Skip to content
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

[BRANCH-1.2][SPARK-4583][MLLIB] LogLoss for GradientBoostedTrees fix + doc updates #3474

Closed
wants to merge 2 commits into from
Closed

[BRANCH-1.2][SPARK-4583][MLLIB] LogLoss for GradientBoostedTrees fix + doc updates #3474

wants to merge 2 commits into from

Conversation

mengxr
Copy link
Contributor

@mengxr mengxr commented Nov 26, 2014

We reverted #3439 in branch-1.2 due to missing import o.a.s.SparkContext._, which is no longer needed in master (#3262). This PR adds #3439 back to branch-1.2 with correct imports.

Github is out-of-sync now. The real changes are the last two commits.

@jkbradley @mengxr
[SPARK-4583] [mllib] LogLoss for GradientBoostedTrees fix + doc updates
6b5564a 
Currently, the LogLoss used by GradientBoostedTrees has 2 issues:
* the gradient (and therefore loss) does not match that used by Friedman (1999)
* the error computation uses 0/1 accuracy, not log loss

This PR updates LogLoss.
It also adds some doc for boosting and forests.

I tested it on sample data and made sure the log loss is monotonically decreasing with each boosting iteration.

CC: mengxr manishamde codedeft

Author: Joseph K. Bradley <joseph@databricks.com>

Closes apache#3439 from jkbradley/gbt-loss-fix and squashes the following commits:

cfec17e [Joseph K. Bradley] removed forgotten temp comments
a27eb6d [Joseph K. Bradley] corrections to last log loss commit
ed5da2c [Joseph K. Bradley] updated LogLoss (boosting) for numerical stability
5e52bff [Joseph K. Bradley] * Removed the 1/2 from SquaredError.  This also required updating the test suite since it effectively doubles the gradient and loss. * Added doc for developers within RandomForest. * Small cleanup in test suite (generating data only once)
e57897a [Joseph K. Bradley] Fixed LogLoss for GradientBoostedTrees, and updated doc for losses, forests, and boosting

(cherry picked from commit c251fd7)
Signed-off-by: Xiangrui Meng <meng@databricks.com>
@SparkQA
Copy link

SparkQA commented Nov 26, 2014

Test build #23883 has started for PR 3474 at commit 8797bbb.

  • This patch merges cleanly.

@SparkQA
Copy link

SparkQA commented Nov 26, 2014

Test build #23883 has finished for PR 3474 at commit 8797bbb.

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

@AmplabJenkins
Copy link

Test PASSed.
Refer to this link for build results (access rights to CI server needed):
https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/23883/
Test PASSed.

@SparkQA
Copy link

SparkQA commented Nov 26, 2014

Test build #23890 has started for PR 3474 at commit aca2abb.

  • This patch merges cleanly.

@SparkQA
Copy link

SparkQA commented Nov 26, 2014

Test build #23890 has finished for PR 3474 at commit aca2abb.

  • This patch fails Spark unit tests.
  • This patch merges cleanly.
  • This patch adds no public classes.

@AmplabJenkins
Copy link

Test FAILed.
Refer to this link for build results (access rights to CI server needed):
https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/23890/
Test FAILed.

@mengxr
Copy link
Contributor Author

mengxr commented Nov 26, 2014

test this please

@SparkQA
Copy link

SparkQA commented Nov 26, 2014

Test build #23897 has started for PR 3474 at commit aca2abb.

  • This patch merges cleanly.

@SparkQA
Copy link

SparkQA commented Nov 26, 2014

Test build #23897 has finished for PR 3474 at commit aca2abb.

  • This patch fails Spark unit tests.
  • This patch merges cleanly.
  • This patch adds no public classes.

@AmplabJenkins
Copy link

Test FAILed.
Refer to this link for build results (access rights to CI server needed):
https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/23897/
Test FAILed.

@mengxr
Copy link
Contributor Author

mengxr commented Nov 26, 2014

@tdas I saw this test also failed in branch-1.2:

org.apache.spark.streaming.CheckpointSuite.recovery with file input stream

test this please

@mengxr
Copy link
Contributor Author

mengxr commented Nov 26, 2014

test this please

@SparkQA
Copy link

SparkQA commented Nov 26, 2014

Test build #23904 has started for PR 3474 at commit aca2abb.

  • This patch merges cleanly.

@SparkQA
Copy link

SparkQA commented Nov 26, 2014

Test build #23904 has finished for PR 3474 at commit aca2abb.

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

@AmplabJenkins
Copy link

Test PASSed.
Refer to this link for build results (access rights to CI server needed):
https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/23904/
Test PASSed.

asfgit pushed a commit that referenced this pull request Nov 26, 2014
@jkbradley @mengxr
[BRANCH-1.2][SPARK-4583][MLLIB] LogLoss for GradientBoostedTrees fix …
69550f7 
…+ doc updates

We reverted #3439 in branch-1.2 due to missing `import o.a.s.SparkContext._`, which is no longer needed in master (#3262). This PR adds #3439 back to branch-1.2 with correct imports.

Github is out-of-sync now. The real changes are the last two commits.

Author: Joseph K. Bradley <joseph@databricks.com>
Author: Xiangrui Meng <meng@databricks.com>

Closes #3474 from mengxr/SPARK-4583-1.2 and squashes the following commits:

aca2abb [Xiangrui Meng] add import o.a.s.SparkContext._ for v1.2
6b5564a [Joseph K. Bradley] [SPARK-4583] [mllib] LogLoss for GradientBoostedTrees fix + doc updates
@mengxr
Copy link
Contributor Author

mengxr commented Nov 26, 2014

Merged into branch-1.2.

@mengxr mengxr closed this Nov 26, 2014
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.
4 participants
@mengxr @SparkQA @AmplabJenkins @jkbradley