def test_automl_time_series_classification_pickle_generated_pipeline(
mock_binary_fit, mock_multi_fit, mock_binary_score,
mock_multiclass_score, problem_type, X_y_binary, X_y_multi):
if problem_type == ProblemTypes.TIME_SERIES_BINARY:
X, y = X_y_binary
pipeline = GeneratedPipelineTimeSeriesBinary
else:
X, y = X_y_multi
pipeline = GeneratedPipelineTimeSeriesMulticlass
configuration = {
"gap": 0,
"max_delay": 0,
'delay_target': False,
'delay_features': True
}
a = AutoMLSearch(X_train=X,
y_train=y,
problem_type=problem_type,
problem_configuration=configuration)
a.search()
for i, row in a.rankings.iterrows():
assert a.get_pipeline(row['id']).__class__ == pipeline
assert pickle.loads(pickle.dumps(a.get_pipeline(row['id'])))
def test_automl_time_series_regression_pickle_generated_pipeline(mock_fit, mock_score, X_y_regression):
X, y = X_y_regression
configuration = {"gap": 0, "max_delay": 0, 'delay_target': False, 'delay_features': True}
a = AutoMLSearch(X_train=X, y_train=y, problem_type="time series regression", problem_configuration=configuration)
a.search()
for i, row in a.rankings.iterrows():
assert a.get_pipeline(row['id']).__class__ == GeneratedPipelineTimeSeriesRegression
assert pickle.loads(pickle.dumps(a.get_pipeline(row['id'])))
def test_init(X_y_regression):
X, y = X_y_regression
automl = AutoMLSearch(X_train=X,
y_train=y,
problem_type='regression',
objective="R2",
max_iterations=3,
n_jobs=1)
automl.search()
assert automl.n_jobs == 1
assert isinstance(automl.rankings, pd.DataFrame)
assert isinstance(automl.best_pipeline, PipelineBase)
automl.best_pipeline.predict(X)
# test with dataframes
automl = AutoMLSearch(pd.DataFrame(X),
pd.Series(y),
problem_type='regression',
objective="R2",
max_iterations=3,
n_jobs=1)
automl.search()
assert isinstance(automl.rankings, pd.DataFrame)
assert isinstance(automl.full_rankings, pd.DataFrame)
assert isinstance(automl.best_pipeline, PipelineBase)
automl.best_pipeline.predict(X)
assert isinstance(automl.get_pipeline(0), PipelineBase)
def test_init(X_y_binary):
X, y = X_y_binary
automl = AutoMLSearch(X_train=X,
y_train=y,
problem_type='binary',
max_iterations=1,
n_jobs=1)
automl.search()
assert automl.n_jobs == 1
assert isinstance(automl.rankings, pd.DataFrame)
assert isinstance(automl.best_pipeline, PipelineBase)
automl.best_pipeline.predict(X)
# test with dataframes
automl = AutoMLSearch(pd.DataFrame(X),
pd.Series(y),
problem_type='binary',
max_iterations=1,
n_jobs=1)
automl.search()
assert isinstance(automl.rankings, pd.DataFrame)
assert isinstance(automl.full_rankings, pd.DataFrame)
assert isinstance(automl.best_pipeline, PipelineBase)
assert isinstance(automl.get_pipeline(0), PipelineBase)
assert automl.objective.name == 'Log Loss Binary'
automl.best_pipeline.predict(X)
def test_automl_pickle_generated_pipeline(mock_binary_score, mock_binary_fit, mock_multi_score, mock_multi_fit,
problem_type, X_y_binary, X_y_multi):
if problem_type == ProblemTypes.BINARY:
X, y = X_y_binary
pipeline = GeneratedPipelineBinary
elif problem_type == ProblemTypes.MULTICLASS:
X, y = X_y_multi
pipeline = GeneratedPipelineMulticlass
a = AutoMLSearch(X_train=X, y_train=y, problem_type=problem_type)
a.search()
for i, row in a.rankings.iterrows():
assert a.get_pipeline(row['id']).__class__ == pipeline
assert pickle.loads(pickle.dumps(a.get_pipeline(row['id'])))
def test_automl_pickle_generated_pipeline(mock_regression_score, mock_regression_fit, X_y_regression):
class RegressionPipelineCustoms(RegressionPipeline):
custom_name = "Custom Regression Name"
component_graph = ["Imputer", "Linear Regressor"]
custom_hyperparameters = {"Imputer": {"numeric_impute_strategy": "most_frequent"}}
X, y = X_y_regression
pipeline = GeneratedPipelineRegression
a = AutoMLSearch(X_train=X, y_train=y, problem_type='regression')
a.search()
a.add_to_rankings(RegressionPipelineCustoms({}))
seen_name = False
for i, row in a.rankings.iterrows():
automl_pipeline = a.get_pipeline(row['id'])
assert automl_pipeline.__class__ == pipeline
assert pickle.loads(pickle.dumps(automl_pipeline))
if automl_pipeline.custom_name == RegressionPipelineCustoms.custom_name:
seen_name = True
assert automl_pipeline.custom_hyperparameters == RegressionPipelineCustoms.custom_hyperparameters
assert automl_pipeline.component_graph == RegressionPipelineCustoms.component_graph
assert seen_name
def test_automl_pickle_generated_pipeline(mock_regression_fit,
mock_regression_score,
X_y_regression):
mock_regression_score.return_value = {"R2": 1.0}
class RegressionPipelineCustom(RegressionPipeline):
custom_name = "Custom Regression Name"
component_graph = ["Imputer", "Linear Regressor"]
custom_hyperparameters = {
"Imputer": {
"numeric_impute_strategy": "most_frequent"
}
}
X, y = X_y_regression
pipeline = GeneratedPipelineRegression
allowed_estimators = get_estimators('regression')
allowed_pipelines = [
make_pipeline(X, y, estimator, problem_type='regression')
for estimator in allowed_estimators
]
allowed_pipelines.append(RegressionPipelineCustom)
a = AutoMLSearch(X_train=X,
y_train=y,
problem_type='regression',
allowed_pipelines=allowed_pipelines)
a.search()
a.add_to_rankings(RegressionPipelineCustom({}))
seen_name = False
for i, row in a.rankings.iterrows():
automl_pipeline = a.get_pipeline(row['id'])
assert automl_pipeline.__class__ == pipeline
assert pickle.loads(pickle.dumps(automl_pipeline))
if automl_pipeline.custom_name == RegressionPipelineCustom.custom_name:
seen_name = True
assert automl_pipeline.custom_hyperparameters == RegressionPipelineCustom.custom_hyperparameters
assert automl_pipeline.component_graph == RegressionPipelineCustom.component_graph
assert seen_name
