Description
We deprecated `runs` in Spark 1.6 (SPARK-11358). In 2.0, we can either remove `runs` or make it no effect (with warning messages). So we can simplify the implementation. I prefer the latter for better binary compatibility.
Attachments
Issue Links
- blocks
-
SPARK-11560 Optimize KMeans implementation / remove 'runs' from implementation
-
- Resolved
-
- is related to
-
SPARK-12363 PowerIterationClustering test case failed if we deprecated KMeans.setRuns
-
- Resolved
-
- relates to
-
SPARK-11358 Deprecate `runs` in k-means
-
- Resolved
-
- links to
(1 links to)