ValueError: unet/mvdiffusion.models.unet_mv2d_condition.py as defined in `model_index.json` does not exist in flamehaze1115/wonder3d-v1.0 and is not a module in 'diffusers/pipelines'.

[YangJunghoon]

i got this error when i run python mv.py --uid YOUR_EXAMPLE_ID
-my packages version -
huggingface-hub 0.26.2
diffusers 0.31.0

• if i want test my custom img, how can i setting preprocessed
  512*512 size
  etc.... (ffc_res, mask)

[AbyssBadger0]

i got this error when i run python mv.py --uid YOUR_EXAMPLE_ID -my packages version - huggingface-hub 0.26.2 diffusers 0.31.0

• if i want test my custom img, how can i setting preprocessed
  512*512 size
  etc.... (ffc_res, mask)

我也遇到了这个问题，已经解决了，不过不太清楚是哪一步解决了关键问题
我做过的事：
1.在DrawingSpinUp/2_charactor_reconstructor中 git clone https://huggingface.co/flamehaze1115/wonder3d-v1.0
2.修改2_charactor_reconstructor/configs/mvdiffusion-joint-ortho-6views.yaml这个文件中第二行为如下
pretrained_model_name_or_path: './wonder3d-v1.0'
3.重新指定版本
diffusers[torch]==0.19.3
huggingface_hub==0.25.0

如果你想使用自定义的数据集可以参考这个：
#26 (comment)

[LordLiang]

Hello, the version of diffusers is important. Wonder3D is only tested on diffusers[torch]==0.19.3, may have conflicts with newer versions of diffusers.