def setUp(self):
study_configuration_json = {
"goal":
"MAXIMIZE",
"maxTrials":
5,
"maxParallelTrials":
1,
"randomInitTrials":
1,
"params": [{
"parameterName": "hidden2",
"type": "DISCRETE",
"feasiblePoints": "8, 16, 32, 64",
"scalingType": "LINEAR"
}, {
"parameterName": "optimizer",
"type": "CATEGORICAL",
"feasiblePoints": "sgd, adagrad, adam, ftrl",
"scalingType": "LINEAR"
}]
}
study_configuration = json.dumps(study_configuration_json)
self.study = Study.create("ChocolateGridSearchStudy",
study_configuration)
def setUp(self):
study_configuration_json = {
"goal":
"MAXIMIZE",
"maxTrials":
5,
"maxParallelTrials":
1,
"params": [{
"parameterName": "hidden1",
"type": "INTEGER",
"minValue": 40,
"maxValue": 400,
"scallingType": "LINEAR"
}]
}
study_configuration = json.dumps(study_configuration_json)
self.study = Study.create("RandomSearchStudy", study_configuration)
trial1 = Trial.create(self.study.id, "RandomSearchTrial1")
trial2 = Trial.create(self.study.id, "RandomSearchTrial2")
self.trials = [trial1, trial2]
TrialMetric.create(trial1.id, 10, 0.5)
TrialMetric.create(trial1.id, 20, 0.6)
TrialMetric.create(trial2.id, 10, 0.6)
TrialMetric.create(trial2.id, 20, 0.5)
def create_study(self,
study_name,
study_configuration,
algorithm="BayesianOptimization"):
study = Study.create(study_name, study_configuration, algorithm)
return study
def setUp(self):
study_configuration_json = {
"goal":
"MAXIMIZE",
"maxTrials":
5,
"maxParallelTrials":
1,
"randomInitTrials":
1,
"params": [{
"parameterName": "l1_normalization",
"type": "DOUBLE",
"minValue": 0.01,
"maxValue": 0.99,
"scalingType": "LINEAR"
}, {
"parameterName": "learning_rate",
"type": "DOUBLE",
"minValue": 0.01,
"maxValue": 0.5,
"scalingType": "LINEAR"
}]
}
study_configuration = json.dumps(study_configuration_json)
self.study = Study.create("TpeStudy", study_configuration)
def setUp(self):
study_configuration_json = {
"goal":
"MAXIMIZE",
"maxTrials":
5,
"maxParallelTrials":
1,
"randomInitTrials":
1,
"params": [{
"parameterName": "hidden1",
"type": "INTEGER",
"minValue": 1,
"maxValue": 10,
"scalingType": "LINEAR"
}, {
"parameterName": "learning_rate",
"type": "DOUBLE",
"minValue": 0.01,
"maxValue": 0.5,
"scalingType": "LINEAR"
}]
}
study_configuration = json.dumps(study_configuration_json)
self.study = Study.create("SkoptBayesianOptimizationStudy",
study_configuration)
def setUp(self):
study_configuration_json = {
"goal":
"MAXIMIZE",
"maxTrials":
5,
"maxParallelTrials":
1,
"params": [{
"parameterName": "hidden1",
"type": "INTEGER",
"minValue": 40,
"maxValue": 400,
"scallingType": "LINEAR"
}]
}
study_configuration = json.dumps(study_configuration_json)
self.study = Study.create("RandomSearchStudy", study_configuration)
trial1 = Trial.create(self.study.id, "RandomSearchTrial1")
trial2 = Trial.create(self.study.id, "RandomSearchTrial2")
self.trials = [trial1, trial2]
TrialMetric.create(trial1.id, 10, 0.5)
TrialMetric.create(trial1.id, 20, 0.6)
TrialMetric.create(trial2.id, 10, 0.6)
TrialMetric.create(trial2.id, 20, 0.5)
def v1_studies(request):
# Create the study
if request.method == "POST":
data = json.loads(request.body)
name = data["name"]
study_configuration = json.dumps(data["study_configuration"])
algorithm = data.get("algorithm", "RandomSearchAlgorithm")
study = Study.create(name, study_configuration, algorithm)
return JsonResponse({"data": study.to_json()})
# List the studies
elif request.method == "GET":
studies = Study.objects.all()
response_data = [study.to_json() for study in studies]
return JsonResponse({"data": response_data})
else:
return JsonResponse({"error": "Unsupported http method"})
def setUp(self):
study_configuration_json = {
"goal":
"MAXIMIZE",
"maxTrials":
5,
"maxParallelTrials":
1,
"params": [{
"parameterName": "hidden1",
"type": "INTEGER",
"minValue": 40,
"maxValue": 400,
"scallingType": "LINEAR"
}]
}
study_configuration = json.dumps(study_configuration_json)
self.study = Study.create("GridSearchStudy", study_configuration)
self.trials = []
def setUp(self):
study_configuration_json = {
"goal":
"MAXIMIZE",
"maxTrials":
5,
"maxParallelTrials":
1,
"params": [{
"parameterName": "hidden1",
"type": "INTEGER",
"minValue": 40,
"maxValue": 400,
"scallingType": "LINEAR"
}]
}
study_configuration = json.dumps(study_configuration_json)
self.study = Study.create("GridSearchStudy", study_configuration)
self.trials = []
def setUp(self):
study_configuration_json = {
"goal":
"MAXIMIZE",
"maxTrials":
5,
"maxParallelTrials":
1,
"randomInitTrials":
1,
"params": [{
"parameterName": "hidden1",
"type": "INTEGER",
"minValue": 1,
"maxValue": 10,
"scalingType": "LINEAR"
}, {
"parameterName": "learning_rate",
"type": "DOUBLE",
"minValue": 0.01,
"maxValue": 0.5,
"scalingType": "LINEAR"
}, {
"parameterName": "hidden2",
"type": "DISCRETE",
"feasiblePoints": "8, 16, 32, 64",
"scalingType": "LINEAR"
}, {
"parameterName": "optimizer",
"type": "CATEGORICAL",
"feasiblePoints": "sgd, adagrad, adam, ftrl",
"scalingType": "LINEAR"
}, {
"parameterName": "batch_normalization",
"type": "CATEGORICAL",
"feasiblePoints": "true, false",
"scalingType": "LINEAR"
}]
}
study_configuration = json.dumps(study_configuration_json)
self.study = Study.create("RandomSearchStudy", study_configuration)