def test_roc_curve_fails_correctly_without_predict_proba(self):
dataset = load_demo_dataset("iris")
svc = Model(SVC(gamma="scale"))
result = svc.score_estimator(dataset)
with pytest.raises(VizError):
result.plot.roc_curve()
plt.close()
def test_dataset_y_from_fetchopenml_with_two_target_columns_works(self):
dataset = load_demo_dataset(
"openml",
name="blood-transfusion-service-center",
target_column=["V1", "V2"],
)
features_y = dataset.y
assert features_y.shape == (748, 2)
def test_load_prediction_data_works_as_expected(self):
dataset = load_demo_dataset("iris")
dataset.create_train_test(stratify=True)
feature_pipeline = Pipeline([("scale", DFStandardScaler())])
model = Model(LogisticRegression(), feature_pipeline=feature_pipeline)
model.train_estimator(dataset)
result = model.make_prediction(dataset, 5)
expected = pd.DataFrame({"Prediction": [0]})
pd.testing.assert_frame_equal(result, expected, check_dtype=False)
def test_pr_curve_fails_correctly_without_predict_proba(self):
"""
Expect that the plot will raise an exception if the estimator
does not have a predict_proba method
"""
dataset = load_demo_dataset("iris")
svc = Model(SVC(gamma="scale"))
result = svc.score_estimator(dataset)
with pytest.raises(VizError):
result.plot.precision_recall_curve()
plt.close()
def classifier_result(self) -> Result:
"""Setup a classiifer Result"""
dataset = load_demo_dataset("iris")
model = Model(LogisticRegression())
return model.score_estimator(dataset)
def result_cv(self, model: Model) -> Result:
"""Setup a Result from a cross-validated scoring"""
dataset = load_demo_dataset("boston")
return model.score_estimator(dataset, cv=2)
def result(self, model: Model) -> Result:
"""Setup a Result from a score_estimator without cv"""
dataset = load_demo_dataset("boston")
return model.score_estimator(dataset)
def regression_result(self) -> Result:
"""Setup a regression Result"""
dataset = load_demo_dataset("boston")
model = Model(LinearRegression())
return model.score_estimator(dataset)
def dataset(self):
return load_demo_dataset("iris")
def classification_result(self) -> Result:
"""Setup a classification Result"""
dataset = load_demo_dataset("breast_cancer")
model = Model(LogisticRegression())
return model.score_estimator(dataset)
def iris_result(self) -> Result:
dataset = load_demo_dataset("iris")
model = Model(LogisticRegression())
return model.score_estimator(dataset)
def test_dataset_x_from_fetchopenml_with_parameters_works(self):
dataset = load_demo_dataset("openml",
name="blood-transfusion-service-center",
target_column="V1")
features_x = dataset.x
assert features_x.shape == (748, 4)
def test_dataset_from_fetchopenml_works(self):
dataset = load_demo_dataset("openml", name="miceprotein")
assert len(dataset.x) == 1080
def load_dataset_iris(self) -> Dataset:
return load_demo_dataset("iris")
def dataset(self):
"""Setup a Dataset"""
return load_demo_dataset("iris")
