search

AILab-CVC / VideoCrafter

VideoCrafter2: Overcoming Data Limitations for High-Quality Video Diffusion Models
https://ailab-cvc.github.io/videocrafter2/
Other
4.57k stars 341 forks source link

AttributeError: 'VisionTransformer' object has no attribute 'input_patchnorm' #39

Closed Laidawang closed 1 year ago

Laidawang commented 1 year ago

when i try to run sh scripts/run_image2video.sh i got AttributeError: 'VisionTransformer' object has no attribute 'input_patchnorm' All models have been downloaded and placed in the designated location.

LetifT commented 1 year ago

I just ran into the same error

alittleanimal commented 1 year ago

same problem

quartets commented 1 year ago

Same problem

jiangyurong609 commented 1 year ago

same here

447806664 commented 1 year ago

same problem

kanik155 commented 1 year ago

It is an open_clip version issue. pip install open-clip-torch==2.12.0

ZombaSY commented 1 year ago

It is an open_clip version issue. pip install open-clip-torch==2.12.0

Currently solved with the package versioning:

scutpaul commented 1 year ago

Thanks to @kanik155 @ZombaSY, we'll fix this in a recent commit!