che.ender

Fine tune the Janus-Pro-7B model using ms swift

7

Fine tune the Janus-Pro-7B model using ms swift, And add a proxy service interface, Integrate Swift with fastapi_cient py The client can directly use fastapi_cient Py doesn't need to be...

尝试使用CPU主对 Qwen2.5-vl-3b进行微调，可以进行入训练过程， -------------------------------------------------------------------------------------------------- INFO:swift] Successfully registered post_encode hook: ['Qwen2_5_VLForConditionalGeneration']. Train: 0%| | 0/12945 [00:00

一种是转换成 transformers、llama、swift格式 ，然后出现了 ValueError: The checkpoint you are trying to load has model type `multi_modality` but Transformers does not recognize this architecture ，看了一下 应该是transformers 里没有配置文件中的multi_modality类别 第二种方法是onnxruntime,先启动模型，然后传输给onnxruntime,有了multi_modality类型了， 又报了 NotImplementedError: Module [MultiModalityCausalLM]...