[MRG] EHN add BalancedBaggingClassifier

[glemaitre]

Reference Issue

closes #252
closes #149

What does this implement/fix? Explain your changes.

Implement EasyEnsemble as a meta-estimator. We call it BalancedBaggingClassifier since it is what it does.

Any other comments?

[codecov]

Codecov Report

❗ No coverage uploaded for pull request base (master@2e7c070). Click here to learn what that means.
The diff coverage is 100%.

@@ Coverage Diff @@ ## master #315 +/- ## ======================================== Coverage ? 98.1% ======================================== Files ? 68 Lines ? 4071 Branches ? 0 ======================================== Hits ? 3994 Misses ? 77 Partials ? 0

Impacted Files	Coverage Δ
imblearn/ensemble/balance_cascade.py	100% <ø> (ø)
imblearn/ensemble/easy_ensemble.py	100% <ø> (ø)
imblearn/ensemble/__init__.py	100% <100%> (ø)
imblearn/ensemble/tests/test_easy_ensemble.py	100% <100%> (ø)
imblearn/ensemble/tests/test_classifier.py	100% <100%> (ø)
imblearn/ensemble/classifier.py	100% <100%> (ø)

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 2e7c070...341cf95. Read the comment docs.

[glemaitre]

@massich @chkoar review time ;)
The test are the same than in scikit-learn with sample_weight which is not supported.

[glemaitre]

@chkoar I would be inclined to move the meta-estimator. I am not sure where indeed.
If we think about it, all predictors would probably some ensemble which use a specific sampling (SMOTEBoost, etc.) It might make sense to have a subpackage predictor with inside different module ensemble.py, smote.py ...

[massich]

despite the rebase and the nosetest asserts LGTM

[amueller]

sweet :)

[glemaitre]

We have for EuroScipy next week ;) We got in release mode chasing scikit-learn :)