sberbank-ai-lab/RePlay

rewrite DataPreparator

monkey0head opened this issue ยท 0 comments

๐Ÿš€ Feature Request

Datapreparator should become a class used to:

  • convert dataframe to spark
  • check dataframe for NaNs
  • convert timestamp

Motivation

To make a preprocessing sequence:
DataPreparator (get spark dataframe with valid data types) - Splitters/Filters (split/crop dataframe) - Indexer (create indexers for train at first)