Skip to content

Conversation

@yinxusen
Copy link
Contributor

What changes were proposed in this pull request?

mllib.LDAExample uses ML pipeline and MLlib LDA algorithm. The former transforms original data into MLVector format, while the latter uses MLlibVector format.

How was this patch tested?

Test manually.

@SparkQA
Copy link

SparkQA commented Jul 15, 2016

Test build #62358 has finished for PR 14212 at commit 596aba6.

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

@SparkQA
Copy link

SparkQA commented Jul 21, 2016

Test build #62696 has finished for PR 14212 at commit de5eeab.

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

@yinxusen
Copy link
Contributor Author

Ping @mengxr @srowen

@MLnick
Copy link
Contributor

MLnick commented Aug 1, 2016

The fix LGTM - I'm in two minds about whether this example is "good" (illustrating using both ml and mllib APIs together) or "bad" (for the same reason, since there is LDA available in ml).

And we already have a LatentDirichletExample in mllib... seems to me that these mllib examples could use a clean up.

@yinxusen
Copy link
Contributor Author

yinxusen commented Aug 1, 2016

@MLnick They serve different purpose. This one is for users who have built their tools upon it. The LatentDirichletAllocationExample is for ML docs.

@asfgit asfgit closed this in dd8514f Aug 2, 2016
@yanboliang
Copy link
Contributor

LGTM, merged into master/branch-2.0. Thanks!

asfgit pushed a commit that referenced this pull request Aug 2, 2016
…LVector instead of MLlib Vector

## What changes were proposed in this pull request?

mllib.LDAExample uses ML pipeline and MLlib LDA algorithm. The former transforms original data into MLVector format, while the latter uses MLlibVector format.

## How was this patch tested?

Test manually.

Author: Xusen Yin <[email protected]>

Closes #14212 from yinxusen/SPARK-16558.

(cherry picked from commit dd8514f)
Signed-off-by: Yanbo Liang <[email protected]>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants