Description
The LogisticRegression and LinearRegression models support training with a weight column, but the corresponding evaluators do not support computing metrics using those weights. This breaks model selection using CrossValidator.
Attachments
Issue Links
- is a child of
-
SPARK-18693 BinaryClassificationEvaluator, RegressionEvaluator, and MulticlassClassificationEvaluator should use sample weight data
- Resolved
- links to