FP meets Feature Engineering & ML

The talk goes through a better approach to handle the end to data pipeline that involves scalable feature generation from big data right through to the data science pipeline that include mode training, model selection and scoring with pure abstractions in Scala, and discuss ZIO’s involvement as the target effect in handling concurrency and overcome some of the performance issues in Spark-ML. The talk will be focussing on Spark as backend of my abstraction.

Copyright © 2018-2019 Lunaya Consulting d.o.o.