actableai.utils.tests package

Submodules

actableai.utils.tests.test__init__ module

actableai.utils.tests.test__init__.test_get_type_special_no_ag()

actableai.utils.tests.test__init__.test_is_fitted()

actableai.utils.tests.test__init__.test_is_fitted_2()

actableai.utils.tests.test_categorical_numerical_convert module

actableai.utils.tests.test_categorical_numerical_convert.test_categorical_to_num()

Check if conversion of categorical features to enumerated, and conversion back to numerical yields the same dataframe as the original.

actableai.utils.tests.test_dataset_generator module

class actableai.utils.tests.test_dataset_generator.TestGenerateDataset

Bases: object

test_generate_and_save_simple_dataset(tmp_path)

test_generate_simple_dataset()

class actableai.utils.tests.test_dataset_generator.TestGenerateDatasetParameters

Bases: object

pytestmark = [Mark(name='parametrize', args=('save_parameters', [True, False]), kwargs={}), Mark(name='parametrize', args=('save_output', [True, False]), kwargs={})]

test_columns(tmp_path, save_output, save_parameters, columns, column_type)

test_columns_mixed_type(tmp_path, save_output, save_parameters, columns, random_state)

test_date_column_default(tmp_path, save_output, save_parameters)

test_date_column_end(tmp_path, save_output, save_parameters)

test_date_column_end_freq(tmp_path, save_output, save_parameters)

test_date_column_freq(tmp_path, save_output, save_parameters, freq)

test_date_column_start(tmp_path, save_output, save_parameters)

test_date_column_start_end(tmp_path, save_output, save_parameters)

test_date_column_start_end_freq(tmp_path, save_output, save_parameters)

test_date_column_start_freq(tmp_path, save_output, save_parameters)

test_name_one_column(tmp_path, save_output, save_parameters, column_type)

test_name_three_columns(tmp_path, save_output, save_parameters)

test_no_name_one_column(tmp_path, save_output, save_parameters, column_type)

test_no_name_three_columns(tmp_path, save_output, save_parameters)

test_no_type(tmp_path, save_output, save_parameters)

test_number_column_default(tmp_path, save_output, save_parameters)

test_number_column_default_float(tmp_path, save_output, save_parameters)

test_number_column_default_range(tmp_path, save_output, save_parameters)

test_number_column_float(tmp_path, save_output, save_parameters)

test_number_column_float_range(tmp_path, save_output, save_parameters, range_min, range_max)

test_number_column_int(tmp_path, save_output, save_parameters)

test_number_column_int_range(tmp_path, save_output, save_parameters, range_min, range_max)

test_one_value(tmp_path, save_output, save_parameters)

test_random_state_one_column(tmp_path, save_output, save_parameters, random_state, column_type)

test_random_state_three_columns(tmp_path, save_output, save_parameters, random_state)

test_rows(tmp_path, save_output, save_parameters, rows)

test_text_column_categories(tmp_path, save_output, save_parameters, n_categories)

test_text_column_default(tmp_path, save_output, save_parameters)

test_text_column_default_categories(tmp_path, save_output, save_parameters)

test_text_column_default_range(tmp_path, save_output, save_parameters)

test_text_column_default_word_range(tmp_path, save_output, save_parameters)

test_text_column_range(tmp_path, save_output, save_parameters, range_min, range_max)

test_text_column_word_range(tmp_path, save_output, save_parameters, range_min, range_max)

test_three_values(tmp_path, save_output, save_parameters)

actableai.utils.tests.test_dataset_generator.call_dataset_generator(tmp_path: pathlib.Path, columns_parameters: List[dict], rows: int, save_output: bool, save_parameters: bool, random_state: Optional[int] = None) → pandas.core.frame.DataFrame

Call dataset generator

Parameters

• tmp_path (The temporary path from pytest, this is where the output will be saved if needed) –
• columns_parameters (The dataset columns parameters) –
• rows (The number of rows to generate) –
• save_output (If true will save the output as a csv to then read it and return the content) –
• save_parameters (If true will save the parameters as a json to then parse them and return the generated dataset) –
• random_state (The random state) –

Return type

Either the generated dataframe directly, or the read dataframe from the generated csv file

actableai.utils.tests.test_parse_datetime module

class actableai.utils.tests.test_parse_datetime.TestParseDatetime

Bases: object

test_datetime_contain_milliseconds()

test_mixed_datetime()

test_multi_format_datetime()

test_non_datetime()

test_simple_datetime()

test_year_day_month_format()

actableai.utils.tests.test_pdp_ice module

actableai.utils.tests.test_pdp_ice.classification_task()

actableai.utils.tests.test_pdp_ice.regression_task()

actableai.utils.tests.test_pdp_ice.test_pdp_ice_classification(classification_task, tmp_path)

Check if PDP and ICE for classification tasks runs without errors, and that the outputs are of the expected dimensions

actableai.utils.tests.test_pdp_ice.test_pdp_ice_classification_null(classification_task, tmp_path)

Check if PDP and ICE for classification tasks runs without errors, and that the outputs are of the expected dimensions

In this case, each row contains at least one column with a null value; ensure that PDP/ICE can still be computed

This also tests for handling of null values in numerical and categorical columns

actableai.utils.tests.test_pdp_ice.test_pdp_ice_regression(regression_task, tmp_path)

Check if PDP and ICE for regression tasks runs without errors, and that the outputs are of the expected dimensions

actableai.utils.tests.test_pdp_ice.test_pdp_ice_regression_null(regression_task, tmp_path)

Check if PDP and ICE for regression tasks runs without errors, and that the outputs are of the expected dimensions

In this case, each row contains at least one column with a null value; ensure that PDP/ICE can still be computed

This also tests for handling of null values in numerical and categorical columns

actableai.utils.tests.test_sanitize module

actableai.utils.tests.test_sanitize.test_sanitize()

Module contents