Will torchvision's new API effect pytorch lightning? #11354
Unanswered
bw4sz
asked this question in
Lightning Trainer API: Trainer, LightningModule, LightningDataModule
Replies: 1 comment
-
We should be good, Flash might be another story though, cc @ethanwharris for awareness |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Torchvision is prototyping a new API for calling multiple model model weights from the same architecture.
https://pytorch.org/blog/introducing-torchvision-new-multi-weight-support-api/
and looking for public comment:
pytorch/vision#5088
I think this community should have a quick scan and see if there are any reasons why the new API may interfere with any of the PL design. In some ways it duplicates the checkpoint loading idea and is working in parallel with the data module train_dataloader design by trying to store information on transforms and other pre-processing techniques.
Thoughts?
Beta Was this translation helpful? Give feedback.
All reactions