tensorflow/models · error · ValueError

`tfds_name` is %s, but `tfds_split` is not specified.

Error message

`tfds_name` is %s, but `tfds_split` is not specified.

What it means

Error "`tfds_name` is %s, but `tfds_split` is not specified." thrown in tensorflow/models.

Source

Thrown at official/core/input_reader.py:283

    """
    if params.input_path and params.tfds_name:
      raise ValueError('At most one of `input_path` and `tfds_name` can be '
                       'specified, but got %s and %s.' %
                       (params.input_path, params.tfds_name))

    if (isinstance(params.input_path, cfg.base_config.Config) or
        isinstance(params.tfds_name, cfg.base_config.Config)
        ) and combine_fn is None:
      raise ValueError(
          'A combine_fn is required if `input_path` or `tfds_name` is a dict.')

    self._tfds_name = params.tfds_name
    self._tfds_data_dir = params.tfds_data_dir
    self._matched_files = None
    if not params.input_path:
      # Read dataset from TFDS.
      if not params.tfds_split:
        raise ValueError(
            '`tfds_name` is %s, but `tfds_split` is not specified.' %
            params.tfds_name)
    else:
      self._matched_files = self.get_files(params.input_path)

    self._global_batch_size = params.global_batch_size
    self._is_training = params.is_training
    self._drop_remainder = params.drop_remainder
    self._shuffle_buffer_size = params.shuffle_buffer_size
    self._cache = params.cache
    self._cycle_length = params.cycle_length
    self._block_length = params.block_length
    self._deterministic = params.deterministic
    self._sharding = params.sharding
    self._tfds_split = params.tfds_split
    self._tfds_as_supervised = params.tfds_as_supervised
    self._tfds_skip_decoding_feature = params.tfds_skip_decoding_feature

View on GitHub (pinned to e006f5f0d5)

When it happens

Trigger: Thrown at official/core/input_reader.py:283 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/fd237b3bdb9818fd. Report an issue: GitHub.