Skip to content

[SPARK-14409][ML] Adding a RankingEvaluator to ML #12461

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Closed
wants to merge 4 commits into from
Closed

[SPARK-14409][ML] Adding a RankingEvaluator to ML #12461

wants to merge 4 commits into from

Conversation

@yongtang
Copy link
Contributor

@yongtang yongtang commented Apr 18, 2016 •
edited
Loading

What changes were proposed in this pull request?

This patch adds the implementation of Mean Rreciprocal Rank (MRR) in mllib.evaluation, and adds a RankingEvaluator to ML (ml.evaluation)

How was this patch tested?

Additional test cast has been added to cover Mean Rreciprocal Rank (MRR).

MLnick
MLnick reviewed Apr 18, 2016
View reviewed changes
mllib/src/main/scala/org/apache/spark/mllib/evaluation/RankingMetrics.scala Outdated
Copy link
Contributor

@MLnick MLnick Apr 18, 2016

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It would be good to expand the doc string here with more detail about what is MRR and how it is computed (see the ndcg doc string above).

Copy link
Contributor Author

@yongtang yongtang Apr 20, 2016

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks @MLnick I just updated the docs.

@MLnick
Copy link
Contributor

MLnick commented Apr 18, 2016

ok to test

@SparkQA
Copy link

SparkQA commented Apr 18, 2016

Test build #56105 has finished for PR 12461 at commit 2838c24.

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

@yongtang yongtang changed the title [SPARK-14409][ML][WIP] Adding a RankingEvaluator to ML [SPARK-14409][ML] Adding a RankingEvaluator to ML Apr 20, 2016
@yongtang
Copy link
Contributor Author

yongtang commented Apr 20, 2016

@MLnick I added the implementation for ml.evaluation.RankingEvaluator. Please let me know if there are any issues.

@SparkQA
Copy link

SparkQA commented Apr 20, 2016

Test build #56317 has finished for PR 12461 at commit af75c66.

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

MLnick
MLnick reviewed Apr 20, 2016
View reviewed changes
mllib/src/main/scala/org/apache/spark/ml/evaluation/RankingEvaluator.scala Outdated
Copy link
Contributor

@MLnick MLnick Apr 20, 2016

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

these columns must in fact be ArrayType

@yongtang
Copy link
Contributor Author

yongtang commented Apr 21, 2016

Thanks @MLnick @srowen let me take some time and address all the issues.

@MLnick MLnick mentioned this pull request Apr 21, 2016
Closed
3 tasks
@yongtang
Copy link
Contributor Author

yongtang commented Apr 26, 2016

@MLnick @srowen I just updated the pull request to wrap the RankingEvaluator into calling RankingMetrics. Was finally able to fix the exception issue I previously encountered. Sorry for the delay and let me know if there are any issues.

@SparkQA
Copy link

SparkQA commented Apr 26, 2016

Test build #56952 has finished for PR 12461 at commit a35d961.

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

@SparkQA
Copy link

SparkQA commented May 2, 2016

Test build #57506 has finished for PR 12461 at commit c50c469.

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

jayzes pushed a commit to ello/spark-jobs that referenced this pull request Jan 22, 2017
Split the model into Builder, Tuner and Recommender
e66722a 
Until apache/spark#12461 is merged, the ML pipeline doesn’t have the same ability to generate top recommendations for a set of users. So, we’ll tune the model with an ML pipeline, build it from the same data with the RDD-based ALS implementation, then have a component to load that model and generate recs into an intermediate store TBD.
yongtang added 4 commits March 2, 2017 11:51
@yongtang
[SPARK-14409][ML][WIP] Adding a RankingEvaluator to ML
bd324cd 
This patch tries to add the implementation of Mean Rreciprocal Rank (MRR)
in mllib.evaluation, as a first step toward adding a RankingEvaluator to ML.

Additional test cast has been added to cover Mean Rreciprocal Rank (MRR).

This patch is a work in progress.
@yongtang
[SPARK-14409][ML][WIP] Adding a RankingEvaluator to ML
05580e0 
This patch tries to add the implementation of RankingEvaluator in ml.evaluation.

Additional test cast has been added to cover the implementation.
@yongtang
[SPARK-14409][ML][WIP] Adding a RankingEvaluator to ML
19ea63b 
This patch tries to update the comment for mean reciprocal rank.
@yongtang
[SPARK-14409][ML] Adding a RankingEvaluator to ML
78f13ad 
This patch tries to consolidate ml.evaluation and mllib.evaluation
so that RankingEvaluator wraps RankingMetrics.
@SparkQA
Copy link

SparkQA commented Mar 3, 2017

Test build #73804 has finished for PR 12461 at commit 78f13ad.

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

@yongtang
Copy link
Contributor Author

yongtang commented Mar 4, 2017

/cc @daniloascione please take a look.

@AmplabJenkins
Copy link

AmplabJenkins commented Jan 18, 2018

Can one of the admins verify this patch?

@yongtang yongtang closed this Jan 18, 2018
@yongtang yongtang deleted the SPARK-14409 branch January 18, 2018 18:14
@flolas
Copy link

flolas commented Jan 13, 2019 •
edited
Loading

Hi, is there any plans for merging this implementation of RankingEvaluator?. I would be glad to help if needed.

Thanks.

@srowen
Copy link
Member

srowen commented Jan 13, 2019

WYSIWYG here. This one was abandoned and I don't see any other activity. If you want to take a crack at a clean change that follows on this, sure, we can look at it.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

6 participants

@yongtang @MLnick @SparkQA @AmplabJenkins @flolas @srowen