[SPARK-28914] GMM Returning Clusters With Equal Probability

• Type: Sub-task
• Status: Resolved
• Priority: Major
• Resolution: Incomplete
• Affects Version/s: 2.4.3
• Component/s: MLlib
• Labels: bulk-closed

Environment

https://gist.github.com/sharmanirek/172c9408e8393462ae54dfae83764413

Test Data & notebook: 

https://drive.google.com/open?id=1YEBsUJv9p2XaTQzsKI9GxLKVD7oe1tan

• Assignee: Unassigned
• Reporter: Nirek Sharma
• Created: 29/Aug/19 15:14
• Updated: 25/May/21 01:52
• Resolved: 25/May/21 01:41

Description

Fitting and Transforming with a GMM yields all categories with equal probability. See the below gist demonstrating the code to reproduce the problem.