9.13.2. sklearn.cross_validation.LeavePOut

class sklearn.cross_validation.LeavePOut(n, p, indices=False)

Leave-P-Out cross validation iterator

Provides train/test indices to split data in train test sets. The test set is built using p samples while the remaining samples form the training set.

Due to the high number of iterations which grows with the number of samples this cross validation method can be very costly. For large datasets one should favor KFold, StratifiedKFold or ShuffleSplit.

Parameters :

n: int :

Total number of elements

p: int :

Size of the test sets

indices: boolean, optional (default False) :

Return train/test split with integer indices or boolean mask. Integer indices are useful when dealing with sparse matrices that cannot be indexed by boolean masks.

Examples

>>> from sklearn import cross_validation
>>> X = np.array([[1, 2], [3, 4], [5, 6], [7, 8]])
>>> y = np.array([1, 2, 3, 4])
>>> lpo = cross_validation.LeavePOut(4, 2)
>>> len(lpo)
6
>>> print lpo
sklearn.cross_validation.LeavePOut(n=4, p=2)
>>> for train_index, test_index in lpo:
...    print "TRAIN:", train_index, "TEST:", test_index
...    X_train, X_test = X[train_index], X[test_index]
...    y_train, y_test = y[train_index], y[test_index]
TRAIN: [False False  True  True] TEST: [ True  True False False]
TRAIN: [False  True False  True] TEST: [ True False  True False]
TRAIN: [False  True  True False] TEST: [ True False False  True]
TRAIN: [ True False False  True] TEST: [False  True  True False]
TRAIN: [ True False  True False] TEST: [False  True False  True]
TRAIN: [ True  True False False] TEST: [False False  True  True]

__init__(n, p, indices=False)