def test_reader__vectorize_empty(self): """Check vectorize method on the empty""" reader = DSReader(dataset_empty_path) reader.make_dictionary() X, y = reader.vectorize() self.assertEqual([], X.tolist()) self.assertEqual([], y.tolist())
def test_reader__split_test_and_train_data_zero_size(self): """Check split_test_and_train_data method with argument size equals to zero""" reader = DSReader(dataset_split_path) reader.make_dictionary() X, y = reader.vectorize() percent = 0 with self.assertRaises(Exception): X_train, y_train, X_test, y_test = reader.split_train_and_test( X, y, percent)
def test_reader__split_test_and_train_data_empty(self): """Check split_test_and_train_data method on the empty dataset""" reader = DSReader(dataset_empty_path) reader.make_dictionary() X, y = reader.vectorize() percent = 0.7 with self.assertRaises(Exception): X_train, y_train, X_test, y_test = reader.split_train_and_test( X, y, percent)
def test_reader__split_test_and_train_data(self): """Check split_test_and_train_data method""" reader = DSReader(dataset_split_path) reader.make_dictionary() X, y = reader.vectorize() percent = 0.7 X_train, y_train, X_test, y_test = reader.split_train_and_test( X, y, percent) self.assertEqual(X_train.shape[0], X.shape[0] * percent) self.assertEqual(X_test.shape[0], X.shape[0] * round(1 - percent, 2)) self.assertEqual(y_train.shape[0], y.shape[0] * percent) self.assertEqual(y_test.shape[0], y.shape[0] * round(1 - percent, 2))
def test_reader__vectorize(self): """Check vectorize method""" x_result = [ [ 'During this webinar we will cover what is DevOps and Cloud Native' ], ['New webinar came up'], [ 'During this webinar we will cover what is DevOps and Cloud Native and storage' ] ] y_result = [1, 1, 0] reader = DSReader(dataset_path) reader.make_dictionary() X, y = reader.vectorize() for i, row in enumerate(X): self.assertEqual(x_result[i], row.tolist()) self.assertEqual(y_result, y.tolist())
# my_dataset.remove_stopwords() # print(my_dataset.dataset) my_dataset1 = DSReader('C:/Users/Masquerade/Downloads/emails.csv') my_dataset1.to_lower() my_dataset1.remove_digits() my_dataset1.remove_punctuation_marks() my_dataset1.remove_duplicates() my_dataset1.remove_stopwords() my_dataset1.remove_stopwords() # print(my_dataset1.dataset) list_email, list_label = my_dataset1.vectorize() print(list_email.shape) print(list_label.shape) X, y = list_email, list_label # X, y = my_dataset1.dataset.email, my_dataset1.dataset.label X_train, X_test, y_train, y_test = train_test_split(X.values, y.values) print("______________________________________________") print(y_test) print("______________________________________________") vectorizer = CountVectorizer() counts = vectorizer.fit_transform(X_train.ravel()) print("______________________________________________")