Announcing Flower Datasets 0.3.0

The Flower Team is excited to announce the release of Flower Datasets 0.3.0!

Flower Datasets (flwr-datasets) is a library that quickly and easily creates datasets for federated learning, federated evaluation, and federated analytics. It was created by the Flower Labs team that also created Flower: A Friendly Federated Learning Framework.

What's new?

• New Partitioning Schemes:

  • Add pathological partitioner (#3623)
  • Add distribution partitioner (#3791)

• New Dataset integration tests and tests improvements:

  • Add tests for usps dataset (#3832)
  • Add tests for mnist-m dataset (#3834)
  • Add tests for femnist dataset (#3840)
  • Add tests for uci-mushrooms dataset (#3841)
  • Add function to perform partial download of dataset for tests (#3860)
  • Add tests for ucf101 dataset (#3842)
  • Add tests for ambient acoustic context (#3843)
  • Add ted-lium dataset to the tested set (#3844)
  • Update tests for ted lium dataset (#3868)

• Docs Improvements

  • Fix docs examples formatting (#3702)
  • Change format of Flower Datasets quickstart tutorial (#3854)
  • Add how to use partitioner docs (#3871)
  • Add details for new datasets (#3879)

Incompatible changes

None