-
Notifications
You must be signed in to change notification settings - Fork 28.9k
[Spark-8092][ml] Allow OneVsRest Classifier feature and label column names to be configurable. #6631
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.
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
Conversation
|
@jkbradley please review. i didn't realize PredictorParams had no setters and setLabelCol, etc were in Predictor which OneVsRest does not extend. This means OneVsRest needs to expose setters for label, features etc that the underlying classifier needs. |
|
Test build #34143 has finished for PR 6631 at commit
|
|
This LGTM. However, after reading this I feel that |
|
Sorry for the delay! Will review now. Jenkins test this please |
|
I guess there are merge conflicts...but will comment anyways |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Is this saying that the base classifier should output "predictionCol"? It probably should send output to a temp col, to be removed.
|
@harsha2010 We should document clearly that the input and output columns of the base classifier are ignored. |
|
Test build #36433 has finished for PR 6631 at commit
|
|
Ping : ) |
|
oops, thanks for reminding me...will update the merged patch tonight |
|
Test build #38041 has finished for PR 6631 at commit
|
|
Could you please add doc to the "classifier" Param indicating that the base classifier input and output columns are ignored in favor of the ones specified in OneVsRest? Other than that, it looks good. |
|
Test build #38178 has finished for PR 6631 at commit
|
|
LGTM Would you mind adding a description before we merge this, e.g., to say that the OneVsRest input/output col params override the base classifier ones? Thanks! |
|
@jkbradley , sure, do you mean in the Spark docs? The description is already in the code docs for the params of OneVsRest. |
|
Oh, no, I mean the PR description (the first comment), to become part of the commit message. |
|
@jkbradley , thanks., added the documentation to the first comment in this PR |
|
Thanks! I'll merge this with master |
The base classifier input and output columns are ignored in favor of the ones specified in OneVsRest.