def transform(self, multilabel_dataset):
features_train = get_instance_features_train(multilabel_dataset)
selected_train = self.selector_train.transform(features_train)
set_instance_features_train(multilabel_dataset, selected_train)
self.filter_total_features_and_update_train(multilabel_dataset)
features_test = get_instance_features_test(multilabel_dataset)
selected_test = self.selector_test.transform(features_test)
set_instance_features_test(multilabel_dataset, selected_test)
self.filter_total_features_and_update_test(multilabel_dataset)
return multilabel_dataset
def classify(self, multilabel_dataset):
features = get_instance_features_test(multilabel_dataset)
results = list(self.classifier.predict(features))
return results
def create_features_dataframe(multilabel_dataset):
features = get_instance_features_test(multilabel_dataset)
total_features = get_total_features_test(multilabel_dataset)
feature_names = Stream(total_features).map(lambda elem: elem[0]).as_list()
return DataFrame(features, columns=feature_names)
def test_get_total_features_test(self):
features = get_instance_features_test(self.multilabel_dataset)
assert type(features) == type(
[]), "total_features_test should be a list"