A neural gym for training deep learning models to carry out geoscientific image segmentation. Works best with labels generated using https://github.com/Doodleverse/dash_doodler
Gym and its companion repo, Zoo, use essentially the same conda environment. The Zoo readme (correctly) advises that the Zoo functions can be accessed from within a Gym conda environment.
Gym and its companion repo, Zoo, use essentially the same conda environment. The Zoo readme (correctly) advises that the Zoo functions can be accessed from within a Gym conda environment.
However, since the introduction of Zoo notebooks (https://github.com/Doodleverse/segmentation_zoo/issues/14), we now need to update the Gym conda environment to include
jupyter