Announcing Flower Datasets 0.3.0

General
1

image
image1200×628 74.7 KB

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

2 Likes