tensorflow/models · error · ValueError

only call `run_customized_training_loop()` with named argume

Error message

only call `run_customized_training_loop()` with named arguments.

What it means

Error "only call `run_customized_training_loop()` with named arguments." thrown in tensorflow/models.

Source

Thrown at official/legacy/bert/model_training_utils.py:218

      allreduce_bytes_per_pack: A non-negative integer. Breaks collective
        operations into packs of certain size. If it's zero, all gradients are
        in one pack. Breaking gradient into packs could enable overlap between
        allreduce and backprop computation. This flag only takes effect when
        explicit_allreduce is set to True.'

  Returns:
      Trained model.

  Raises:
      ValueError: (1) When model returned by `model_fn` does not have optimizer
        attribute or when required parameters are set to none. (2) eval args are
        not specified correctly. (3) metric_fn must be a callable if specified.
        (4) sub_model_checkpoint_name is specified, but `sub_model` returned
        by `model_fn` is None.
  """

  if _sentinel is not None:
    raise ValueError('only call `run_customized_training_loop()` '
                     'with named arguments.')

  required_arguments = [
      strategy, model_fn, loss_fn, model_dir, steps_per_epoch, train_input_fn
  ]

  steps_between_evals = int(steps_per_epoch / num_eval_per_epoch)  # pyrefly: ignore[unsupported-operation]
  if [arg for arg in required_arguments if arg is None]:
    raise ValueError('`strategy`, `model_fn`, `loss_fn`, `model_dir`, '
                     '`steps_per_epoch` and `train_input_fn` are required '
                     'parameters.')
  if not steps_per_loop:
    if tf.config.list_logical_devices('TPU'):
      # One can't fully utilize a TPU with steps_per_loop=1, so in this case
      # default users to a more useful value.
      steps_per_loop = min(1000, steps_between_evals)
    else:
      steps_per_loop = 1

View on GitHub (pinned to e006f5f0d5)

When it happens

Trigger: Thrown at official/legacy/bert/model_training_utils.py:218 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/695a16e8004d63e9. Report an issue: GitHub.