def test_grid_predict_usecase():
clf = InteractiveClassifier.from_json(
"tests/test_classification/demo-data.json")
pipe = Pipeline([
("id", PipeTransformer(identity)),
("mod", clf),
])
grid = GridSearchCV(pipe, cv=5, param_grid={})
df = load_penguins(as_frame=True).dropna()
X, y = df.drop(columns=["species", "island", "sex"]), df["species"]
preds = grid.fit(X, y).predict_proba(X)
assert preds.shape[0] == df.shape[0]
assert preds.shape[1] == 3
def test_grid_predict_usecase():
tfm = InteractivePreprocessor.from_json(
"tests/test_classification/demo-data.json")
pipe = Pipeline([
(
"features",
FeatureUnion([("original", PipeTransformer(identity)),
("new_feats", tfm)]),
),
])
df = load_penguins(as_frame=True).dropna()
X, y = df.drop(columns=["species", "island", "sex"]), df["species"]
preds = pipe.fit(X, y).transform(X)
assert preds.shape[0] == df.shape[0]
assert preds.shape[1] == X.shape[1] + 3
def test_grid_predict():
clf = InteractiveOutlierDetector.from_json(
"tests/test_classification/demo-data.json")
pipe = Pipeline([
("id", PipeTransformer(identity)),
("mod", clf),
])
grid = GridSearchCV(
pipe,
cv=5,
param_grid={},
scoring={"acc": make_scorer(accuracy_score)},
refit="acc",
)
df = load_penguins(as_frame=True).dropna()
X = df.drop(columns=["species", "island", "sex"])
y = (np.random.random(df.shape[0]) < 0.1).astype(int)
preds = grid.fit(X, y).predict(X)
assert preds.shape[0] == df.shape[0]
def test_works_with_pipeline_gridsearch(random_xy_dataset_clf):
X, y = random_xy_dataset_clf
pipe = Pipeline([("pipe", PipeTransformer(func=double, factor=1)),
("mod", GaussianNB())])
grid = GridSearchCV(pipe, cv=2, param_grid={"pipe__factor": [1, 2, 3]})
grid.fit(X, y).predict(X)
def test_basic_example(factor):
np.random.seed(42)
X = np.random.normal(0, 1, (1000, 4))
tfm = PipeTransformer(func=double, factor=factor)
X_tfm = tfm.fit_transform(X)
assert np.all(np.isclose(X * factor, X_tfm))
def test_estimator_checks(test_fn):
clf = PipeTransformer(func=double)
test_fn(PipeTransformer.__name__, clf)
def test_works_with_partial_fit(random_xy_dataset_clf):
X, y = random_xy_dataset_clf
pipe = PipeTransformer(func=double, factor=2)
assert np.all(np.isclose(pipe.partial_fit(X, y).transform(X), X * 2))
