Hello @glemaitre! Thanks for updating the PR.

• There are no PEP8 issues in the file examples/ensemble/plot_comparison_ensemble_classifier.py !

• There are no PEP8 issues in the file imblearn/ensemble/__init__.py !

• There are no PEP8 issues in the file imblearn/ensemble/_bagging.py !

• There are no PEP8 issues in the file imblearn/ensemble/_forest.py !

• There are no PEP8 issues in the file imblearn/ensemble/tests/test_forest.py !

• There are no PEP8 issues in the file imblearn/under_sampling/_prototype_selection/_random_under_sampler.py !

• There are no PEP8 issues in the file imblearn/utils/_validation.py !

Comment last updated on September 06, 2018 at 12:29 Hours UTC

@chkoar I made a quick implementation of a balanced random forest classifier.
I tried to keep the changes minimal. The issue is that most of the code rely that the base estimators are trees, calling some private functions. Therefore, we cannot easily use pipeline as in the bagging case.

If you could have a look at it. It would be nice to have a second opinion.

Note that this can work only with the release 0.20 which is the reason for the failing.

We don't implement this via Bagging in order to get feature importances out of the box, right?

There's some issue with the init. I'll check it out

Codecov Report

Merging #459 into master will increase coverage by <.01%.
The diff coverage is 98.96%.

@@            Coverage Diff            @@
##           master    #459      +/-   ##
=========================================
+ Coverage   98.69%   98.7%   +<.01%     
=========================================
  Files          75      77       +2     
  Lines        4538    4720     +182     
=========================================
+ Hits         4479    4659     +180     
- Misses         59      61       +2

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 839df67...a21418b. Read the comment docs.