error:RuntimeError: Error(s) in loading state_dict for CLIPVisionModel: size mismatch for vision_model.embeddings.class_embedding: copying a param with shape torch.Size([1024]) from checkpoint, the shape in current model is torch.Size([768]).

[zapqqqwe]

为什么报错呢，下载的LanguageBind/LanguageBind_Video_merge 和 LanguageBind/LanguageBind_Image 在本地同时，在config.json文件修改了mm_video_tower 和 mm_image_tower 分别为本地的位置，但是报错，我看好像clip的隐藏层768但是设置的为1024，怎么解决呢

[zapqqqwe]

huggface里面的LanguageBind/LanguageBind_Video_merge 和 LanguageBind/LanguageBind_Image

[Liu98C]

huggface里面的LanguageBind/LanguageBind_Video_merge 和 LanguageBind/LanguageBind_Image

请问您解决了吗

[Liu98C]

huggface里面的LanguageBind/LanguageBind_Video_merge 和 LanguageBind/LanguageBind_Image

#57 (comment)