def test_concat(self):
dmd1 = self.get_data(is_classification=False)
dmd2 = self.get_data(is_classification=True)
self.assertEqual(dmd1.n_samples, dmd2.n_samples)
dmd = DMD.concat([dmd1, dmd2], axis=0)
self.assertEqual(dmd.n_samples, 2 * dmd2.n_samples)
self.assertEqual(dmd._x.shape[0], 2 * dmd1._y.shape[0])
self.assertEqual(dmd._x.shape[0], 2 * dmd1._samples_meta.shape[0])
self.assertEqual(dmd.n_features, dmd2.n_features)
def prepare_dataset_for_score_quality(cls, dmd_train: DMD, dmd_test: DMD):
'''
:param dmd_train: train set
:param dmd_test: test set
:return: dataset with target of test/train
'''
dmd = DMD.concat([dmd_train, dmd_test])
new_label = [0] * dmd_train.n_samples + [1] * dmd_test.n_samples
dmd.set_target(new_label)
train, test = dmd.split(ratio=dmd_test.n_samples / (dmd_train.n_samples + dmd_test.n_samples))
return train, test