image_encoder?

[zhanghh-031102]

May I ask why there is no image_encoder in the pre-trained model?

[zhanghh-031102]

Is stabilityai/stable-video-diffusion-img2vid the corresponding one in huggingface? The error message indicates that config.json is missing. Which config.json is it (there is no config.json under the main of stabilityai/stable-video-diffusion-img2vid)

[zhanghh-031102]

OSError: Error no file named pytorch_model.bin, tf_model.h5, model.ckpt.index or flax_model.msgpack found in directory stabilityai/stable-video-diffusion-img2vid

[weijiawu]

1. The image_encoder has already been included in the pre-trained code of stabilityai/stable-video-diffusion-img2vid. For details, please refer to the pipeline/pipeline_svd_DragAnything.py file.

2. Could you post a screenshot of the error bug you encountered? It would help me understand the specific error better.

[zhanghh-031102]

Where can I download this from? Thanks for your answer。 在 2024-03-21 00:14:45，"weijiawu" ***@***.***> 写道： The image_encoder has already been included in the pre-trained code of stabilityai/stable-video-diffusion-img2vid. For details, please refer to the pipeline/pipeline_svd_DragAnything.py file. Could you post a screenshot of the error bug you encountered? It would help me understand the specific error better. — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: ***@***.***>

[weijiawu]

Where can I download this from? Thanks for your answer。 在 2024-03-21 00:14:45，"weijiawu" @.> 写道： The image_encoder has already been included in the pre-trained code of stabilityai/stable-video-diffusion-img2vid. For details, please refer to the pipeline/pipeline_svd_DragAnything.py file. Could you post a screenshot of the error bug you encountered? It would help me understand the specific error better. — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.>

https://huggingface.co/stabilityai/stable-video-diffusion-img2vid

[zhanghh-031102]

Since I cannot download directly from pycharm through "pretrained_model_name_or_path": "stabilityai/stable-video-diffusion-img2vid", I can only manually enter hugging face to download all the files in stabilityai/stable-video-diffusion-img2vid to the local. and then execute Then I found that an error was reported A config file is missing Please tell me how to solve this 在 2024-03-21 00:30:55，"weijiawu" ***@***.***> 写道： Where can I download this from? Thanks for your answer。 在 2024-03-21 00:14:45，"weijiawu" @.> 写道： The image_encoder has already been included in the pre-trained code of stabilityai/stable-video-diffusion-img2vid. For details, please refer to the pipeline/pipeline_svd_DragAnything.py file. Could you post a screenshot of the error bug you encountered? It would help me understand the specific error better. — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.> https://huggingface.co/stabilityai/stable-video-diffusion-img2vid — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: ***@***.***>

[PeisenZhao]

@zhanghh-031102

As you mentioned, "I can only manually enter Hugging Face to download all the files in stabilityai/stable-video-diffusion-img2vid to the local." The "config.json" file in "image_encoder" will be renamed to "image_encoder_config.json" due to manual downloading. You will need to manually change the file name.