Hi All,
I've tried GridsearchCV with RandomForestClassifier()
clf = GridSearchCV(RandomForestClassifier(), tuned_parameters, cv=5,
scoring='accuracy')
for a multi-label problem where the output is a list of lists of 20
zeros or ones.
[[1, 0, 1, 1, 1, 0, 0, 1, 1, 1, 1, 1, 1, 1, 0, 1, 0, 0, 1
for legacy reasons, multilabel targets need to be passed as an array (or a
sparse matrix if supported by the classifier). lists of lists are not
supported but may be in the near future.
___
scikit-learn mailing list
scikit-learn@python.org
https://mail.py
Joel, thank you. It helps. One step forward.
-Dmitry
2017-12-10 23:09 GMT+03:00 Joel Nothman :
> for legacy reasons, multilabel targets need to be passed as an array (or a
> sparse matrix if supported by the classifier). lists of lists are not
> supported but may be in the near future.
>
> _
