spotify/featran
A Scala library for feature transformation and engineering in machine learning workflows.
Not currently ranked — collecting fresh signals.
star history
Featran is a Scala library that simplifies feature engineering for machine learning by providing common transformation operations like scaling, encoding, and normalization. It implements a two-step pattern combining global aggregation followed by element-wise mapping, supporting algorithms such as Min-Max scaling and one-hot encoding. The library integrates with multiple data processing backends including Spark, Flink, Scalding, and Scio.
Frequently asked
- What is spotify/featran?
- A Scala library for feature transformation and engineering in machine learning workflows.
- Is featran open source?
- Yes — spotify/featran is open source, released under the Apache-2.0 license.
- What language is featran written in?
- spotify/featran is primarily written in Scala.
- How popular is featran?
- spotify/featran has 475 stars on GitHub.
- Where can I find featran?
- spotify/featran is on GitHub at https://github.com/spotify/featran.