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

Switch from segmengation-models-pytorch to torchseg #1

Merged
merged 3 commits into from
Jun 3, 2024

Conversation

zacharielegault
Copy link
Contributor

@zacharielegault zacharielegault commented May 25, 2024

SMP appears to not be very actively maintained anymore.

torchseg is a fork that

  • Supports recent timm versions, whereas SMP pins timm==0.9.7
  • Supports ViT encoders (not necessary at the moment, but could be useful in the future)
  • Has fewer dependencies (SMP notably depends on efficientnet-pytorch and pretrainedmodels, both of which are also unmaintained, and are superseded by timm anyway)

Minimum Python version is bumped to 3.9 to match torchseg (3.5 has long reached end-of-life status).

Also:
- add missing dependencies to pyproject.toml
- Bump min python version to match torchseg
@isaaccorley
Copy link

@zacharielegault Glad to see torchseg being used! Feel free to make an issue on our repo if you run into any bugs or want a new feature. We are happy to take any feedback!

@ClementPla ClementPla marked this pull request as ready for review June 3, 2024 13:33
@ClementPla ClementPla merged commit 9a14ed1 into ClementPla:main Jun 3, 2024
@zacharielegault
Copy link
Contributor Author

@isaaccorley Thanks for your work on torchseg! I'll let you know if we run into anything.

@notprime
Copy link

@zacharielegault amazing to see torchseg being finally used! let's us know if you bump or find anything strange, we are planning a lots of amazing features 😎

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.

4 participants