speech_recognition/FunASR
1
0
Fork 0
mirror of https://github.com/modelscope/FunASR synced 2025-09-15 14:48:36 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

Merge pull request #2267 from modelscope/dev_sx2

Browse Source 
fix model_wrapper and seaco finetune
This commit is contained in:
Shi Xian 2024-12-05 15:15:38 +08:00 committed by GitHub
commit 0efc87352c
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 2 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
examples/industrial_data_pretraining/seaco_paraformer/finetune.sh
View File

@ -78,5 +78,6 @@ torchrun $DISTRIBUTED_ARGS \
++train_conf.avg_nbest_model=10 \
++train_conf.use_deepspeed=false \
++train_conf.deepspeed_config=${deepspeed_config} \
++train_conf.find_unused_parameters=true \
++optim_conf.lr=0.0002 \
++output_dir="${output_dir}" &> ${log_file}

2
funasr/bin/train_ds.py
View File

@ -134,7 +134,7 @@ def main(**kwargs):
**kwargs.get("train_conf"),
)
model = trainer.warp_model(model)
model = trainer.warp_model(model, **kwargs)
kwargs["device"] = int(os.environ.get("LOCAL_RANK", 0))
trainer.device = int(os.environ.get("LOCAL_RANK", 0))