Joel Murray

Is there any progress on this feature? I noticed [this library popped up in pypi](https://pypi.org/project/pillow-jxl-plugin/)

I think it should use the best checkpoint? the load_model_callback for model export uses: `load_model_callback = LoadModelCallback(weights_path=trainer.checkpoint_callback.best_model_path)` And the callbacks uses the ModelCheckpoint class which should support saving the best...

Thank you @jonsneyers for the very informed answer! Having the feature for the grayscale image of palette indices + a palette would be ideal for the project I am working...

@jendalinda that sounds like an interesting approach. I would like to test it and compare file sizes to our current uint16 PNGs, do you have any reference material on using...

Hi folks, looking into this again, I experimented a bit with the encoder and decoder for both our RGB images and our thermal 16 bit PNGs and saw a file...

@olokelo I 2nd @bgorissen request around 16 bit single channel support. I work with imaging systems that produce 16 bit single channel images, like thermal and depth, and the JXL...

This is very good work @olokelo! Is there any plan to add support for writing to JXL from Pillow?

@sovrasov I should probably ask this in that model_api repo but I see you are a contributor: It looks like the create_model function only works with a path to the...

Our integration with OTX SSL hasn't made it into production quite yet, I have some work queued up to determine how much gain we would see with real world datasets....

@harimkang @eunwoosh understandable about configurable input size and timelines. How about the original request? This (and the configurable resolution) are the biggest barriers for our full adoption of this library....