issue generating models with mix vision transformer encoders

[K-D-Gallagher]

I noticed that Mix Vision Transformer encoders are now listed as available. I was trying to generate a unet model with one of these encoders, but have been getting an error that it's not a supported encoder. I tried mit_b0, 3, and 5. Is there a particular prefix to use with these, like with the timm encoders? Thanks.

[qubvel]

Hi, it is not released to pypi, please install latest version from github repo to use these encoders

[qubvel]

$ pip install -U git+https://github.com/qubvel/segmentation_models.pytorch

[K-D-Gallagher]

Ah yes it worked, thank you so much! Looking forward to seeing how these encoders perform

[qubvel]

Let me know if you find it useful for your task)

[K-D-Gallagher]

It definitely boosted performance a bit! Maybe ~2% over my previously best optimized architecture/encoder combo. Thanks!

[qubvel]

Sounds great! Will try to move to PyPI release soon!

[github-actions[bot]]

This issue is stale because it has been open 60 days with no activity. Remove stale label or comment or this will be closed in 7 days.

[github-actions[bot]]

This issue was closed because it has been stalled for 7 days with no activity.