[Query] Weight of evidence on Spark

[Aakash Basu]

Hi guys,

What's the best way to create feature column with Weight of Evidence
calculated for categorical columns on target column (both Binary and
Multi-Class)?

Any insight?

Thanks,
Aakash.