def test_single_valued_choice_to_fixed_param_conversion(self):
representation = {
"name": "test",
"type": "choice",
"values": [1.0],
}
output = parameter_from_json(representation)
self.assertIsInstance(output, FixedParameter)
self.assertEqual(output.value, 1.0)
def with_evaluation_function(
parameters: List[TParameterRepresentation],
evaluation_function: TEvaluationFunction,
experiment_name: Optional[str] = None,
objective_name: Optional[str] = None,
minimize: bool = False,
parameter_constraints: Optional[List[str]] = None,
outcome_constraints: Optional[List[str]] = None,
total_trials: int = 20,
arms_per_trial: int = 1,
wait_time: int = 0,
random_seed: Optional[int] = None,
generation_strategy: Optional[GenerationStrategy] = None,
) -> "OptimizationLoop":
"""Constructs a synchronous `OptimizationLoop` using an evaluation
function."""
exp_parameters = [parameter_from_json(p) for p in parameters]
parameter_map = {p.name: p for p in exp_parameters}
experiment = SimpleExperiment(
name=experiment_name,
search_space=SearchSpace(
parameters=exp_parameters,
parameter_constraints=None
if parameter_constraints is None else [
constraint_from_str(c, parameter_map)
for c in parameter_constraints
],
),
objective_name=objective_name,
evaluation_function=evaluation_function,
minimize=minimize,
outcome_constraints=[
outcome_constraint_from_str(c)
for c in (outcome_constraints or [])
],
)
return OptimizationLoop(
experiment=experiment,
total_trials=total_trials,
arms_per_trial=arms_per_trial,
random_seed=random_seed,
wait_time=wait_time,
generation_strategy=generation_strategy,
)
