def setUp(self):
from main import ProcessCLI
# Obtain from terminal log when run main.py runs `vars(parser.parse_args())`
args = {
'data_set': 4,
'logistic': True,
'kfold': False,
'linear': True,
'knn': 'scikit',
'training_features': [
'gpa',
'gre'
],
'target_feature': 'admit',
'multi_class_features': '',
'exclude_non_numeric': '',
'exclude_non_ordinal': '',
'exclude_out_of_scope': '',
'cleanse_price_format_features': '',
'convert_feature_words_to_digits': '',
'kmeans': True,
'kmeans_qty': 5,
'affiliation_feature': '',
'kfold_qty': 10,
'hyper_optim': True,
'hyper_optim_range': 20,
'suppress_all_plots': False
}
# Suppress all plots
args["suppress_all_plots"] = True
self.process_cli = ProcessCLI(**args)
self.event_mod = self.process_cli.map_cli_args_to_event_config(EVENT)
def setUp(self):
from main import ProcessCLI
# Obtain from terminal log when run main.py runs `vars(parser.parse_args())`
args = {
'data_set':
2,
'logistic':
False,
'kfold':
True,
'linear':
True,
'knn':
'scikit',
'training_features': [
'num-of-doors', 'curb-weight', 'horsepower', 'city-mpg',
'highway-mpg'
],
'target_feature':
'price',
'multi_class_features':
'',
'exclude_non_numeric': [
'make', 'fuel-type', 'aspiration', 'body-style',
'drive-wheels', 'engine-location', 'engine-type', 'fuel-system'
],
'exclude_non_ordinal':
'',
'exclude_out_of_scope': ['symboling', 'normalized-losses'],
'cleanse_price_format_features': ['price'],
'convert_feature_words_to_digits':
['num-of-doors', 'num-of-cylinders'],
'kmeans':
True,
'kmeans_qty':
5,
'affiliation_feature':
'',
'kfold_qty':
10,
'hyper_optim':
True,
'hyper_optim_range':
20,
'suppress_all_plots':
False
}
# Suppress all plots
args["suppress_all_plots"] = True
self.process_cli = ProcessCLI(**args)
self.event_mod = self.process_cli.map_cli_args_to_event_config(EVENT)
class MainExample4NonLogisticTestCase(unittest.TestCase):
"""Tests for `main.py`."""
def setUp(self):
from main import ProcessCLI
# Obtain from terminal log when run main.py runs `vars(parser.parse_args())`
args = {
'data_set': 6,
'logistic': False,
'kfold': True,
'linear': True,
'knn': 'scikit',
'training_features': [
'yearpublished',
'minplaytime',
'minage',
'total_wanters',
'average_weight'
],
'target_feature': 'average_rating',
'multi_class_features': '',
'exclude_non_numeric': [
'type,',
'name'
],
'exclude_non_ordinal': '',
'exclude_out_of_scope': [
'id'
],
'cleanse_price_format_features': '',
'convert_feature_words_to_digits': '',
'kmeans': True,
'kmeans_qty': 5,
'affiliation_feature': 'yearpublished',
'kfold_qty': 10,
'hyper_optim': True,
'hyper_optim_range': 20,
'suppress_all_plots': False
}
# Suppress all plots
args["suppress_all_plots"] = True
self.process_cli = ProcessCLI(**args)
self.event_mod = self.process_cli.map_cli_args_to_event_config(EVENT)
def tearDown(self):
del self.process_cli
del self.event_mod
def test_valid_config_responds_with_expected_results(self):
expected_result = {
'logistic': None,
'linear': {
'pre-hyperparameter_optimisation': {
'model_type': 'linear',
'rmse': 1.2042523903063078
},
'post-hyperparameter_optimisation': {
'model_type': 'linear',
'feature_names': 'yearpublished__minage__total_wanters__average_weight',
'rmse': 1.2096966125815165,
'k_neighbors_qty': 1,
'k_folds_qty': 10,
'k_fold_cross_validation_toggle': True
}
},
'knn': {
'feature_names': 'yearpublished__minplaytime__minage__total_wanters__average_weight',
'rmse': 1.1749918955167673,
'k_neighbors_qty': 15,
'k_folds_qty': 10,
'k_fold_cross_validation_toggle': True
}
}
res = self.process_cli.main(self.event_mod, None)
self.assertGreater(res['linear']['pre-hyperparameter_optimisation']['rmse'], 1.1)
self.assertLess(res['linear']['pre-hyperparameter_optimisation']['rmse'], 1.4)
self.assertGreater(res['linear']['post-hyperparameter_optimisation']['rmse'], 1.1)
self.assertLess(res['linear']['post-hyperparameter_optimisation']['rmse'], 1.4)
self.assertGreater(res['knn']['rmse'], 1.1)
self.assertLess(res['knn']['rmse'], 1.4)
class MainExample4NonLogisticTestCase(unittest.TestCase):
"""Tests for `main.py`."""
def setUp(self):
from main import ProcessCLI
# Obtain from terminal log when run main.py runs `vars(parser.parse_args())`
args = {
'data_set':
5,
'logistic':
False,
'kfold':
True,
'linear':
True,
'knn':
'scikit',
'training_features': [
'vote-bill1', 'vote-bill4', 'vote-bill5', 'vote-bill6',
'vote-bill7', 'vote-bill8'
],
'target_feature':
'extremism',
'multi_class_features':
'',
'exclude_non_numeric':
'',
'exclude_non_ordinal':
'',
'exclude_out_of_scope':
'',
'cleanse_price_format_features':
'',
'convert_feature_words_to_digits':
'',
'kmeans':
True,
'kmeans_qty':
2,
'affiliation_feature':
'party',
'kfold_qty':
10,
'hyper_optim':
True,
'hyper_optim_range':
20,
'suppress_all_plots':
False
}
# Suppress all plots
args["suppress_all_plots"] = True
self.process_cli = ProcessCLI(**args)
self.event_mod = self.process_cli.map_cli_args_to_event_config(EVENT)
def tearDown(self):
del self.process_cli
del self.event_mod
def test_valid_config_responds_with_expected_results(self):
expected_result = {
'logistic': None,
'linear': {
'pre-hyperparameter_optimisation': {
'model_type': 'linear',
'rmse': 4.409405296207438
},
'post-hyperparameter_optimisation': {
'model_type': 'linear',
'feature_names': 'vote-bill1__vote-bill4__vote-bill7',
'rmse': 3.8974455631403573,
'k_neighbors_qty': 1,
'k_folds_qty': 10,
'k_fold_cross_validation_toggle': True
}
},
'knn': {
'feature_names': 'vote-bill1__vote-bill4__vote-bill7',
'rmse': 3.8689053880562674,
'k_neighbors_qty': 5,
'k_folds_qty': 10,
'k_fold_cross_validation_toggle': True
}
}
self.assertEqual(self.process_cli.main(self.event_mod, None),
expected_result)
class MainExample2TestCase(unittest.TestCase):
"""Tests for `main.py`."""
def setUp(self):
from main import ProcessCLI
# Obtain from terminal log when run main.py runs `vars(parser.parse_args())`
args = {
'data_set':
2,
'logistic':
False,
'kfold':
True,
'linear':
True,
'knn':
'scikit',
'training_features': [
'num-of-doors', 'curb-weight', 'horsepower', 'city-mpg',
'highway-mpg'
],
'target_feature':
'price',
'multi_class_features':
'',
'exclude_non_numeric': [
'make', 'fuel-type', 'aspiration', 'body-style',
'drive-wheels', 'engine-location', 'engine-type', 'fuel-system'
],
'exclude_non_ordinal':
'',
'exclude_out_of_scope': ['symboling', 'normalized-losses'],
'cleanse_price_format_features': ['price'],
'convert_feature_words_to_digits':
['num-of-doors', 'num-of-cylinders'],
'kmeans':
True,
'kmeans_qty':
5,
'affiliation_feature':
'',
'kfold_qty':
10,
'hyper_optim':
True,
'hyper_optim_range':
20,
'suppress_all_plots':
False
}
# Suppress all plots
args["suppress_all_plots"] = True
self.process_cli = ProcessCLI(**args)
self.event_mod = self.process_cli.map_cli_args_to_event_config(EVENT)
def tearDown(self):
del self.process_cli
del self.event_mod
def test_valid_config_responds_with_expected_results(self):
expected_result = {
'logistic': None,
'linear': {
'pre-hyperparameter_optimisation': {
'model_type': 'linear',
'rmse': 4276.331113497754
},
'post-hyperparameter_optimisation': {
'model_type': 'linear',
'feature_names': 'num-of-doors__curb-weight',
'rmse': 4235.3551812093447,
'k_neighbors_qty': 1,
'k_folds_qty': 10,
'k_fold_cross_validation_toggle': True
}
},
'knn': {
'feature_names': 'num-of-doors__curb-weight',
'rmse': 3945.9183360130905,
'k_neighbors_qty': 6,
'k_folds_qty': 10,
'k_fold_cross_validation_toggle': True
}
}
res = self.process_cli.main(self.event_mod, None)
self.assertEqual(
res['linear']['pre-hyperparameter_optimisation'],
expected_result['linear']['pre-hyperparameter_optimisation'])
self.assertAlmostEqual(
res['linear']['post-hyperparameter_optimisation']['rmse'],
expected_result['linear']['post-hyperparameter_optimisation']
['rmse'])
self.assertEqual(res['knn'], expected_result['knn'])
class MainExample3LogisticTestCase(unittest.TestCase):
"""Tests for `main.py`."""
def setUp(self):
from main import ProcessCLI
# Obtain from terminal log when run main.py runs `vars(parser.parse_args())`
args = {
'data_set':
3,
'logistic':
True,
'kfold':
False,
'linear':
True,
'knn':
'scikit',
'training_features': [
'cylinders', 'displacement', 'horsepower', 'weight',
'acceleration', 'model-year', 'car-name'
],
'target_feature':
'origin',
'multi_class_features': ['cylinders', 'model-year'],
'exclude_non_numeric': ['car-name'],
'exclude_non_ordinal':
'',
'exclude_out_of_scope':
'',
'cleanse_price_format_features':
'',
'convert_feature_words_to_digits':
'',
'kmeans':
True,
'kmeans_qty':
5,
'affiliation_feature':
'',
'kfold_qty':
10,
'hyper_optim':
True,
'hyper_optim_range':
20,
'suppress_all_plots':
False
}
# Suppress all plots
args["suppress_all_plots"] = True
self.process_cli = ProcessCLI(**args)
self.event_mod = self.process_cli.map_cli_args_to_event_config(EVENT)
def tearDown(self):
del self.process_cli
del self.event_mod
def test_valid_config_responds_with_expected_results(self):
expected_result = {
'logistic': {
'pre-hyperparameter_optimisation': {
'model_type': 'logistic',
'rmse': None,
'accuracy': 0.0792838874680307,
'sensitivity': 0.5,
'specificity': 0.5,
'auc_score': None
}
},
'linear': {
'pre-hyperparameter_optimisation': {
'model_type': 'linear',
'rmse': 0.6360193336863021
}
},
'knn': {
'feature_names': 'cyl_3__cyl_4__cyl_5__mod_74__mod_78',
'rmse': 0.8799866005988728,
'k_neighbors_qty': 6,
'k_folds_qty': 10,
'k_fold_cross_validation_toggle': False
}
}
self.assertEqual(self.process_cli.main(self.event_mod, None),
expected_result)
def setUp(self):
from main import ProcessCLI
# Obtain from terminal log when main.py runs `vars(parser.parse_args())`
args = {
'data_set':
1,
'logistic':
False,
'kfold':
True,
'linear':
True,
'knn':
'scikit',
'training_features':
['accommodates', 'bedrooms', 'bathrooms', 'number_of_reviews'],
'target_feature':
'price',
'multi_class_features':
'',
'exclude_non_numeric': [
'host_response_time', 'host_response_rate',
'host_acceptance_rate', 'host_is_superhost',
'host_listings_count', 'host_verifications',
'host_has_profile_pic', 'host_identity_verified',
'property_type', 'room_type', 'bed_type', 'amenities',
'calendar_updated', 'has_availability', 'requires_license',
'license', 'instant_bookable', 'cancellation_policy',
'require_guest_profile_picture',
'require_guest_phone_verification'
],
'exclude_non_ordinal': ['latitude', 'longitude', 'zipcode'],
'exclude_out_of_scope': [
'id', 'listing_url', 'scrape_id', 'last_scraped', 'name',
'summary', 'space', 'description', 'experiences_offered',
'neighborhood_overview', 'notes', 'transit', 'thumbnail_url',
'medium_url', 'picture_url', 'xl_picture_url', 'host_id',
'host_url', 'host_name', 'host_since', 'host_location',
'host_about', 'host_thumbnail_url', 'host_picture_url',
'host_neighbourhood', 'street', 'neighbourhood',
'neighbourhood_cleansed', 'neighbourhood_group_cleansed',
'city', 'state', 'market', 'smart_location', 'country_code',
'country', 'is_location_exact', 'calendar_last_scraped',
'first_review', 'last_review', 'jurisdiction_names'
],
'cleanse_price_format_features': [
'price', 'weekly_price', 'monthly_price', 'security_deposit',
'cleaning_fee', 'extra_people'
],
'convert_feature_words_to_digits':
'',
'kmeans':
True,
'kmeans_qty':
5,
'affiliation_feature':
'',
'kfold_qty':
10,
'hyper_optim':
True,
'hyper_optim_range':
20,
'suppress_all_plots':
False
}
# Suppress all plots
args["suppress_all_plots"] = True
self.process_cli = ProcessCLI(**args)
self.event_mod = self.process_cli.map_cli_args_to_event_config(EVENT)
class MainExample1TestCase(unittest.TestCase):
"""Tests for `main.py`."""
def setUp(self):
from main import ProcessCLI
# Obtain from terminal log when main.py runs `vars(parser.parse_args())`
args = {
'data_set':
1,
'logistic':
False,
'kfold':
True,
'linear':
True,
'knn':
'scikit',
'training_features':
['accommodates', 'bedrooms', 'bathrooms', 'number_of_reviews'],
'target_feature':
'price',
'multi_class_features':
'',
'exclude_non_numeric': [
'host_response_time', 'host_response_rate',
'host_acceptance_rate', 'host_is_superhost',
'host_listings_count', 'host_verifications',
'host_has_profile_pic', 'host_identity_verified',
'property_type', 'room_type', 'bed_type', 'amenities',
'calendar_updated', 'has_availability', 'requires_license',
'license', 'instant_bookable', 'cancellation_policy',
'require_guest_profile_picture',
'require_guest_phone_verification'
],
'exclude_non_ordinal': ['latitude', 'longitude', 'zipcode'],
'exclude_out_of_scope': [
'id', 'listing_url', 'scrape_id', 'last_scraped', 'name',
'summary', 'space', 'description', 'experiences_offered',
'neighborhood_overview', 'notes', 'transit', 'thumbnail_url',
'medium_url', 'picture_url', 'xl_picture_url', 'host_id',
'host_url', 'host_name', 'host_since', 'host_location',
'host_about', 'host_thumbnail_url', 'host_picture_url',
'host_neighbourhood', 'street', 'neighbourhood',
'neighbourhood_cleansed', 'neighbourhood_group_cleansed',
'city', 'state', 'market', 'smart_location', 'country_code',
'country', 'is_location_exact', 'calendar_last_scraped',
'first_review', 'last_review', 'jurisdiction_names'
],
'cleanse_price_format_features': [
'price', 'weekly_price', 'monthly_price', 'security_deposit',
'cleaning_fee', 'extra_people'
],
'convert_feature_words_to_digits':
'',
'kmeans':
True,
'kmeans_qty':
5,
'affiliation_feature':
'',
'kfold_qty':
10,
'hyper_optim':
True,
'hyper_optim_range':
20,
'suppress_all_plots':
False
}
# Suppress all plots
args["suppress_all_plots"] = True
self.process_cli = ProcessCLI(**args)
self.event_mod = self.process_cli.map_cli_args_to_event_config(EVENT)
def tearDown(self):
del self.process_cli
del self.event_mod
def test_valid_config_responds_with_expected_results(self):
expected_result = {
'logistic': None,
'linear': {
'pre-hyperparameter_optimisation': {
'model_type': 'linear',
'rmse': 111.78870076908206
},
'post-hyperparameter_optimisation': {
'model_type': 'linear',
'feature_names': 'accommodates__bedrooms__bathrooms',
'rmse': 109.86142964502258,
'k_neighbors_qty': 1,
'k_folds_qty': 10,
'k_fold_cross_validation_toggle': True
}
},
'knn': {
'feature_names': 'accommodates__bedrooms__bathrooms',
'rmse': 109.65178843975404,
'k_neighbors_qty': 20,
'k_folds_qty': 10,
'k_fold_cross_validation_toggle': True
}
}
self.assertEqual(self.process_cli.main(self.event_mod, None),
expected_result)
class MainExample4NonLogisticTestCase(unittest.TestCase):
"""Tests for `main.py`."""
def setUp(self):
from main import ProcessCLI
# Obtain from terminal log when run main.py runs `vars(parser.parse_args())`
args = {
'data_set': 4,
'logistic': True,
'kfold': False,
'linear': True,
'knn': 'scikit',
'training_features': [
'gpa',
'gre'
],
'target_feature': 'admit',
'multi_class_features': '',
'exclude_non_numeric': '',
'exclude_non_ordinal': '',
'exclude_out_of_scope': '',
'cleanse_price_format_features': '',
'convert_feature_words_to_digits': '',
'kmeans': True,
'kmeans_qty': 5,
'affiliation_feature': '',
'kfold_qty': 10,
'hyper_optim': True,
'hyper_optim_range': 20,
'suppress_all_plots': False
}
# Suppress all plots
args["suppress_all_plots"] = True
self.process_cli = ProcessCLI(**args)
self.event_mod = self.process_cli.map_cli_args_to_event_config(EVENT)
def tearDown(self):
del self.process_cli
del self.event_mod
def test_valid_config_responds_with_expected_results(self):
expected_result = {
'logistic': {
'pre-hyperparameter_optimisation': {
'model_type': 'logistic',
'rmse': 0.393589629343064,
'accuracy': 0.782608695652174,
'sensitivity': 0.5,
'specificity': 0.5,
'auc_score': 0.85679303278688534
}
},
'linear': {
'pre-hyperparameter_optimisation': {
'model_type': 'linear',
'rmse': 0.39200813287845615
}
},
'knn': {
'feature_names': 'gpa__gre',
'rmse': 0.6196196990320526,
'k_neighbors_qty': 8,
'k_folds_qty': 10,
'k_fold_cross_validation_toggle': False
}
}
self.assertEqual(self.process_cli.main(self.event_mod, None), expected_result)
class MainExample3NonLogisticTestCase(unittest.TestCase):
"""Tests for `main.py`."""
def setUp(self):
from main import ProcessCLI
# Obtain from terminal log when run main.py runs `vars(parser.parse_args())`
args = {
'data_set':
3,
'logistic':
False,
'kfold':
True,
'linear':
True,
'knn':
'scikit',
'training_features': [
'cylinders', 'displacement', 'horsepower', 'weight',
'acceleration', 'model-year', 'car-name'
],
'target_feature':
'mpg',
'multi_class_features':
'',
'exclude_non_numeric': ['car-name'],
'exclude_non_ordinal':
'',
'exclude_out_of_scope':
'',
'cleanse_price_format_features':
'',
'convert_feature_words_to_digits':
'',
'kmeans':
True,
'kmeans_qty':
5,
'affiliation_feature':
'',
'kfold_qty':
10,
'hyper_optim':
True,
'hyper_optim_range':
20,
'suppress_all_plots':
False
}
# Suppress all plots
args["suppress_all_plots"] = True
self.process_cli = ProcessCLI(**args)
self.event_mod = self.process_cli.map_cli_args_to_event_config(EVENT)
def tearDown(self):
del self.process_cli
del self.event_mod
def test_valid_config_responds_with_expected_results(self):
expected_result = {
'logistic': None,
'linear': {
'pre-hyperparameter_optimisation': {
'model_type': 'linear',
'rmse': 6.002793054130211
},
'post-hyperparameter_optimisation': {
'model_type': 'linear',
'feature_names': 'acceleration__model-year',
'rmse': 6.0241861168232917,
'k_neighbors_qty': 1,
'k_folds_qty': 10,
'k_fold_cross_validation_toggle': True
}
},
'knn': {
'feature_names': 'acceleration__model-year',
'rmse': 5.75101716901536,
'k_neighbors_qty': 20,
'k_folds_qty': 10,
'k_fold_cross_validation_toggle': True
}
}
self.assertEqual(self.process_cli.main(self.event_mod, None),
expected_result)