def test_check_lengths_strict_no_exception(self):
x = [{"inputs": [9, 4, 3, 8, 1], "targets": [3, 9, 4, 5]}]
ds = create_default_dataset(x)
task_feature_lengths = {"inputs": 5, "targets": 4}
ds = feature_converters._check_lengths(ds,
task_feature_lengths,
strict=True,
error_label="initial")
list(ds.as_numpy_iterator())
def test_check_lengths_extra_features(self):
x = [{"targets": [3, 9, 4, 5], "targets_pretokenized": "some text"}]
output_types = {"targets": tf.int64, "targets_pretokenized": tf.string}
output_shapes = {"targets": [4], "targets_pretokenized": []}
ds = tf.data.Dataset.from_generator(lambda: x,
output_types=output_types,
output_shapes=output_shapes)
task_feature_lengths = {"targets": 4}
ds = feature_converters._check_lengths(ds,
task_feature_lengths,
strict=True,
error_label="initial")
list(ds.as_numpy_iterator())
def test_check_lengths_strict_exception(self):
x = [{"inputs": [9, 4, 3, 8, 1], "targets": [3, 9, 4, 5]}]
ds = create_default_dataset(x)
task_feature_lengths = {"inputs": 7, "targets": 4}
expected_msg = (
r".*Feature \\'inputs\\' has length not equal to the expected length of"
r" 7 during initial validation.*")
with self.assertRaisesRegex(tf.errors.InvalidArgumentError,
expected_msg):
ds = feature_converters._check_lengths(ds,
task_feature_lengths,
strict=True,
error_label="initial")
list(ds.as_numpy_iterator())