E5D3 WIP: Classes support for all classifiers by erg · Pull Request #1304 · scikit-learn/scikit-learn · GitHub

WIP: Classes support for all classifiers#1304

Closed

erg wants to merge 11 commits intoscikit-learn:masterfrom

Commits on Nov 15, 2012

base: Add code to find unique classes_ given classes or y.
erg
committed
lda: Allow classes to be set in __init__. Add y_inverse_ instead of overwriting y.
erg
committed
base: Check that we expect all the y classes in fit().
erg
committed
naive_bayes: Add support for classes in __init__. Update unit tests. Fix bug where class_prior can't be a numpy array.
erg
committed
preprocessing: Add support for classes to LabelBinarizer. Needs test updates to be sure it's correct.
erg
committed
tree: Add support for classes= parameter for DecisionTreeClassifiers. Move _is_multilabel from preprocessing to base, and rename it to is_multilabel.
erg
committed
tree: Add support for classes= in ExtraTreeClassifier.
erg
committed
ensemble: Adding classes= parameter to ensemble.
erg
committed
base: Rename _is_label_indicator_matrix to is_label_indicator_matrix and use in preprocessing. Use prepare_classes() and have it decide between single or multilabel.
erg
committed
ensemble: Use classes in forest if provided.
erg
committed
base: Implement is_iterable for better error reporting on bad class shapes.
erg
committed

0