featran

A Scala feature transformation library for data science and machine learning

APACHE-2.0 License

Stars
466
Committers
27

Bot releases are hidden (Show)

featran - v0.8.0 Latest Release

Published by regadas over 1 year ago

What's new

This is mainly a maintenance release. Though it formalizes the release of scala 3 support.

What's Changed

New Contributors

Full Changelog: https://github.com/spotify/featran/compare/v0.7.0...v0.8.0

featran -

Published by regadas over 3 years ago

Dependency updates

  • scio v0.10.0
  • TensorFlow v0.3.0
  • Spark v3.1.1
  • Flink v1.12.2
featran -

Published by regadas over 3 years ago

  • Cross Compile to Scala 3.0.0-M3
  • Update TensorFlow to 2.3.1
  • Update XGboost to 1.3.1
  • Update Scio to 0.10.0-beta2
featran -

Published by nevillelyh about 4 years ago

  • Bump Scio to 0.10.0-alpha1
featran -

Published by nevillelyh over 4 years ago

  • Bump Scio to 0.9.0
  • Drop Scala 2.11 support
featran -

Published by nevillelyh almost 5 years ago

  • Add cross build for Scala 2.13
  • Allow empty sequences as input values for HashNHotEncoders (#312)
  • Add contramap method to Transformer (#311)
featran -

Published by nevillelyh about 5 years ago

  • Add Scala 2.12 cross builds for Flink & Spark
  • Decouple XGBoost from Scala binary version
featran -

Published by nevillelyh about 5 years ago

  • Add Indicator transformer
featran -

Published by regadas almost 6 years ago

featran -

Published by nevillelyh about 6 years ago

  • Add java support for extractWithSubsettings #182
featran - v0.2.0

Published by regadas about 6 years ago

  • Introduced FlatReader / FlatWriters (#161)
  • Removed JsonSerializable (#165)
  • Small clean ups and tweaks
featran -

Published by nevillelyh over 6 years ago

  • Add NamedSparseArray #141
  • Normalize feature names in TensorFlow Example #119 #138
featran -

Published by nevillelyh over 6 years ago

  • Add support for partial feature spec extraction #121
  • Treat empty input as missing value in NHot encoders #130
  • Use simulacrum typeclass macro #123 #126 #127 #128
featran -

Published by nevillelyh over 6 years ago

  • Fix TensorFlowFeatureBuilder serialization issue
featran -

Published by jbx over 6 years ago

  • Add RecordExtractor state as a ThreadLocal
  • (Spark users) make JMapWrapper serializable
featran -

Published by nevillelyh over 6 years ago

  • Fix dimensions in OneHotEncoder and TopNOneHotEncoder
featran -

Published by nevillelyh over 6 years ago

  • Bump scio to 0.5.1 and circe to 0.9.1
featran -

Published by nevillelyh over 6 years ago

  • Add encodeMissingValue option to encoders #96
featran -

Published by nevillelyh over 6 years ago

  • TopNOneHotEncoder #89
  • PositionEncoder #92
featran -

Published by ravwojdyla over 6 years ago

  • expose XGBoost LabeledPoint in Java API
  • bump dependencies