def test_generic_estimator_for_classification():
pipeline = make_pipeline(
H2OAutoMLEstimator(estimator_type='classifier',
seed=seed,
init_connection_args=init_connection_args))
pipeline.set_params(h2oautomlestimator__max_models=max_models,
h2oautomlestimator__nfolds=3)
pipeline.named_steps.h2oautomlestimator.exclude_algos = ['XGBoost']
data = _get_data(format='numpy', n_classes=3)
assert isinstance(data.X_train, np.ndarray)
pipeline.fit(data.X_train, data.y_train)
assert len(pipeline.named_steps.h2oautomlestimator._estimator.leaderboard
) == max_models + 2
preds = pipeline.predict(data.X_test)
assert isinstance(preds, np.ndarray)
assert preds.shape == (len(data.X_test), )
probs = pipeline.predict_proba(data.X_test)
assert probs.shape == (len(data.X_test), 3)
assert np.allclose(np.sum(probs, axis=1),
1.), "`predict_proba` didn't return probabilities"
score = pipeline.score(data.X_test, data.y_test)
assert isinstance(score, float)
skl_score = accuracy_score(data.y_test, preds)
assert abs(score - skl_score) < 1e-6, "score={}, skl_score={}".format(
score, skl_score)
def test_params_are_correctly_passed_to_underlying_automl():
estimator = H2OAutoMLEstimator(seed=seed)
estimator.set_params(max_models=5, nfolds=0)
estimator.project_name = "dummy"
assert estimator.estimator is None
estimator._make_estimator() # normally done when calling `fit`
aml = estimator.estimator
assert aml
assert aml.build_control["stopping_criteria"]["seed"] == seed
assert aml.build_control["stopping_criteria"]["max_models"] == 5
assert aml.build_control["nfolds"] == 0
assert aml.build_control["project_name"] == "dummy"
def test_generic_estimator_for_regression():
pipeline = make_pipeline(
H2OAutoMLEstimator(estimator_type='regressor',
seed=seed,
init_connection_args=init_connection_args))
pipeline.set_params(h2oautomlestimator__max_models=max_models,
h2oautomlestimator__nfolds=3)
pipeline.named_steps.h2oautomlestimator.exclude_algos = ['XGBoost']
data = _get_data(format='numpy', n_classes=0)
assert isinstance(data.X_train, np.ndarray)
pipeline.fit(data.X_train, data.y_train)
assert len(pipeline.named_steps.h2oautomlestimator._estimator.leaderboard
) == max_models + 2
preds = pipeline.predict(data.X_test)
assert isinstance(preds, np.ndarray)
assert preds.shape == (len(data.X_test), )
score = pipeline.score(data.X_test, data.y_test)
assert isinstance(score, float)
skl_score = r2_score(data.y_test, preds)
assert abs(score - skl_score) < 1e-6, "score={}, skl_score={}".format(
score, skl_score)