def test_as_training_data_produces_correct_numpy_arrays_with_character_tokenization( self): instance = IndexedSentenceInstance([[1, 2], [3, 1, 2]], [3, 4]) instance.pad({'num_sentence_words': 3, 'num_word_characters': 4}) inputs, label = instance.as_training_data() assert_array_equal(inputs, [[0, 0, 0, 0], [1, 2, 0, 0], [3, 1, 2, 0]]) assert_array_equal(label, [[0], [3], [4]])
def test_as_training_data_produces_correct_numpy_arrays(self): instance = IndexedSentenceInstance([1, 2, 3, 4], [2, 3, 4, 5]) inputs, label = instance.as_training_data() assert_array_equal(inputs, [1, 2, 3, 4]) assert_array_equal(label, [[2], [3], [4], [5]])