def convert(search_space, vals):
assignments = []
for param in search_space.params:
if param.type == INTEGER:
assignments.append(Assignment(param.name, int(vals[param.name][0])))
elif param.type == DOUBLE:
assignments.append(Assignment(param.name, vals[param.name][0]))
elif param.type == CATEGORICAL or param.type == DISCRETE:
assignments.append(
Assignment(param.name, param.list[vals[param.name][0]]))
return assignments
def convert(search_space, skopt_suggested):
assignments = []
for i in range(len(search_space.params)):
param = search_space.params[i]
if param.type == INTEGER:
assignments.append(Assignment(param.name, skopt_suggested[i]))
elif param.type == DOUBLE:
assignments.append(Assignment(param.name, skopt_suggested[i]))
elif param.type == CATEGORICAL or param.type == DISCRETE:
assignments.append(Assignment(param.name, skopt_suggested[i]))
return assignments
def convert(search_space, chocolate_params):
assignments = []
for param in search_space.params:
key = BaseChocolateService.encode(param.name)
if param.type == INTEGER:
assignments.append(
Assignment(param.name, chocolate_params[key]))
elif param.type == DOUBLE:
assignments.append(
Assignment(param.name, chocolate_params[key]))
elif param.type == CATEGORICAL or param.type == DISCRETE:
assignments.append(
Assignment(param.name, param.list[chocolate_params[key]]))
return assignments
def GetSuggestions(self, request, context):
"""
Main function to provide suggestion.
"""
algorithm_name, config = OptimizerConfiguration.convertAlgorithmSpec(
request.experiment.spec.algorithm)
if algorithm_name != "bayesianoptimization":
raise Exception(
"Failed to create the algorithm: {}".format(algorithm_name))
if self.is_first_run:
search_space = HyperParameterSearchSpace.convert(
request.experiment)
self.base_service = BaseSkoptService(
base_estimator=config.base_estimator,
n_initial_points=config.n_initial_points,
acq_func=config.acq_func,
acq_optimizer=config.acq_optimizer,
random_state=config.random_state,
search_space=search_space)
self.is_first_run = False
trials = Trial.convert(request.trials)
new_trials = self.base_service.getSuggestions(trials,
request.request_number)
return api_pb2.GetSuggestionsReply(
parameter_assignments=Assignment.generate(new_trials))
def _ask(self, request_number):
list_of_assignments = []
for _ in range(request_number):
optuna_trial = self.study.ask(fixed_distributions=self._get_optuna_search_space())
assignments = [Assignment(k, v) for k, v in optuna_trial.params.items()]
list_of_assignments.append(assignments)
assignments_key = self._get_assignments_key(assignments)
self.assignments_to_optuna_number[assignments_key].append(optuna_trial.number)
return list_of_assignments
def GetSuggestions(self, request, context):
"""
Main function to provide suggestion.
"""
with self.lock:
if self.study is None:
self.search_space = HyperParameterSearchSpace.convert(request.experiment)
self.study = self._create_study(request.experiment.spec.algorithm, self.search_space)
trials = Trial.convert(request.trials)
if len(trials) != 0:
self._tell(trials)
list_of_assignments = self._ask(request.request_number)
return api_pb2.GetSuggestionsReply(
parameter_assignments=Assignment.generate(list_of_assignments)
)
def GetSuggestions(self, request, context):
"""
Main function to provide suggestion.
"""
if self.is_first_run:
search_space = HyperParameterSearchSpace.convert(
request.experiment)
self.base_service = BaseChocolateService(
algorithm_name=request.experiment.spec.algorithm.
algorithm_name,
search_space=search_space)
self.is_first_run = False
trials = Trial.convert(request.trials)
new_assignments = self.base_service.getSuggestions(
trials, request.request_number)
return api_pb2.GetSuggestionsReply(
parameter_assignments=Assignment.generate(new_assignments))
def GetSuggestions(self, request, context):
"""
Main function to provide suggestion.
"""
name, config = OptimizerConfiguration.convert_algorithm_spec(
request.experiment.spec.algorithm)
if self.is_first_run:
search_space = HyperParameterSearchSpace.convert(
request.experiment)
self.base_service = BaseHyperoptService(algorithm_name=name,
algorithm_conf=config,
search_space=search_space)
self.is_first_run = False
trials = Trial.convert(request.trials)
new_assignments = self.base_service.getSuggestions(
trials, request.current_request_number)
return api_pb2.GetSuggestionsReply(
parameter_assignments=Assignment.generate(new_assignments))
