search

nateraw / replicate-examples

73 stars 20 forks source link

In video background remover replicate #2

Open tahercoolguy opened 3 days ago

tahercoolguy commented 3 days ago

In video background remover replicate . I have successful deployed on replicate through your code . But at the runtime i am getting this error

Prediction failed.

torch.cat(): expected a non-empty list of Tensors

`0 0.0 - 1.5 Traceback (most recent call last): File "/root/.pyenv/versions/3.11.9/lib/python3.11/site-packages/cog/server/worker.py", line 221, in _predict result = predict(**payload) ^^^^^^^^^^^^^^^^^^ File "/src/predict.py", line 25, in predict out = self.pipe(str(video), chunk_duration=1.5) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/src/pipeline.py", line 153, in call full_processed_video = torch.cat(processed_video_clips, dim=1) # Assuming dim=1 is the time dimension ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ RuntimeError: torch.cat(): expected a non-empty list of Tensors

`

And above is snippet

Can you please help what problem it can be

nateraw commented 3 days ago

Could it be a bad input file? Have you tried any of the example files that are in the demo I put on replicate?

nateraw commented 3 days ago

This one for example https://huggingface.co/spaces/nateraw/animegan-v2-for-videos/resolve/main/obama.webm

tahercoolguy commented 2 days ago

Thanks for it . I will try .

tahercoolguy commented 2 days ago

I have used above link i am getting decoding error "Unable to decode video stream"