You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Oct 22, 2023. It is now read-only.
I'm on a Mac using Parallels to run Windows. I've opened the Jupyter Notebook in Runpod, and installed the dependencies. When I try to launch training (!accelerate launch --mixed_precision="fp16" scripts/trainer.py), I get the following error:
I solved this issue by downloading dataloaders_util.py locally and then manually uploading it into the python notebook.
However, when I try to run training again, I get "TypeError: 'ABCMeta' object is not subscriptable", which I'm not sure how so solve.
Any ideas how to solve? Any other information I can supply to help find the error?
The text was updated successfully, but these errors were encountered:
I'm trying to run Stable Tuner on Runpod by following this tutorial: https://www.youtube.com/watch?v=8tzA-1x6FIA
I'm on a Mac using Parallels to run Windows. I've opened the Jupyter Notebook in Runpod, and installed the dependencies. When I try to launch training (!accelerate launch --mixed_precision="fp16" scripts/trainer.py), I get the following error:
I solved this issue by downloading dataloaders_util.py locally and then manually uploading it into the python notebook.
However, when I try to run training again, I get "TypeError: 'ABCMeta' object is not subscriptable", which I'm not sure how so solve.
Any ideas how to solve? Any other information I can supply to help find the error?
The text was updated successfully, but these errors were encountered: