Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Update Rescaling Inputs #286

Merged
merged 2 commits into from
Dec 6, 2023
Merged

Update Rescaling Inputs #286

merged 2 commits into from
Dec 6, 2023

Conversation

anwai98
Copy link
Contributor

@anwai98 anwai98 commented Dec 5, 2023

WIP - @constantinpape I went ahead and added the changes to the LiveCELL dataloaders (as I had the setup already hehe) (you could take over for the other dataloaders, as discussed).
Thanks!

@constantinpape
Copy link
Contributor

@anwai98 I went ahead and implemented an identity transform. For most torch_em.get_default_dataloader this has to be used, otherwise it will fall back to standardize.
(This is not the case for the livecell loader as you correctly figured out, but I use the identity there as well for consistency).

In addition I added a check in the training that throws a warning if the data appears to be normalized.

I have not updated all the data loaders for generalist training yet, we can do this once we update those experiments.

@constantinpape constantinpape merged commit c0cdf1a into dev Dec 6, 2023
3 checks passed
@constantinpape constantinpape deleted the fix-norm branch December 6, 2023 08:35
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants