[SPARK-7651] PySpark GMM predict, predictSoft should fail on bad input

• Type: Bug
• Status: Resolved
• Priority: Minor
• Resolution: Fixed
• Affects Version/s: 1.3.0, 1.3.1, 1.4.0
• Fix Version/s: 1.3.2, 1.4.0
• Component/s: MLlib, PySpark
• Target Version/s: 1.3.2, 1.4.0
• Assignee: Meethu Mathew
• Reporter: Joseph K. Bradley
• Created: 14/May/15 23:00
• Updated: 15/May/15 21:53
• Resolved: 15/May/15 17:51

Description

In PySpark, GaussianMixtureModel predict and predictSoft test if the argument is an RDD and operate correctly if so. But if the argument is not an RDD, they fail silently, returning nothing.

https://github.com/apache/spark/blob/11a1a135d1fe892cd48a9116acc7554846aed84c/python/pyspark/mllib/clustering.py#L176

Instead, they should raise errors.

Issue Links

• links to: [Github] Pull Request #6180 (FlytxtRnD)