def test_import_from_sklearn_pipeline2(self):
from sklearn.feature_selection import SelectKBest, f_regression
from sklearn.pipeline import Pipeline
from sklearn.svm import SVC as SklearnSVC
anova_filter = SelectKBest(f_regression, k=3)
clf = SklearnSVC(kernel="linear")
sklearn_pipeline = Pipeline([("anova", anova_filter), ("svc", clf)])
sklearn_pipeline.fit(self.X_train, self.y_train)
lale_pipeline = import_from_sklearn_pipeline(sklearn_pipeline)
lale_pipeline.predict(self.X_test)
def test_import_from_sklearn_pipeline3(self):
from sklearn.feature_selection import SelectKBest, f_regression
from sklearn.pipeline import Pipeline
from sklearn.svm import SVC as SklearnSVC
anova_filter = SelectKBest(f_regression, k=3)
clf = SklearnSVC(kernel="linear")
sklearn_pipeline = Pipeline([("anova", anova_filter), ("svc", clf)])
lale_pipeline = import_from_sklearn_pipeline(sklearn_pipeline, fitted=False)
with self.assertRaises(
ValueError
): # fitted=False returns a Trainable, so calling predict is invalid.
lale_pipeline.predict(self.X_test)
def test_import_from_sklearn_pipeline(self):
from sklearn.feature_selection import SelectKBest, f_regression
from sklearn.pipeline import Pipeline
from sklearn.svm import SVC as SklearnSVC
anova_filter = SelectKBest(f_regression, k=3)
clf = SklearnSVC(kernel="linear")
sklearn_pipeline = Pipeline([("anova", anova_filter), ("svc", clf)])
lale_pipeline = import_from_sklearn_pipeline(sklearn_pipeline)
for i, pipeline_step in enumerate(sklearn_pipeline.named_steps):
sklearn_step_params = sklearn_pipeline.named_steps[
pipeline_step].get_params()
lale_sklearn_params = lale_pipeline.steps(
)[i]._impl._wrapped_model.get_params()
self.assertEqual(sklearn_step_params, lale_sklearn_params)
self.assert_equal_predictions(sklearn_pipeline, lale_pipeline)
def test_import_from_sklearn_pipeline(self):
from sklearn.feature_selection import SelectKBest, f_regression
from sklearn.pipeline import Pipeline
from sklearn.svm import SVC as SklearnSVC
anova_filter = SelectKBest(f_regression, k=3)
clf = SklearnSVC(kernel="linear")
sklearn_pipeline = Pipeline([("anova", anova_filter), ("svc", clf)])
lale_pipeline = typing.cast(
lale.operators.TrainablePipeline,
import_from_sklearn_pipeline(sklearn_pipeline),
)
for i, pipeline_step in enumerate(sklearn_pipeline.named_steps):
sklearn_step_params = sklearn_pipeline.named_steps[
pipeline_step].get_params()
lale_sklearn_params = self.get_sklearn_params(
lale_pipeline.steps_list()[i])
self.assertEqual(sklearn_step_params, lale_sklearn_params)
self.assert_equal_predictions(sklearn_pipeline, lale_pipeline)
def _get_ml_model(self, cores_for_training: int = 2, X=None):
return SklearnSVC(**self._parameters)