search

scikit-learn-contrib / imbalanced-learn

A Python Package to Tackle the Curse of Imbalanced Datasets in Machine Learning
https://imbalanced-learn.org
MIT License
6.85k stars 1.28k forks source link

BRFclassifier #926

Closed github844268529 closed 2 years ago

github844268529 commented 2 years ago

Is the balanced random forest first randomly undersampled and then bootstrapped?

Or do you first bootstrap it and then random undersampling?

I want to figure out what happens to the data in each decision tree.

I hope you can reply me in your busy schedule. Thank you very much

hayesall commented 2 years ago

Resample, then apply bootstrapping when fitting DecisionTreeClassifiers.

This step takes place in imblearn.ensemble._forest._local_parallel_build_trees:

https://github.com/scikit-learn-contrib/imbalanced-learn/blob/9f8830e13207ddf85596831593da0183b36cd1fb/imblearn/ensemble/_forest.py#L42-L73