Process stuck

[piyushK52]

I am trying to run the process through the instructions given (this command in particular - sh scripts/text_condition/sample_video.sh).. but after a series of warnings the process freezes .

workspace/Open-Sora-Plan/venv/lib/python3.10/site-packages/torch/nn/modules/module.py:2025: UserWarning: for classifier.6.weight: copying from a non-meta parameter in the checkpoint to a meta parameter in the current model, which is a no-op. (Did you mean to pass `assign=True` to assign items in the state dictionary to their corresponding key in the module instead of copying them in place?)
  warnings.warn(f'for {key}: copying from a non-meta parameter in the checkpoint to a meta '
/workspace/Open-Sora-Plan/venv/lib/python3.10/site-packages/torch/nn/modules/module.py:2025: UserWarning: for classifier.6.bias: copying from a non-meta parameter in the checkpoint to a meta parameter in the current model, which is a no-op. (Did you mean to pass `assign=True` to assign items in the state dictionary to their corresponding key in the module instead of copying them in place?)
  warnings.warn(f'for {key}: copying from a non-meta parameter in the checkpoint to a meta '
/workspace/Open-Sora-Plan/venv/lib/python3.10/site-packages/torch/nn/modules/module.py:2025: UserWarning: for lin0.model.1.weight: copying from a non-meta parameter in the checkpoint to a meta parameter in the current model, which is a no-op. (Did you mean to pass `assign=True` to assign items in the state dictionary to their corresponding key in the module instead of copying them in place?)
  warnings.warn(f'for {key}: copying from a non-meta parameter in the checkpoint to a meta '
/workspace/Open-Sora-Plan/venv/lib/python3.10/site-packages/torch/nn/modules/module.py:2025: UserWarning: for lin1.model.1.weight: copying from a non-meta parameter in the checkpoint to a meta parameter in the current model, which is a no-op. (Did you mean to pass `assign=True` to assign items in the state dictionary to their corresponding key in the module instead of copying them in place?)
  warnings.warn(f'for {key}: copying from a non-meta parameter in the checkpoint to a meta '
/workspace/Open-Sora-Plan/venv/lib/python3.10/site-packages/torch/nn/modules/module.py:2025: UserWarning: for lin2.model.1.weight: copying from a non-meta parameter in the checkpoint to a meta parameter in the current model, which is a no-op. (Did you mean to pass `assign=True` to assign items in the state dictionary to their corresponding key in the module instead of copying them in place?)
  warnings.warn(f'for {key}: copying from a non-meta parameter in the checkpoint to a meta '
/workspace/Open-Sora-Plan/venv/lib/python3.10/site-packages/torch/nn/modules/module.py:2025: UserWarning: for lin3.model.1.weight: copying from a non-meta parameter in the checkpoint to a meta parameter in the current model, which is a no-op. (Did you mean to pass `assign=True` to assign items in the state dictionary to their corresponding key in the module instead of copying them in place?)
  warnings.warn(f'for {key}: copying from a non-meta parameter in the checkpoint to a meta '
/workspace/Open-Sora-Plan/venv/lib/python3.10/site-packages/torch/nn/modules/module.py:2025: UserWarning: for lin4.model.1.weight: copying from a non-meta parameter in the checkpoint to a meta parameter in the current model, which is a no-op. (Did you mean to pass `assign=True` to assign items in the state dictionary to their corresponding key in the module instead of copying them in place?)
  warnings.warn(f'for {key}: copying from a non-meta parameter in the checkpoint to a meta '
loaded pretrained LPIPS loss from taming/modules/autoencoder/lpips/vgg.pth

I am using a A100 with 80GB vram and 128GB ram

[qqingzheng]

During this phase, a download will be carried out. If you are a user in China, you may attempt to use a VPN for the download.

[piyushK52]

Thanks for the quick response. I am outside China.. multiple models were already downloaded automatically before coming to this step (I remember seeing a pytorch model of size 9GB..).. Can you give some estimate as to how long I should wait at this step and what's the total generation time in general (let's say for a single clip) ?

[qqingzheng]

Sorry, it seems that you have already downloaded the model. This step may involve generating a video, and I am not sure how long it will take to wait. You may wait for responses from other contributors within the repository. Alternatively, you can try debugging to determine where the code is stuck.

[piyushK52]

Got it. Thanks for the response, will debug a little and provide my findings here.

[IranQin]

@piyushK52 Hello, I also met the same problem, how did you solve it?