[SPARK-9077] Improve error message for decision trees when numExamples < maxCategoriesPerFeature

• Type: Improvement
• Status: Resolved
• Priority: Trivial
• Resolution: Fixed
• Fix Version/s: 1.5.0
• Component/s: MLlib
• Labels: starter
• Target Version/s: 1.5.0
• Assignee: Sean R. Owen
• Reporter: Joseph K. Bradley
• Shepherd: Joseph K. Bradley
• Created: 15/Jul/15 20:30
• Updated: 31/Jul/15 00:26
• Resolved: 31/Jul/15 00:26

Description

See SPARK-9075's discussion for details. We should improve the current error message to recommend that the user remove the high-arity categorical features.

Issue Links

• relates to: SPARK-9075 DecisionTreeMetadata - setting maxPossibleBins to numExamples is incorrect.
• links to: [Github] Pull Request #7800 (srowen)