def test_autosklearn2_classification_methods_returns_self_sparse(dask_client):
X_train, y_train, X_test, y_test = putil.get_dataset('breast_cancer',
make_sparse=True)
automl = AutoSklearn2Classifier(time_left_for_this_task=60,
ensemble_size=0,
delete_tmp_folder_after_terminate=False,
dask_client=dask_client)
automl_fitted = automl.fit(X_train, y_train)
assert automl is automl_fitted
automl_ensemble_fitted = automl.fit_ensemble(y_train, ensemble_size=5)
assert automl is automl_ensemble_fitted
automl_refitted = automl.refit(X_train.copy(), y_train.copy())
assert automl is automl_refitted
predictions = automl_fitted.predict(X_test)
assert sklearn.metrics.accuracy_score(
y_test, predictions) >= 2 / 3, print_debug_information(automl)
assert "boosting" not in str(
automl.get_configuration_space(X=X_train, y=y_train))
pickle.dumps(automl_fitted)
def test_classification_methods_returns_self(self):
X_train, y_train, X_test, y_test = putil.get_dataset('iris')
automl = AutoSklearn2Classifier(time_left_for_this_task=60, ensemble_size=0,)
automl_fitted = automl.fit(X_train, y_train)
self.assertIs(automl, automl_fitted)
automl_ensemble_fitted = automl.fit_ensemble(y_train, ensemble_size=5)
self.assertIs(automl, automl_ensemble_fitted)
automl_refitted = automl.refit(X_train.copy(), y_train.copy())
self.assertIs(automl, automl_refitted)
def init(df, param):
params = deepcopy(param['options']['params'])
params.pop('algo', None)
params.pop('mode', None)
params.pop('dataset_name', None)
for key in params:
try:
if params[key].isdigit():
params[key] = int(params[key])
except:
pass
model = {}
model["model"] = AutoSklearn2Classifier(**params)
return model
def test_autosklearn2_classification_methods_returns_self(dask_client):
X_train, y_train, X_test, y_test = putil.get_dataset('iris')
automl = AutoSklearn2Classifier(time_left_for_this_task=60,
ensemble_size=0,
dask_client=dask_client)
automl_fitted = automl.fit(X_train, y_train)
assert automl is automl_fitted
automl_ensemble_fitted = automl.fit_ensemble(y_train, ensemble_size=5)
assert automl is automl_ensemble_fitted
automl_refitted = automl.refit(X_train.copy(), y_train.copy())
assert automl is automl_refitted
predictions = automl_fitted.predict(X_test)
assert sklearn.metrics.accuracy_score(y_test, predictions) >= 2 / 3
pickle.dumps(automl_fitted)
to_encode = [
"Gender",
"Customer Type",
"Type of Travel",
"Class",
]
for col in to_encode:
orig_data[col] = LabelEncoder().fit_transform(orig_data[col])
orig_data["satisfaction"].replace("satisfied", 1, inplace=True)
orig_data["satisfaction"].replace("neutral or dissatisfied", 0, inplace=True)
scaler = StandardScaler()
for col in orig_data.columns:
if col != target:
orig_data[col] = scaler.fit_transform(orig_data[[col]])
data_X = orig_data.loc[:, [x for x in orig_data.columns if x != target]]
data_Y = orig_data.loc[:, target]
data_X_train, data_X_test, data_y_train, data_y_test = train_test_split(
data_X, data_Y, test_size=0.3, random_state=1)
cls = AutoSklearn2Classifier(time_left_for_this_task=60)
print("Fitting..")
cls.fit(data_X_train, data_y_train)
pred = cls.predict(data_X_test)
print("AutoML precision_score", precision_score(data_y_test, pred))
print("AutoML recall_score", recall_score(data_y_test, pred))
print("AutoML roc_auc_score", roc_auc_score(data_y_test, pred))