def test_create_with_validation_dataset(self):
#Create experiment with validation with dataset.
# add vald dataset
cols = ['sepal length', 'sepal width', 'petal length', 'petal width']
target = 'class'
df = pd.read_csv('tests/data/test_1_vald.csv')
dc = DatasetClient(self.project.hid)
vald_dataset = dc.add_dataset_if_not_exists(df[cols], df[target])
# add experiment
ec = ExperimentClient(self.project.hid)
self.assertNotEqual(ec, None)
# there should be none experiments
experiments = ec.get_experiments()
self.assertEqual(experiments, [])
# create new experiment
experiment = ec.add_experiment_if_not_exists(
self.dataset, vald_dataset, self.expt_title, self.project.task,
self.validation_kfolds, self.validation_shuffle,
self.validation_stratify, 0.72, self.algorithms, self.metric,
self.tuning_mode, self.time_constraint, self.create_enseble)
self.assertNotEqual(experiment, None)
self.assertEqual(experiment.title, self.expt_title)
self.assertEqual(experiment.validation_scheme, "With dataset")
def test_add_existing_dataset(self):
# setup dataset client
dc = DatasetClient(self.project.hid)
self.assertNotEqual(dc, None)
# get initial number of datasets
init_datasets_cnt = len(dc.get_datasets())
# add dataset
dc.add_dataset_if_not_exists(self.X, self.y)
# get datasets
datasets = dc.get_datasets()
self.assertEqual(len(datasets), init_datasets_cnt + 1)
# add the same dataset
# it shouldn't be added
dc.add_dataset_if_not_exists(self.X, self.y)
# number of all datasets in project should be 1
datasets = dc.get_datasets()
self.assertEqual(len(datasets), init_datasets_cnt + 1)
def test_create_and_delete(self):
# setup dataset client
dc = DatasetClient(self.project.hid)
self.assertNotEqual(dc, None)
# get initial number of datasets
init_datasets_cnt = len(dc.get_datasets())
# add dataset
my_dataset_1 = dc.add_dataset_if_not_exists(self.X, self.y)
my_dataset_2 = dc.add_dataset_if_not_exists(self.X, y = None)
# get datasets
datasets = dc.get_datasets()
self.assertEqual(len(datasets), init_datasets_cnt+2)
# delete added dataset
dc.delete_dataset(my_dataset_1.hid)
# check number of datasets
datasets = dc.get_datasets()
self.assertEqual(len(datasets), init_datasets_cnt+1)
def test_add_dataset_for_prediction(self):
# setup dataset client
dc = DatasetClient(self.project.hid)
self.assertNotEqual(dc, None)
# get datasets, there should be none
datasets = dc.get_datasets()
self.assertEqual(len(datasets), 0)
# add dataset
my_dataset = dc.add_dataset_if_not_exists(self.X, None)
self.assertNotEqual(my_dataset, None)
# get datasets
datasets = dc.get_datasets()
self.assertEqual(len(datasets), 1)
my_dataset_2 = dc.get_dataset(my_dataset.hid)
self.assertEqual(my_dataset.hid, my_dataset_2.hid)
self.assertEqual(my_dataset.title, my_dataset_2.title)
def test_add_dataset_for_training(self):
# setup dataset client
dc = DatasetClient(self.project.hid)
self.assertNotEqual(dc, None)
# get datasets, there should be none
datasets = dc.get_datasets()
self.assertEqual(len(datasets), 0)
# add dataset
my_dataset = dc.add_dataset_if_not_exists(self.X, self.y)
self.assertNotEqual(my_dataset, None)
# get datasets
datasets = dc.get_datasets()
self.assertEqual(len(datasets), 1)
my_dataset_2 = dc.get_dataset(my_dataset.hid)
self.assertEqual(my_dataset.hid, my_dataset_2.hid)
self.assertEqual(my_dataset.title, my_dataset_2.title)
# test __str__ method
self.assertTrue('id' in str(my_dataset_2))
self.assertTrue('title' in str(my_dataset_2))
self.assertTrue('file' in str(my_dataset_2))
def setUp(self):
proj_title = 'Test project-01' + get_postfix()
proj_task = 'bin_class'
self.expt_title = 'Test experiment-01'
self.validation_kfolds = 5
self.validation_shuffle = True
self.validation_stratify = True
self.validation_train_split = None
self.algorithms = ['xgb']
self.metric = 'logloss'
self.tuning_mode = 'Normal'
self.time_constraint = 1
self.create_enseble = False
# setup project
self.project_client = ProjectClient()
self.project = self.project_client.create_project(title=proj_title,
task=proj_task)
# add training data
df = pd.read_csv('tests/data/test_1.csv')
cols = ['sepal length', 'sepal width', 'petal length', 'petal width']
target = 'class'
dc = DatasetClient(self.project.hid)
self.dataset = dc.add_dataset_if_not_exists(df[cols], df[target])