def test_complete_and_get_new_suggestions(self):
tpeAlgorithm = TpeAlgorithm()
new_trials = tpeAlgorithm.get_new_suggestions(self.study.id, [], 1)
new_trials[0].status = "Completed"
new_trials[0].objective_value = 0.6
new_trials[0].save()
new_trials = tpeAlgorithm.get_new_suggestions(self.study.id, [], 1)
new_trials[0].status = "Completed"
new_trials[0].objective_value = 0.7
new_trials[0].save()
new_trials = tpeAlgorithm.get_new_suggestions(self.study.id, [], 3)
# Assert getting two trials
self.assertEqual(len(new_trials), 3)
# Assert getting the trials
new_trial = new_trials[0]
new_parameter_values = new_trial.parameter_values
new_parameter_values_json = json.loads(new_parameter_values)
self.assertTrue(
0.99 >= new_parameter_values_json["l1_normalization"] >= 0.01)
self.assertTrue(
0.5 >= new_parameter_values_json["learning_rate"] >= 0.01)
self.assertTrue(
new_parameter_values_json["hidden2"] in [8, 16, 32, 64])
self.assertTrue(new_parameter_values_json["optimizer"] in
["sgd", "adagrad", "adam", "ftrl"])
def test_get_multiple_new_suggestions(self):
tpeAlgorithm = TpeAlgorithm()
# Assert getting one trial
new_trials = tpeAlgorithm.get_new_suggestions(self.study.id, number=1)
self.assertEqual(len(new_trials), 1)
# Assert getting multiple trials
new_trials = tpeAlgorithm.get_new_suggestions(self.study.id, number=10)
self.assertEqual(len(new_trials), 10)
def test_get_new_suggestions(self):
tpeAlgorithm = TpeAlgorithm()
new_trials = tpeAlgorithm.get_new_suggestions(
self.study.id, number=1)
new_trial = new_trials[0]
new_parameter_values_json = json.loads(new_trial.parameter_values)
#self.assertTrue(0.99 >= new_parameter_values_json["l1_normalization"] >= 0.01)
#self.assertTrue(0.5 >= new_parameter_values_json["learning_rate"] >= 0.01)
self.assertTrue(new_parameter_values_json["l1_normalization"] >= 0.01)
self.assertTrue(new_parameter_values_json["learning_rate"] >= 0.01)
def test_get_new_suggestions(self):
tpeAlgorithm = TpeAlgorithm()
new_trials = tpeAlgorithm.get_new_suggestions(self.study.id, number=1)
new_trial = new_trials[0]
new_parameter_values_json = json.loads(new_trial.parameter_values)
self.assertTrue(
0.99 >= new_parameter_values_json["l1_normalization"] >= 0.01)
self.assertTrue(
0.5 >= new_parameter_values_json["learning_rate"] >= 0.01)
self.assertTrue(
new_parameter_values_json["hidden2"] in [8, 16, 32, 64])
self.assertTrue(new_parameter_values_json["optimizer"] in
["sgd", "adagrad", "adam", "ftrl"])