sunneam

基于damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch模型微调,运行finetune.py报错

4

 环境:linux python=3.9.0 torch=2.1.1 funasr=0.8.4 modelscope=1.9.1

[4090c] 2023-11-23 05:36:47,586 (build_trainer:153) INFO: The training was resumed using /home/cnhis/whyme/FunASR/egs_modelscope/asr/TEMPLATE/checkpoint/checkpoint.pb [4090c] 2023-11-23 05:36:47,604 (build_trainer:260) INFO: 4/100epoch started [4090c] 2023-11-23 05:36:50,002 (build_trainer:302) INFO: 4epoch results: [train] time=2.05 seconds, total_count=0, gpu_max_cached_mem_GB=1.723,...

训练方言数据时acc为0

11

#### What is your question? 使用开源方言数据集训练,Loss值很高,验证集效果很差,日志中best acc都是0 #### Code