[SPARK-6518][MLlib][Example][DOC] Add example code and user guide for bisecting k-means

[yu-iskw]

This PR includes only an example code in order to finish it quickly.
I'll send another PR for the docs soon.

[yu-iskw]

Jenkins, retest this please

[SparkQA]

Test build #46657 has finished for PR 9952 at commit 7c8005d.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):\n * case class Params(\n

[yu-iskw]

@jkbradley could you review this PR?

[jkbradley]

When you modify this and [https://github.com//pull/9968] to use the include_example functionality, it may make sense to change this example code to use fixed parameters, rather than making it a command-line tool.

[yu-iskw]

All right. I'll fix it.

[SparkQA]

Test build #46790 has finished for PR 9952 at commit 86f6085.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds no public classes.

[yu-iskw]

@jkbradley could you review it?

[yu-iskw]

I moved the documentation to this PR from #9968. Because the PR depends on this PR.

I modified the docs for BisectingKMeans with include_example.

[SparkQA]

Test build #47095 has finished for PR 9952 at commit 2f4c2a5.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds no public classes.

[yu-iskw]

@jkbradley ping

[jkbradley]

Thanks! I'll review today

[jkbradley]

I'll go ahead and review this, but if you have time to add a Java example, that would be awesome--thanks!

[jkbradley]

Before this line (which is getting into details), it would be good to give a high-level description of the use of bisecting k-means. E.g., "Bisecting K-means can often be much faster than regular K-means, but it will generally produce a different clustering."

[jkbradley]

That should be it.

[jkbradley]

Also, can you please add the "[DOC]" tag to the PR title? Thanks!

[yu-iskw]

@jkbradley sure, done.

[jkbradley]

Thanks, I'll check back for updates for the few remaining comments.

[yu-iskw]

ping @jkbradley

[jkbradley]

@yu-iskw There are several comments of mine (above) which have not yet been addressed. Could you please update this PR based on those comments? I'll then take another look.

[yu-iskw]

Oh, I'm terribly sorry about that. Pushing the update was failed....

• Modified what you pointed out
• Add a Java example and its doc

[SparkQA]

Test build #47639 has finished for PR 9952 at commit 7c36c1d.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):\n * public class JavaBisectingKMeansExample\n

[jkbradley]

I would exclude JavaSparkContext and SparkConf.

[jkbradley]

No problem. I just added some comments and am generating the docs now.

[jkbradley]

The generated docs look good, so just those few comments.

[yu-iskw]

@jkbradley thanks for the review. I modified the wrong indentations and excluded the import statements for SparkConf and JavaSparkContext.

[SparkQA]

Test build #47780 has finished for PR 9952 at commit 3066696.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):\n * public class JavaBisectingKMeansExample\n

[jkbradley]

LGTM, merging with master and branch-1.6
Thanks!

[yu-iskw]

Thank you for merging it!