Efficient Datamodules customized for the Mila/ComputeCanada clusters.

NOTE: This is a work in progress. Please let us know what you think, what features you'd like to see, what datasets we're forgetting, etc, using the issues tab.

This can be installed with pip:

pip install "mila_datamodules @ git https://www.github.com/mila-iqia/mila_datamodules.git"

TODO: Add more detailed usage instructions and examples.

$ mila_datamodules prepare <dataset_name>

To see the list of available dataset preprocessing functions, use:

$ mila_datamodules prepare --help

TODO: Show users how to add a preparation function for their own dataset, and describe the advantages of doing so.