Skip to content

[MRG] EHN add note to create balanced RF #373

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.

Sign up for GitHub

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

Merged
merged 1 commit into from
Nov 24, 2017
Merged

[MRG] EHN add note to create balanced RF #373

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
24 changes: 20 additions & 4 deletions doc/ensemble.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -100,13 +100,29 @@ takes the same parameters than the scikit-learn
... ratio='auto',
... replacement=False,
... random_state=0)
>>> bbc.fit(X, y) # doctest: +ELLIPSIS
>>> bbc.fit(X_train, y_train) # doctest: +ELLIPSIS
BalancedBaggingClassifier(...)
>>> y_pred = bbc.predict(X_test)
>>> confusion_matrix(y_test, y_pred)
array([[ 12, 0, 0],
[ 1, 54, 4],
[ 49, 53, 1077]])
array([[ 9, 1, 2],
[ 0, 55, 4],
[ 42, 46, 1091]])

It also possible to turn a balanced bagging classifier into a balanced random
forest using a decision tree classifier and setting the parameter
``max_features='auto'`. It allows to randomly select a subset of features for
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

missing backtick after auto

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks Andy

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks for adding the example, I think it's very helpful! (I just linked my students to it ;)

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We will try to find sometime to check the sickit-learn PR to push it further

each tree::

>>> brf = BalancedBaggingClassifier(
... base_estimator=DecisionTreeClassifier(max_features='auto'),
... random_state=0)
>>> brf.fit(X_train, y_train) # doctest: +ELLIPSIS
BalancedBaggingClassifier(...)
>>> y_pred = brf.predict(X_test)
>>> confusion_matrix(y_test, y_pred)
array([[ 9, 1, 2],
[ 0, 54, 5],
[ 31, 34, 1114]])

See
:ref:`sphx_glr_auto_examples_ensemble_plot_comparison_bagging_classifier.py`.
6 changes: 6 additions & 0 deletions doc/whats_new/v0.0.4.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,12 @@ Version 0.4 (under development)
Changelog
---------

Enhancement
...........

- Add a documentation node to create a balanced random forest from a balanced
bagging classifier. :issue:`372` by :user:`Guillaume Lemaitre <glemaitre>`.

Bug fixes
.........

Expand Down
22 changes: 22 additions & 0 deletions examples/ensemble/plot_comparison_bagging_classifier.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -26,6 +26,7 @@

from sklearn.model_selection import train_test_split
from sklearn.ensemble import BaggingClassifier
from sklearn.tree import DecisionTreeClassifier
from sklearn.metrics import confusion_matrix

from imblearn.datasets import fetch_datasets
Expand Down Expand Up @@ -99,4 +100,25 @@ def plot_confusion_matrix(cm, classes,
plot_confusion_matrix(cm_balanced_bagging, classes=np.unique(ozone.target),
title='Confusion matrix using BalancedBaggingClassifier')

###############################################################################
# Turning the balanced bagging classifier into a balanced random forest
###############################################################################
# It is possible to turn the ``BalancedBaggingClassifier`` into a balanced
# random forest by using a ``DecisionTreeClassifier`` with
# ``max_features='auto'``. We illustrate such changes below.

balanced_random_forest = BalancedBaggingClassifier(
base_estimator=DecisionTreeClassifier(max_features='auto'),
random_state=0)

balanced_random_forest.fit(X_train, y_train)
print('Classification results using a balanced random forest classifier on'
' imbalanced data')
y_pred_balanced_rf = balanced_random_forest.predict(X_test)
print(classification_report_imbalanced(y_test, y_pred_balanced_rf))
cm_bagging = confusion_matrix(y_test, y_pred_balanced_rf)
plt.figure()
plot_confusion_matrix(cm_bagging, classes=np.unique(ozone.target),
title='Confusion matrix using balanced random forest')

plt.show()
8 changes: 6 additions & 2 deletions imblearn/ensemble/classifier.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -8,12 +8,10 @@

import numpy as np

import sklearn
from sklearn.base import clone
from sklearn.ensemble import BaggingClassifier
from sklearn.tree import DecisionTreeClassifier
from sklearn.ensemble.bagging import _generate_bagging_indices
from sklearn.utils import indices_to_mask

from ..pipeline import Pipeline
from ..under_sampling import RandomUnderSampler
Expand Down Expand Up @@ -136,6 +134,9 @@ class BalancedBaggingClassifier(BaggingClassifier):

Notes
-----
This is possible to turn this classifier into a balanced random forest [5]_
by passing a :class:`sklearn.tree.DecisionTreeClassifier` with
`max_features='auto'` as a base estimator.

See
:ref:`sphx_glr_auto_examples_ensemble_plot_comparison_bagging_classifier.py`.
Expand All @@ -155,6 +156,9 @@ class BalancedBaggingClassifier(BaggingClassifier):
1998.
.. [4] G. Louppe and P. Geurts, "Ensembles on Random Patches", Machine
Learning and Knowledge Discovery in Databases, 346-361, 2012.
.. [5] Chen, Chao, Andy Liaw, and Leo Breiman. "Using random forest to
learn imbalanced data." University of California, Berkeley 110,
2004.

Examples
--------
Expand Down