SmallZZTT

>=1.11

Hi， @shuluoshu I have a same question ,have you solved this problem? As your question ,can I think `val_data_idx.txt` is generated from `extract_rgbd_data.m`? But run `extract_rgbd_data.m`, I only get four...

感谢大佬回复，但是我这里使用的就是最新的代码 ```python def save_pretrained(self, output_dir: str): state_dict = self.hf_model.state_dict() state_dict = type(state_dict)( {k: v.clone().cpu() for k, v in state_dict.items()}) self.hf_model.save_pretrained(output_dir, state_dict=state_dict) ``` 另外transformers 版本 4.33.0

我重新测试了一下，好像是因为mix models导致的，下面是我的代码 ```python from LM_Cocktail import mix_models, mix_models_with_data # Mix fine-tuned model and base model; then save it to output_path: ./mixed_model_1 model = mix_models( model_names_or_paths=["/models/bge-reranker-base", "/sft_model/0221-4/checkpoint-150"], model_type='encoder', weights=[0.5, 0.5], #...