How to use custom classifiers in ensemble classifiers in sklearn?

Question

How to use custom classifiers in ensemble classifiers in sklearn?

I read that the built-in ensemble methods in sklearn use decision trees as basic classifiers. Can I use custom classifiers instead?

+5

scikit-learn machine-learning nlp scikits

Karthick May 09 '12 at 14:39

source share

3 answers

sklearn.ensemble.AdaBoostClassifier, : scikit-learn.org/stable/modules/generated/sklearn.ensemble.AdaBoostClassifier.html base_estimator .

+3

andy 08 '14 23:31

I don't know if this helps, but you can easily compile / combine custom classifiers using the Pipeline utilities: http://scikit-learn.org/stable/tutorial/statistical_inference/putting_together.html#pipelining

+2

user1151446 May 22 '13 at 12:33

source share

Fred foo · Accepted Answer · 2012-05-09T16:10:13+0000

If you mean random forest classes, then no, this is currently not possible. The ability to allow other grades was discussed on the scikit-learn mailing list last January, but I don't think any actual code came out of this discussion.

How to use custom classifiers in ensemble classifiers in sklearn?

More articles: