boax.experiments.optimization

Contents

boax.experiments.optimization#

boax.experiments.optimization(parameters, *, seed=0, num_warm_up_steps=12, batch_size=1, acquisition=None, surrogate=None)#

Setup for a bayesian optimization experiment.

Example

>>> experiment = optimization([{'name': 'x', 'type': 'range', 'bounds': [0.0, 1.0]}])
Parameters:

  • parameters (list[dict[str, Any]]) – List of parameters to be optimized via bayesian optimization. Each parameter is described by a dictionary with a ‘name’, a ‘type’ of either fixed, range, or log_range, and a ‘value’ for fixed parameters, or ‘bounds’ for range and log_range parameters.

  • seed (int) – The initial random seed.

Return type:

Trial[State[TypeVar(T)]]

Returns:

A trail object with next and best functions.

Raises:

ValueError – If given parameters cannot be parsed or don’t match requirements.

Contents