tensorflow/models · error · ValueError

Must use drop_remainder=True with CombinationDatasetInputRea

Error message

Must use drop_remainder=True with CombinationDatasetInputReader

What it means

Error "Must use drop_remainder=True with CombinationDatasetInputReader" thrown in tensorflow/models.

Source

Thrown at official/vision/dataloaders/input_reader.py:193

    """
    super().__init__(
        params=params,
        dataset_fn=dataset_fn,
        decoder_fn=decoder_fn,
        combine_fn=combine_fn,
        sample_fn=sample_fn,
        parser_fn=parser_fn,
        transform_and_batch_fn=transform_and_batch_fn,
        postprocess_fn=postprocess_fn)

    self._pseudo_label_file_pattern = params.pseudo_label_data.input_path
    self._pseudo_label_dataset_fn = pseudo_label_dataset_fn
    self._pseudo_label_data_ratio = params.pseudo_label_data.data_ratio
    self._pseudo_label_batch_size = params.pseudo_label_data.global_batch_size
    self._pseudo_label_matched_files = input_reader.match_files(
        self._pseudo_label_file_pattern)
    if not self._drop_remainder:
      raise ValueError(
          'Must use drop_remainder=True with CombinationDatasetInputReader')

  def read(
      self,
      input_context: Optional[tf.distribute.InputContext] = None
  ) -> tf.data.Dataset:
    """Generates a tf.data.Dataset object."""

    labeled_batch_size, pl_batch_size = calculate_batch_sizes(
        self._global_batch_size, self._pseudo_label_data_ratio,
        self._pseudo_label_batch_size)

    if not labeled_batch_size and pl_batch_size:
      raise ValueError(
          'Invalid batch_size: {} and pseudo_label_data_ratio: {}, '
          'resulting in a 0 batch size for one of the datasets.'.format(
              self._global_batch_size, self._pseudo_label_data_ratio))

View on GitHub (pinned to e006f5f0d5)

When it happens

Trigger: Thrown at official/vision/dataloaders/input_reader.py:193 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/8327749901c2874b. Report an issue: GitHub.