def fit(self, multilabel_dataset):
features = get_instance_features_train(multilabel_dataset)
labels = get_instance_labels_train(multilabel_dataset)
self.classifier.fit(features, labels)
return self
def fit(self, multilabel_dataset):
features = get_instance_features_train(multilabel_dataset)
labels = get_instance_labels_train(multilabel_dataset)
features_matrix = csr_matrix(features)
self.selector_train.fit(features_matrix, labels)
self.selector_test.fit(features_matrix, labels)
return self.selector_train
def partition_labels(multilabel_dataset):
total_labels = get_total_labels_train(multilabel_dataset)
instance_labels = get_instance_labels_train(multilabel_dataset)
partitioned = []
for column_number in range(0, len(total_labels)):
column_as_list = get_column_as_list(instance_labels, column_number)
partitioned.append(column_as_list)
return partitioned
def test_get_instance_labels_train(self):
labels = get_instance_labels_train(self.multilabel_dataset)
assert type(labels) == type(
[]), "instance_labels_train should be a list"