run app.py报错

[txiang163]

在运行app.py时出现以下错误
(my_lm) xt@ji-jupyter-6713700621420699648-master-0:~/txiang/LMFlow/service$ python app.py
Explicitly passing a revision is encouraged when loading a configuration with custom code to ensure no malicious code has been contributed in a newer revision.
Failed to use RAM optimized load. Automatically use original load instead.
Explicitly passing a revision is encouraged when loading a configuration with custom code to ensure no malicious code has been contributed in a newer revision.
Traceback (most recent call last):
File "/home/xt/txiang/LMFlow/service/../src/lmflow/models/hf_decoder_model.py", line 192, in init
self.backend_model = AutoModelForCausalLM.from_pretrained(
File "/home/xt/anaconda3/envs/my_lm/lib/python3.9/site-packages/transformers/models/auto/auto_factory.py", line 474, in from_pretrained
raise ValueError(
ValueError: Unrecognized configuration class <class 'transformers_modules.configuration_chatglm.ChatGLMConfig'> for this kind of AutoModel: AutoModelForCausalLM.
Model type should be one of BartConfig, BertConfig, BertGenerationConfig, BigBirdConfig, BigBirdPegasusConfig, BioGptConfig, BlenderbotConfig, BlenderbotSmallConfig, BloomConfig, CamembertConfig, CodeGenConfig, CTRLConfig, Data2VecTextConfig, ElectraConfig, ErnieConfig, GitConfig, GPT2Config, GPT2Config, GPTNeoConfig, GPTNeoXConfig, GPTNeoXJapaneseConfig, GPTJConfig, LlamaConfig, MarianConfig, MBartConfig, MegaConfig, MegatronBertConfig, MvpConfig, OpenAIGPTConfig, OPTConfig, PegasusConfig, PLBartConfig, ProphetNetConfig, QDQBertConfig, ReformerConfig, RemBertConfig, RobertaConfig, RobertaPreLayerNormConfig, RoCBertConfig, RoFormerConfig, Speech2Text2Config, TransfoXLConfig, TrOCRConfig, XGLMConfig, XLMConfig, XLMProphetNetConfig, XLMRobertaConfig, XLMRobertaXLConfig, XLNetConfig, XmodConfig.

我是在hugging face models里面把chatglm下载下来放到了output_models目录下，运行的时候说无法识别ChatGLMConfig这个配置类，这个问题该如何解决？

[research4pan]

Thanks for your interest in LMFlow! Currently we haven't merged the support for chatglm yet. The corresponding pull request is this one. We will let you know once the merge is completed. Thanks for your understanding 🙏

[txiang163]

So what are the models on huggingface that you currently support?

[research4pan]

We support all decoder-only models in huggingface, such as gpt2, gpt2-large, gpt2-xl, gpt-neo-2.7b, galactica, bloom, llama etc. The aforementioned pull request will introduce more supports of all encoder-decoder models in huggingface, such as T5. Hope that answers your question. Thanks 😄

[research4pan]

As for the issue not displayed here (but in my mailbox), you may run the ./scripts/run_chatbot.sh under the project directory. Running it under ./scripts may result in a model-not-found error. Hope that helps. Thanks~

[shizhediao]

This issue has been marked as stale because it has not had recent activity. If you think this still needs to be addressed please feel free to reopen this issue. Thanks