JustEachGroup

class mastml.legos.data_splitters.JustEachGroup

Bases: sklearn.base.BaseEstimator, sklearn.base.TransformerMixin

Class to train the model on one group at a time and test it on the rest of the data This class wraps scikit-learn’s LeavePGroupsOut with P set to n-1. More information is available at: http://scikit-learn.org/stable/modules/generated/sklearn.model_selection.LeavePGroupsOut.html

Args:

None (only object instance)

Methods:

get_n_splits: method to calculate the number of splits to perform

Args:

groups: (numpy array), array of group labels

Returns:

(int), number of unique groups, indicating number of splits to perform

split: method to perform split into train indices and test indices

Args:

X: (numpy array), array of X features y: (numpy array), array of y data groups: (numpy array), array of group labels

Returns:

(numpy array), array of train and test indices

Methods Summary

get_n_splits([X, y, groups])
split(X, y, groups)

Methods Documentation

get_n_splits(X=None, y=None, groups=None)

split(X, y, groups)