evalml.tuners.RandomSearchTuner

class evalml.tuners.RandomSearchTuner(space, random_state=0, with_replacement=False, replacement_max_attempts=10)[source]

Random Search Optimizer

Example

>>> tuner = RandomSearchTuner([(1,10), ['A', 'B']], random_state=0)
>>> print(tuner.propose())
(6, 'B')
>>> print(tuner.propose())
(4, 'B')
>>> print(tuner.propose())
(5, 'A')

Methods

__init__

Sets up check for duplication if needed.

add

Not applicable to random search tuner as generated parameters are not dependent on scores of previous parameters.

is_search_space_exhausted

Checks if it is possible to generate a set of valid parameters.

propose

Generate a unique set of parameters.