check_random_state

sklearn.utils.check_random_state(seed)

Turn seed into a np.random.RandomState instance.

Parameters:

seedNone, int or instance of RandomState

If seed is None, return the RandomState singleton used by np.random. If seed is an int, return a new RandomState instance seeded with seed. If seed is already a RandomState instance, return it. Otherwise raise ValueError.

Returns:

numpy.random.RandomState

The random state object based on seed parameter.

Examples

>>> from sklearn.utils.validation import check_random_state
>>> check_random_state(42)
RandomState(MT19937) at 0x...

Gallery examples

Empirical evaluation of the impact of k-means initialization

Empirical evaluation of the impact of k-means initialization

MNIST classification using multinomial logistic + L1

MNIST classification using multinomial logistic + L1

Manifold Learning methods on a severed sphere

Manifold Learning methods on a severed sphere

Face completion with a multi-output estimators

Face completion with a multi-output estimators

Isotonic Regression

Isotonic Regression