tensorflow/models · error · ValueError

Eval metric being checked against stop conditions must be a

Error message

Eval metric being checked against stop conditions must be a number.

What it means

Error "Eval metric being checked against stop conditions must be a number." thrown in tensorflow/models.

Source

Thrown at official/utils/misc/model_helpers.py:46

  Args:
    stop_threshold: float, the threshold above which a model should stop
      training.
    eval_metric: float, the current value of the relevant metric to check.

  Returns:
    True if training should stop, False otherwise.

  Raises:
    ValueError: if either stop_threshold or eval_metric is not a number
  """
  if stop_threshold is None:
    return False

  if not isinstance(stop_threshold, numbers.Number):
    raise ValueError("Threshold for checking stop conditions must be a number.")
  if not isinstance(eval_metric, numbers.Number):
    raise ValueError("Eval metric being checked against stop conditions "
                     "must be a number.")

  if eval_metric >= stop_threshold:
    logging.info("Stop threshold of {} was passed with metric value {}.".format(
        stop_threshold, eval_metric))
    return True

  return False


def generate_synthetic_data(input_shape,
                            input_value=0,
                            input_dtype=None,
                            label_shape=None,
                            label_value=0,
                            label_dtype=None):
  """Create a repeating dataset with constant values.

View on GitHub (pinned to e006f5f0d5)

When it happens

Trigger: Thrown at official/utils/misc/model_helpers.py:46 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/40e6a6608cfe9b0f. Report an issue: GitHub.