[BUG/Help] p-tuningv2的时候报了No huggingface_hub attribute hf_api

[sungh66]

Is there an existing issue for this?

• [X] I have searched the existing issues

Current Behavior

尝试使用ptuningv2来微调模型，用的是提供的衣服数据集，用的是6b-int8的预训练权重（本地下载）

Expected Behavior

No response

Steps To Reproduce

我使用了本地下载的6b-int8权重，并覆盖了https://huggingface.co/THUDM/chatglm-6b-int8的clone，能正常部署cli_demo.py， 在尝试用ptuningv2微调时，将提供数据集解压放到此目录，训练参数如下 CUDA_VISIBLE_DEVICES=0 python3 main.py
--do_train
--train_file AdvertiseGen/train.json
--validation_file AdvertiseGen/dev.json
--prompt_column content
--response_column summary
--overwrite_cache
--model_name_or_path /home/Downloads/repo/ChatGLM/chatglm-6b-int8
--output_dir output/adgen-chatglm-6b-int8-pt-$PRE_SEQ_LEN-$LR
--overwrite_output_dir
--max_source_length 64
--max_target_length 64
--per_device_train_batch_size 1
--per_device_eval_batch_size 1
--gradient_accumulation_steps 16
--predict_with_generate
--max_steps 3000
--logging_steps 10
--save_steps 1000
--learning_rate $LR
--pre_seq_len $PRE_SEQ_LEN
--quantization_bit 4

报错如下 Traceback (most recent call last): File "//Downloads/repo/ChatGLM/ChatGLM-6B/ptuning/main.py", line 27, in from datasets import load_dataset File "/anaconda3/envs/chatglm-pt/lib/python3.9/site-packages/datasets/init.py", line 37, in from .builder import ArrowBasedBuilder, BeamBasedBuilder, BuilderConfig, DatasetBuilder, GeneratorBasedBuilder File "/anaconda3/envs/chatglm-pt/lib/python3.9/site-packages/datasets/builder.py", line 44, in from .data_files import DataFilesDict, _sanitize_patterns File "/anaconda3/envs/chatglm-pt/lib/python3.9/site-packages/datasets/data_files.py", line 120, in dataset_info: huggingface_hub.hf_api.DatasetInfo, File "/anaconda3/envs/chatglm-pt/lib/python3.9/site-packages/huggingface_hub/init.py", line 317, in getattr raise AttributeError(f"No {package_name} attribute {name}") AttributeError: No huggingface_hub attribute hf_api

Environment

- OS:ubuntu18
- Python:3.9
- Transformers:4.27.1
- PyTorch:1.12
- CUDA Support (`python -c "import torch; print(torch.cuda.is_available())"`) :true

Anything else?

No response

[willyzw1221]

+1

[YFAN1020]

+1