Rémi Tavon

My team has the same need, except we don't do overlapping predictions. Currently, we read an image tile by tile (with rasterio's Window), then stack predictions in a large array....

@dphogan This PR would be ready for review. I don't think I can launch CI tests myself. Am I wrong? Also, I added unit tests for Path object inputs, but...

> @dphogan This PR would be ready for review. I don't think I can launch CI tests myself. Am I wrong? Also, I added unit tests for Path object inputs,...

@rbavery have you a chance to take a look? I'll be leaving next monday until mi-august, was trying to close up a few things before leave :)

@rbavery friendly reminder about this PR and allowing CI tests for external contributions. Keep us posted.

@IAINATTWATER I reached out to development seed this morning trying to get in touch with @rbavery . We (Natural Resources Canada) benefit from solaris quite a bit, but if we...

@gaussiangit meanwhile, did you check the last developments? https://github.com/CosmiQ/solaris/blob/0.5.0/ROADMAP.md I guess @rbavery will soon drop the current main branch and replace it with 0.5.0.­.. EDIT: this conda environment is **not**...

@IAINATTWATER environment seems broken actually. Check #462

With Hydra, using the [Optuna plugin ](https://hydra.cc/docs/next/plugins/optuna_sweeper/) seems a better choice for hyperparameter optimization than hyperopt. The plugins leverages Hydra's [multirun feature](https://hydra.cc/docs/1.0/tutorials/basic/running_your_app/multi-run/).

Once data is written to hdf5, it's difficult to verify their content. Could there be a quick way to do quality control (visualize with matplotlib, etc.)? It could be on...