sxela
patreon
For people having pytorch\xformers\torchvision.transforms.functional_tensor error
🕑 Added 2024-03-14 15:33:10 +0000 UTCGoogle Colab has updated its env, so we need to fix a few things for it to work.
Installers for v0.20+ have been updated. Those notebooks should work without any manual fixes.
For people having pytorch\xformers error (xFormers can't load C++/CUDA extensions) and all sorts of import errors like ModuleNotFoundError: No module named 'torchvision.transforms.functional_tensor':
1) run Prepare folders & Install cell
2) check "skip install" after running it
3) add a new code cell below it with this code:
!pip uninstall torch xformers -y
!python -m pip install torch==2.0.1 torchvision==0.15.2 --extra-index-url https://download.pytorch.org/whl/cu118 xformers==0.0.21and run it.
4) delete or comment that newly added cell to make sure you don't run it every time you restart the env.
5) restart your env and run all as usual
Comments
Warpfusion
Make sure to join discord, I reply faster there, as Patreon only notifies via email
Igor Wewers
okay β thank you also for the fast response! I'm amazed by this notebook! Cheers
Warpfusion
It's Inside braces () in your output frames names
Igor Wewers
Ok, got it thanks! How can one find run number if the env is restarted? sorry if that seems like a dumb question
Warpfusion
It's either skip checkbox or skip selection in drop-down. If you don't have that, then you're using an older version. Then you will have to start the render cell, cancel it and go to video export, then manually input run number there
Igor Wewers
Yes i got my frames saved βοΈ Okay i will double check that as i must have missed it π€
Warpfusion
But default the output frames are saved on your Google drive. You can also run the notebook and skip the render (there is a checkbox or drop-down option for that in cell 4), and then go to save video cell.
Igor Wewers
Hey there, thank you for the recent tut you uploaded! i obviously liked and commented on YT. However there's one question that arose: after a successful run i sometimes struggle with the video generation section, due to google colab being in a half active state but active due to background execution. i seem to have never successfully generated a video unfortunately and now I#m asking if you could provide or help me find a solution so i can run this cell independently with my generated frames if some goes south in a 14h run
Warpfusion
Does it work now? without extra tweaks
Warpfusion
Does it work now? without extra tweaks
Footless Pirate
All working better now with your code, only if I don't restart it
Pablo Azevedo
Hi! You can show this steps in scrren shots? Please :D . I cant resolve this problem :(
Footless Pirate
This works on v24.6, not v30 or v27. Thanks for the help been trying all day. I hate when Google Colab updated its env lol
