KMnO4-zx/huanhuan-chat

无法使用peftmodel

MZGuo111 opened this issue · 1 comments

MZGuo111 commented

实在不好意思，我没有发现讨论区，只能占用您issue区的资源了，请见谅！
我尝试在本地部署huanhuanchat，我成功调用了chatglm，但是无法使用微调后的模型，我的代码如下：

from peft import PeftModel
from transformers import AutoTokenizer, AutoModel

model_path = "THUDM/chatglm2-6b"
model = AutoModel.from_pretrained(model_path, trust_remote_code=True).half().cuda()
tokenizer = AutoTokenizer.from_pretrained(model_path, trust_remote_code=True)
# 到这里都没有问题


#   给你的模型加上嬛嬛LoRA! 
model = PeftModel.from_pretrained(model, "output/sft").half()
model.eval()

报错信息如下：

HTTPError: 401 Client Error: Unauthorized for url: https://huggingface.co/output/sft/resolve/main/adapter_config.json

The above exception was the direct cause of the following exception:

RepositoryNotFoundError                   Traceback (most recent call last)
    176 try:
--> 177     config_file = hf_hub_download(
    178         model_id,
    179         CONFIG_NAME,
    180         **hf_hub_download_kwargs,
    181     )
    182 except Exception:

File [~/anaconda3/envs/deeplearning/lib/python3.8/site-packages/huggingface_hub/utils/_validators.py:120](https://vscode-remote+ssh-002dremote-002b7b22686f73744e616d65223a225461726765744d616368696e65227d.vscode-resource.vscode-cdn.net/home/mguo/dl/llm/~/anaconda3/envs/deeplearning/lib/python3.8/site-packages/huggingface_hub/utils/_validators.py:120), in validate_hf_hub_args.._inner_fn(*args, **kwargs)
    118     kwargs = smoothly_deprecate_use_auth_token(fn_name=fn.__name__, has_token=has_token, kwargs=kwargs)
--> 120 return fn(*args, **kwargs)

File [~/anaconda3/envs/deeplearning/lib/python3.8/site-packages/huggingface_hub/file_download.py:1195](https://vscode-remote+ssh-002dremote-002b7b22686f73744e616d65223a225461726765744d616368696e65227d.vscode-resource.vscode-cdn.net/home/mguo/dl/llm/~/anaconda3/envs/deeplearning/lib/python3.8/site-packages/huggingface_hub/file_download.py:1195), in hf_hub_download(repo_id, filename, subfolder, repo_type, revision, library_name, library_version, cache_dir, local_dir, local_dir_use_symlinks, user_agent, force_download, force_filename, proxies, etag_timeout, resume_download, token, local_files_only, legacy_cache_layout)
...
--> 183         raise ValueError(f"Can't find '{CONFIG_NAME}' at '{model_id}'")
    185 loaded_attributes = cls.from_json_file(config_file)
    186 return loaded_attributes["peft_type"]

ValueError: Can't find 'adapter_config.json' at 'output/sft'

请问这个如何解决呢？

KMnO4-zx commented

请查看output/sft目录下是否存在这两个文件，以及文件是否下载完整，可以尝试单独下载这两个文件，以及请问您是在根目录下运行该文件的嘛，chatglm2-6b模型需要下载到本地，然后将THUDM/chatglm2-6b路径更换为下载好的chatglm'模型路径。期待你的成功调用~
image