Colab? about text-to-video-finetuning HOT 12 CLOSED

Turns out the problem was just me. It takes a while and without any progressbar it seems to hang but it still working.
Continuing on.

from text-to-video-finetuning.

The following ensures the install happens in the right place (i.e. this is how to get the location to persist when doing cd in Google Colab):

!git clone https://github.com/ExponentialML/Text-To-Video-Finetuning.git
%cd Text-To-Video-Finetuning
!git lfs install
!git clone https://huggingface.co/damo-vilab/text-to-video-ms-1.7b ./models/model_scope_diffusers/

notice the %cd ...

from text-to-video-finetuning.

I made a quick attempt at it but kept getting

Error: Failed to call git rev-parse --git-dir: exit status 128
Git LFS initialized.

from text-to-video-finetuning.

The other problem I ran into was it not installing the models correctly in the proper directory because each command seems to run from root and not from last known location:

!git clone https://github.com/ExponentialML/Text-To-Video-Finetuning.git
!cd Text-To-Video-Finetuning
!git lfs install
!git clone https://huggingface.co/damo-vilab/text-to-video-ms-1.7b ./models/model_scope_diffusers/

Which means the process just hangs

from text-to-video-finetuning.

The other problem I ran into was it not installing the models correctly in the proper directory because each command seems to run from root and not from last known location:

!git clone https://github.com/ExponentialML/Text-To-Video-Finetuning.git
!cd Text-To-Video-Finetuning
!git lfs install
!git clone https://huggingface.co/damo-vilab/text-to-video-ms-1.7b ./models/model_scope_diffusers/

Which means the process just hangs

Did you find a fix for the issues yet? Because I don't have a PC rn and I don't know how to run the code anyways.

from text-to-video-finetuning.

Still looking. Not sure why git is hanging but can replicate the issue both on Colab and on my laptop

from text-to-video-finetuning.

Update: ran into the same issue here: #61 and suspected my process missed something. Going to regen another fine-tune and update when I can

from text-to-video-finetuning.

The other problem I ran into was it not installing the models correctly in the proper directory because each command seems to run from root and not from last known location:

!git clone https://github.com/ExponentialML/Text-To-Video-Finetuning.git
!cd Text-To-Video-Finetuning
!git lfs install
!git clone https://huggingface.co/damo-vilab/text-to-video-ms-1.7b ./models/model_scope_diffusers/

Which means the process just hangs

Did you find a fix for the issues yet? Because I don't have a PC rn and I don't know how to run the code anyways.

This is solved. Takes longer to download the weights than I expected and lack of progress indicator threw me off

from text-to-video-finetuning.

@patrickjonesdotca would you be able to share your colab notebook? I'm trying to get this to run in Colab as well and am running into the same issues mentioned above with install & inference
Thanks in advance!

from text-to-video-finetuning.

The following ensures the install happens in the right place (i.e. this is how to get the location to persist when doing cd in Google Colab):

!git clone https://github.com/ExponentialML/Text-To-Video-Finetuning.git
%cd Text-To-Video-Finetuning
!git lfs install
!git clone https://huggingface.co/damo-vilab/text-to-video-ms-1.7b ./models/model_scope_diffusers/

notice the %cd ...

Thank you!

from text-to-video-finetuning.

Hey, when trying
!python train.py --config /path/to/Text-To-Video-Finetuning/configs/v2/low_vram_config_example.yaml

get :
Traceback (most recent call last): File "/content/drive/MyDrive/Text-To-Video-Finetuning/train.py", line 998, in <module> main(**OmegaConf.load(args.config)) File "/content/drive/MyDrive/Text-To-Video-Finetuning/train.py", line 577, in main accelerator = Accelerator( TypeError: Accelerator.__init__() got an unexpected keyword argument 'logging_dir'

Any idea ?

from text-to-video-finetuning.

There is a finetuning Colab here: https://github.com/camenduru/Text-To-Video-Finetuning-colab

Please let me know if there are any issues, and I'll re-open.

from text-to-video-finetuning.