sramirez / spark-MDLP-discretization

Spark implementation of Fayyad's discretizer based on Minimum Description Length Principle (MDLP)
Apache License 2.0
44 stars 27 forks source link

Issue 8 null label policy bb #17

Closed barrybecker4 closed 8 years ago

barrybecker4 commented 8 years ago

Fixes for #8 and #16 on this branch. The original fix for #8 used a check like _.label == null but that would never work because it always returns false. I changed the check to _.label.isNaN. I also have it through an IllegalArgumentException instead of logError so it fails fast and we can verify it in the unit test.

16 was pretty interesting. See notes in that issue.