This pr is ready for review.

This is so funny, @arjoly are you reviewing your own pull request? :)

Yep ;-)

@rsivapr ENH means enhancement(s)

@arjoly That was quick! :) I literally deleted that post within a second when I realized it must mean that.

I received an email when you post a message on my pull request. ;-)

(ping @glouppe )

My first question would be: does this ever work? And can we get an example of this vs OVR with one dataset where OVR works and one where this works?
Also, I think we should warn the user that this can only produce label combinations that actually exist in the trainingset (making your remark about overfitting a bit more explicit maybe?)

Otherwise this looks good, good job :)

I am not entirely happy with the testing as the real use case is only tested via a hard-coded result. I think I would like it best if the transformation would be done by hand there for a small problem and in an obvious way and compare against the estimator. But maybe that is overkill. wdyt?

@amueller @jnothman Thanks for the review !!! I will try to find some time to work on all your comments.

My first question would be: does this ever work? And can we get an example of this vs OVR with one dataset where OVR works and one where this works?

Yes, this works. For instance on the yeast dataset, the lps meta-estimator shine on several metrics compare to ovr

{'hamming_loss': {'dummy': 0.23298021498675806,
                  'lps svm': 0.25775042841564105,
                  'ova svm': 0.23298021498675806},
 'jaccard': {'dummy': 0.33653822852296145,
             'lps svm': 0.43881512586528965,
             'ova svm': 0.33653822852296145},
 'macro-f1': {'dummy': 0.12221934801958166,
              'lps svm': 0.23575486447032259,
              'ova svm': 0.12221934801958166},
 'micro-f1': {'dummy': 0.47828362114076395,
              'lps svm': 0.56270648870093831,
              'ova svm': 0.47828362114076395},
 'samples-f1': {'dummy': 0.45689163011954953,
                'lps svm': 0.547173739867307,
                'ova svm': 0.45689163011954953},
 'subset_accuracy': {'dummy': 0.017448200654307525,
                     'lps svm': 0.14612868047982552,
                     'ova svm': 0.017448200654307525},
 'weighted-f1': {'dummy': 0.30083303670803041,
                 'lps svm': 0.43848139536413128,
                 'ova svm': 0.30083303670803041}}

Should I add the script to the examples?

I am not entirely happy with the testing as the real use case is only tested via a hard-coded result. I think I would like it best if the transformation would be done by hand there for a small problem and in an obvious way and compare against the estimator. But maybe that is overkill. wdyt?

Do you suggest to create a LabelPowerSetTransformer?

Hm strange, commons tests are failing and do not detect that this is a meta estimator.

rebase on top of master

Coverage increased (+0.01%) when pulling b5f1a9c on arjoly:labelpowerset into 0807e19 on scikit-learn:master.

@amueller and @vene Is it good for you?

Writeup of our discussion:

• add a test for the zero-label class being handled correctly in predict_proba
• marginalize to get p(label | x) (btw how would this relate with what OvR gets?

Apart from this lgtm, 👍

marginalize to get p(label | x) (btw how would this relate with what OvR gets?

There is an api discepancy for that classes.

see #2451 for more information

Isn't there some way of renormalizing the output of OvR to be comparable?

Isn't there some way of renormalizing the output of OvR to be comparable?

I answer to you totally off apparently. Yes, there is a way. The one we discussed today. I am working.

I need to add tests whenever some label set are missing, i.e. label set presented at fit are not the same whenver you predict.

LabelPowerSetClassifier doesn't have a classes_ attribute. Should it have one?

LabelPowerSetClassifier doesn't have a classes_ attribute. Should it have one?

yes, it should have one

Switch from MRG to WIP since I am progressing slowly on this.

Hi everyone, this seems to be almost in a completed state but not merged yet. Was there any reason for doing so ? I would also like to ask the same about classifier chain algorithm in #3727 and MultiOutput Bagging in #4848. I would like to complete these PRs in case no one else is currently working on these and if it is okay. Thanks !

@maniteja123 I think you can feel free to take this over if you like, I think @arjoly is busy.

given the lack of requests for this, I'd suggest moving this to scikit-learn-extras.

given the lack of requests for this, I'd suggest moving this to scikit-learn-extras.

I agree. Should we close this and create an issue there?