[SPARK-14931] Mismatched default Param values between pipelines in Spark and PySpark

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Fix Version/s: 2.0.0
• Component/s: ML, PySpark
• Labels: ML PySpark
• Target Version/s: 2.0.0
• Assignee: Xusen Yin
• Reporter: Xusen Yin
• Shepherd: Joseph K. Bradley
• Created: 26/Apr/16 21:51
• Updated: 02/May/16 03:20
• Resolved: 01/May/16 19:29

Description

Mismatched default values between pipelines in Spark and PySpark lead to different pipelines in PySpark after saving and loading.

Find generic ways to check JavaParams then fix them.

Issue Links

• Is contained by: SPARK-14771 Python ML Param and UID issues
• is related to: SPARK-14706 Python ML persistence integration test
• links to: [Github] Pull Request #12738 (yinxusen)
• links to: [Github] Pull Request #12816 (jkbradley)