search

EIDOSLAB / torchstain

Stain normalization tools for histological analysis and computational pathology
MIT License
118 stars 20 forks source link

Simplified imports and refactored #14

Closed andreped closed 2 years ago

andreped commented 2 years ago

As discussed in https://github.com/EIDOSLAB/torchstain/pull/13#issue-1330100162, I have simplified the imports, such that it works as expected. Simply use torchstain.MacenkoNormalizer(backend) as before, but works seemlessly with all backends.

As I believe this is now ready to be released, I also did some refactoring in the README and whatnot. I made a suggestion to edit the description of the tool in the README. You can edit as you please.

I set to merge with the backends branch. You can merge with main after if you approve of edits.

Changes:

carloalbertobarbano commented 2 years ago

Thanks. I had some work started on this already, I will see tomorrow about merging them :+1:

andreped commented 2 years ago

Thanks. I had some work started on this already, I will see tomorrow about merging them 👍

After taking a nap, I got an idea of how to solve the problem, and it seems like the nap worked :p Quite easy in the end. Just me being silly

carloalbertobarbano commented 2 years ago

Merged into the backends branch. I kept the "core" backage named to "base" as it is more appropriate for providing base classes. I also moved the numpy implementation to its own backend package, for coherence with the rest of the codebase

andreped commented 2 years ago

I also moved the numpy implementation to its own backend package, for coherence with the rest of the codebase

Good idea. I was thinking about doing the same, but I was lazy :p