def test_wine():
X, y = demos.load_wine()
assert X.shape == (178, 13)
assert y.shape == (178, )
assert isinstance(X, ww.DataTable)
assert isinstance(y, ww.DataColumn)
X, y = demos.load_wine(return_pandas=True)
assert X.shape == (178, 13)
assert y.shape == (178, )
assert isinstance(X, pd.DataFrame)
assert isinstance(y, pd.Series)
def test_partial_dependence_multiclass(
logistic_regression_multiclass_pipeline_class):
X, y = load_wine()
pipeline = logistic_regression_multiclass_pipeline_class(
parameters={"Logistic Regression Classifier": {
"n_jobs": 1
}})
pipeline.fit(X, y)
num_classes = y.to_series().nunique()
grid_resolution = 20
one_way_part_dep = partial_dependence(pipeline=pipeline,
X=X,
features="magnesium",
grid_resolution=grid_resolution)
assert "class_label" in one_way_part_dep.columns
assert one_way_part_dep["class_label"].nunique() == num_classes
assert len(one_way_part_dep.index) == num_classes * grid_resolution
assert list(one_way_part_dep.columns) == [
"feature_values", "partial_dependence", "class_label"
]
two_way_part_dep = partial_dependence(pipeline=pipeline,
X=X,
features=("magnesium", "alcohol"),
grid_resolution=grid_resolution)
assert "class_label" in two_way_part_dep.columns
assert two_way_part_dep["class_label"].nunique() == num_classes
assert len(two_way_part_dep.index) == num_classes * grid_resolution
assert len(two_way_part_dep.columns) == grid_resolution + 1
def test_pipeline_has_classes_property(
logistic_regression_binary_pipeline_class,
logistic_regression_multiclass_pipeline_class, problem_type, use_ints):
if problem_type == "binary":
X, y = load_breast_cancer(return_pandas=True)
pipeline = logistic_regression_binary_pipeline_class(
parameters={"Logistic Regression Classifier": {
"n_jobs": 1
}})
if use_ints:
y = y.map({'malignant': 0, 'benign': 1})
answer = [0, 1]
else:
answer = ["benign", "malignant"]
elif problem_type == "multi":
X, y = load_wine(return_pandas=True)
pipeline = logistic_regression_multiclass_pipeline_class(
parameters={"Logistic Regression Classifier": {
"n_jobs": 1
}})
if use_ints:
y = y.map({"class_0": 0, "class_1": 1, "class_2": 2})
answer = [0, 1, 2]
else:
answer = ["class_0", "class_1", "class_2"]
with pytest.raises(
AttributeError,
match="Cannot access class names before fitting the pipeline."):
pipeline.classes_
pipeline.fit(X, y)
pd.testing.assert_series_equal(pd.Series(pipeline.classes_),
pd.Series(answer))
def test_graph_partial_dependence_multiclass(logistic_regression_multiclass_pipeline_class):
go = pytest.importorskip('plotly.graph_objects', reason='Skipping plotting test because plotly not installed')
X, y = load_wine()
pipeline = logistic_regression_multiclass_pipeline_class(parameters={"Logistic Regression Classifier": {"n_jobs": 1}})
pipeline.fit(X, y)
# Test one-way without class labels
fig_one_way_no_class_labels = graph_partial_dependence(pipeline, X, features='magnesium', grid_resolution=20)
assert isinstance(fig_one_way_no_class_labels, go.Figure)
fig_dict = fig_one_way_no_class_labels.to_dict()
assert len(fig_dict['data']) == len(pipeline.classes_)
for data, label in zip(fig_dict['data'], pipeline.classes_):
assert len(data['x']) == 20
assert len(data['y']) == 20
assert data['name'] == label
# Check that all the subplots axes have the same range
for suplot_1_axis, suplot_2_axis in [('axis2', 'axis3'), ('axis2', 'axis4'), ('axis3', 'axis4')]:
for axis_type in ['x', 'y']:
assert fig_dict['layout'][axis_type + suplot_1_axis]['range'] == fig_dict['layout'][axis_type + suplot_2_axis]['range']
# Test one-way with class labels
fig_one_way_class_labels = graph_partial_dependence(pipeline, X, features='magnesium', class_label='class_1', grid_resolution=20)
assert isinstance(fig_one_way_class_labels, go.Figure)
fig_dict = fig_one_way_class_labels.to_dict()
assert len(fig_dict['data']) == 1
assert len(fig_dict['data'][0]['x']) == 20
assert len(fig_dict['data'][0]['y']) == 20
assert fig_dict['data'][0]['name'] == 'class_1'
msg = "Class wine is not one of the classes the pipeline was fit on: class_0, class_1, class_2"
with pytest.raises(ValueError, match=msg):
graph_partial_dependence(pipeline, X, features='alcohol', class_label='wine')
# Test two-way without class labels
fig_two_way_no_class_labels = graph_partial_dependence(pipeline, X, features=('magnesium', 'alcohol'), grid_resolution=20)
assert isinstance(fig_two_way_no_class_labels, go.Figure)
fig_dict = fig_two_way_no_class_labels.to_dict()
assert len(fig_dict['data']) == 3, "Figure does not have partial dependence data for each class."
assert all([len(fig_dict["data"][i]['x']) == 20 for i in range(3)])
assert all([len(fig_dict["data"][i]['y']) == 20 for i in range(3)])
assert [fig_dict["data"][i]['name'] for i in range(3)] == ["class_0", "class_1", "class_2"]
# Check that all the subplots axes have the same range
for suplot_1_axis, suplot_2_axis in [('axis', 'axis2'), ('axis', 'axis3'), ('axis2', 'axis3')]:
for axis_type in ['x', 'y']:
assert fig_dict['layout'][axis_type + suplot_1_axis]['range'] == fig_dict['layout'][axis_type + suplot_2_axis]['range']
# Test two-way with class labels
fig_two_way_class_labels = graph_partial_dependence(pipeline, X, features=('magnesium', 'alcohol'), class_label='class_1', grid_resolution=20)
assert isinstance(fig_two_way_class_labels, go.Figure)
fig_dict = fig_two_way_class_labels.to_dict()
assert len(fig_dict['data']) == 1
assert len(fig_dict['data'][0]['x']) == 20
assert len(fig_dict['data'][0]['y']) == 20
assert fig_dict['data'][0]['name'] == 'class_1'
msg = "Class wine is not one of the classes the pipeline was fit on: class_0, class_1, class_2"
with pytest.raises(ValueError, match=msg):
graph_partial_dependence(pipeline, X, features='alcohol', class_label='wine')
def test_invalid_targets_regression_pipeline(target_type, dummy_regression_pipeline_class):
X, y = load_wine(return_pandas=True)
if target_type == "category":
y = pd.Series(y).astype("category")
if target_type == "bool":
X, y = load_breast_cancer(return_pandas=True)
y = y.map({"malignant": False, "benign": True})
mock_regression_pipeline = dummy_regression_pipeline_class(parameters={})
with pytest.raises(ValueError, match="Regression pipeline can only handle numeric target data"):
mock_regression_pipeline.fit(X, y)
def test_partial_dependence_multiclass_categorical(class_label,
logistic_regression_multiclass_pipeline_class):
pytest.importorskip('plotly.graph_objects', reason='Skipping plotting test because plotly not installed')
X, y = load_wine()
X['categorical_column'] = ww.DataColumn(pd.Series([i % 3 for i in range(X.shape[0])]).astype(str),
logical_type="Categorical")
X['categorical_column_2'] = ww.DataColumn(pd.Series([i % 6 for i in range(X.shape[0])]).astype(str),
logical_type="Categorical")
pipeline = logistic_regression_multiclass_pipeline_class({"Logistic Regression Classifier": {"n_jobs": 1}})
pipeline.fit(X, y)
fig = graph_partial_dependence(pipeline, X, features='categorical_column', class_label=class_label,
grid_resolution=5)
for i, plot_data in enumerate(fig.to_dict()['data']):
assert plot_data['type'] == 'bar'
assert plot_data['x'].tolist() == ['0', '1', '2']
if class_label is None:
assert plot_data['name'] == f'class_{i}'
else:
assert plot_data['name'] == class_label
fig = graph_partial_dependence(pipeline, X, features=('alcohol', 'categorical_column'), class_label=class_label,
grid_resolution=5)
for i, plot_data in enumerate(fig.to_dict()['data']):
assert plot_data['type'] == 'contour'
assert fig.to_dict()['layout']['yaxis']['ticktext'] == ['0', '1', '2']
if class_label is None:
assert plot_data['name'] == f'class_{i}'
else:
assert plot_data['name'] == class_label
fig = graph_partial_dependence(pipeline, X, features=('categorical_column_2', 'categorical_column'),
class_label=class_label, grid_resolution=5)
for i, plot_data in enumerate(fig.to_dict()['data']):
assert plot_data['type'] == 'contour'
assert fig.to_dict()['layout']['xaxis']['ticktext'] == ['0', '1', '2']
assert fig.to_dict()['layout']['yaxis']['ticktext'] == ['0', '1', '2', '3', '4', '5']
if class_label is None:
assert plot_data['name'] == f'class_{i}'
else:
assert plot_data['name'] == class_label
