ValueError: The checkpoint you are trying to load has model type `multi_modality` but Transformers does not recognize this architecture. This could be because of an issue with the checkpoint, or because your version of Transformers is out of date. #60
您好！
我下载该模型搭配LLamafactory框架，在做api部署的时候，报以下错误
[2024-10-01 00:15:35,483] [INFO] [real_accelerator.py:203:get_accelerator] Setting ds_accelerator to cuda (auto detect)
[INFO|configuration_utils.py:670] 2024-10-01 00:15:38,538 >> loading configuration file /mnt/ssd2/models/deepseek-vl-7b-chat/config.json
Traceback (most recent call last):
File "/home/ubuntu/miniconda3/envs/panc_math_vscode/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py", line 1023, in from_pretrained
config_class = CONFIG_MAPPING[config_dict["model_type"]]
File "/home/ubuntu/miniconda3/envs/panc_math_vscode/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py", line 725, in getitem
raise KeyError(key)
KeyError: 'multi_modality'
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "/home/ubuntu/miniconda3/envs/panc_math_vscode/bin/llamafactory-cli", line 8, in
sys.exit(main())
File "/home/ubuntu/pqj/math/LLaMA-Factory/src/llamafactory/cli.py", line 79, in main
run_api()
File "/home/ubuntu/pqj/math/LLaMA-Factory/src/llamafactory/api/app.py", line 129, in run_api
chat_model = ChatModel()
File "/home/ubuntu/pqj/math/LLaMA-Factory/src/llamafactory/chat/chat_model.py", line 52, in init
self.engine: "BaseEngine" = HuggingfaceEngine(model_args, data_args, finetuning_args, generating_args)
File "/home/ubuntu/pqj/math/LLaMA-Factory/src/llamafactory/chat/hf_engine.py", line 54, in init
tokenizer_module = load_tokenizer(model_args)
File "/home/ubuntu/pqj/math/LLaMA-Factory/src/llamafactory/model/loader.py", line 69, in load_tokenizer
config = load_config(model_args)
File "/home/ubuntu/pqj/math/LLaMA-Factory/src/llamafactory/model/loader.py", line 122, in load_config
return AutoConfig.from_pretrained(model_args.model_name_or_path, **init_kwargs)
File "/home/ubuntu/miniconda3/envs/panc_math_vscode/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py", line 1025, in from_pretrained
raise ValueError(
ValueError: The checkpoint you are trying to load has model type multi_modality but Transformers does not recognize this architecture. This could be because of an issue with the checkpoint, or because your version of Transformers is out of date.
请问怎么修改，我的transformer=4.45.0，具体环境附上图。
LIMYOONA8
commented
1 week ago
您好！ 我下载该模型搭配LLamafactory框架，在做api部署的时候，报以下错误 [2024-10-01 00:15:35,483] [INFO] [real_accelerator.py:203:get_accelerator] Setting ds_accelerator to cuda (auto detect) [INFO|configuration_utils.py:670] 2024-10-01 00:15:38,538 >> loading configuration file /mnt/ssd2/models/deepseek-vl-7b-chat/config.json Traceback (most recent call last): File "/home/ubuntu/miniconda3/envs/panc_math_vscode/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py", line 1023, in from_pretrained config_class = CONFIG_MAPPING[config_dict["model_type"]] File "/home/ubuntu/miniconda3/envs/panc_math_vscode/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py", line 725, in getitem raise KeyError(key) KeyError: 'multi_modality'
During handling of the above exception, another exception occurred:
Traceback (most recent call last): File "/home/ubuntu/miniconda3/envs/panc_math_vscode/bin/llamafactory-cli", line 8, in
sys.exit(main())
File "/home/ubuntu/pqj/math/LLaMA-Factory/src/llamafactory/cli.py", line 79, in main
run_api()
File "/home/ubuntu/pqj/math/LLaMA-Factory/src/llamafactory/api/app.py", line 129, in run_api
chat_model = ChatModel()
File "/home/ubuntu/pqj/math/LLaMA-Factory/src/llamafactory/chat/chat_model.py", line 52, in init
self.engine: "BaseEngine" = HuggingfaceEngine(model_args, data_args, finetuning_args, generating_args)
File "/home/ubuntu/pqj/math/LLaMA-Factory/src/llamafactory/chat/hf_engine.py", line 54, in init
tokenizer_module = load_tokenizer(model_args)
File "/home/ubuntu/pqj/math/LLaMA-Factory/src/llamafactory/model/loader.py", line 69, in load_tokenizer
config = load_config(model_args)
File "/home/ubuntu/pqj/math/LLaMA-Factory/src/llamafactory/model/loader.py", line 122, in load_config
return AutoConfig.from_pretrained(model_args.model_name_or_path, **init_kwargs)
File "/home/ubuntu/miniconda3/envs/panc_math_vscode/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py", line 1025, in from_pretrained
raise ValueError(
ValueError: The checkpoint you are trying to load has model type 
multi_modalitybut Transformers does not recognize this architecture. This could be because of an issue with the checkpoint, or because your version of Transformers is out of date. 请问怎么修改，我的transformer=4.45.0，具体环境附上图。