tensorflow/models · error · ValueError

Dataset must specify a path for the data files.

Error message

Dataset must specify a path for the data files.

What it means

Error "Dataset must specify a path for the data files." thrown in tensorflow/models.

Source

Thrown at official/legacy/image_classification/dataset_factory.py:367

        interleave_cycle_length=10,
        interleave_block_length=1,
        input_context=self.input_context)

    dataset = builder.as_dataset(
        split=self.config.split,
        as_supervised=True,
        shuffle_files=True,
        decoders=decoders,
        read_config=read_config)

    return dataset

  def load_records(self) -> tf.data.Dataset:
    """Return a dataset loading files with TFRecords."""
    logging.info('Using TFRecords to load data.')
    if self.config.filenames is None:
      if self.config.data_dir is None:
        raise ValueError('Dataset must specify a path for the data files.')

      file_pattern = os.path.join(self.config.data_dir,
                                  '{}*'.format(self.config.split))
      dataset = tf.data.Dataset.list_files(file_pattern, shuffle=False)
    else:
      dataset = tf.data.Dataset.from_tensor_slices(self.config.filenames)

    return dataset

  def load_synthetic(self) -> tf.data.Dataset:
    """Return a dataset generating dummy synthetic data."""
    logging.info('Generating a synthetic dataset.')

    def generate_data(_):
      image = tf.zeros([self.image_size, self.image_size, self.num_channels],
                       dtype=self.dtype)
      label = tf.zeros([1], dtype=tf.int32)
      return image, label

View on GitHub (pinned to e006f5f0d5)

When it happens

Trigger: Thrown at official/legacy/image_classification/dataset_factory.py:367 when the library encounters an invalid state.

Common situations: See trigger scenarios.

AI-assisted analysis of tensorflow/models@e006f5f0d5 (2026-08-24). Data as JSON: /api/errors/bbd3755efa777852. Report an issue: GitHub.