
self.manual_backward(loss)
if batch_idx > 0 and batch_idx % 4 == 0:
opt.step()
opt.zero_grad()
scheduler.step()
self.log(
"total_loss",
loss,
on_step=True,
on_epoch=True,
prog_bar=True,
sync_dist=True,
)
self.log(
"lr",
scheduler.get_last_lr()[0],
on_epoch=True,
prog_bar=True,
sync_dist=True,
)
self.log(
f"top_{self.top_k}_acc",
acc,
on_step=True,
on_epoch=True,
prog_bar=True,
sync_dist=True,
)
def validation_step(self, batch: Dict, batch_idx: int):
return
def configure_optimizers(self):
model_parameters = self.model.parameters()
parameters_names = []
parameters_names.append(
[name_param_pair[0] for name_param_pair in self.model.named_parameters()]
)
lm_opt = ScaledAdam(
model_parameters,
lr=0.01,
betas=(0.9, 0.95),
clipping_scale=2.0,
parameters_names=parameters_names,
show_dominant_parameters=False,
clipping_update_period=1000,
)
return {
"optimizer": lm_opt,
"lr_scheduler": {
"scheduler": WarmupCosineLRSchedule(
lm_opt,
init_lr=self.config["optimizer"]["lr_init"],
peak_lr=self.config["optimizer"]["lr"],
end_lr=self.config["optimizer"]["lr_end"],
warmup_steps=self.config["optimizer"]["warmup_steps"],
total_steps=self.config["optimizer"]["decay_steps"],
)
},
}