tensorflow/models · error · ValueError

segmentation_groundtruth_padded_size ([height, width]) needs

Error message

segmentation_groundtruth_padded_size ([height, width]) needs to bespecified when segmentation_resize_eval_groundtruth is False.

What it means

Error "segmentation_groundtruth_padded_size ([height, width]) needs to bespecified when segmentation_resize_eval_groundtruth is False." thrown in tensorflow/models.

Source

Thrown at official/projects/maskconver/dataloaders/multiscale_maskconver_input.py:145

        by the PQ evaluator.
      level: `int`, output level, used to generate target assignments.
      mask_target_level: `int`, target level for panoptic masks, default is 1.
      num_panoptic_categories: `int`, number of panoptic categories.
      num_thing_categories: `int1, number of thing categories.
      gaussian_iou: `float`, used for generating the center heatmaps.
      aug_type: An optional Augmentation object with params for AutoAugment.
      dtype: `str`, data type. One of {`bfloat16`, `float32`, `float16`}.
    """
    super().__init__()

    self.aug_rand_hflip = aug_rand_hflip
    self._segmentation_resize_eval_groundtruth = (
        segmentation_resize_eval_groundtruth
    )
    if (not segmentation_resize_eval_groundtruth) and (
        segmentation_groundtruth_padded_size is None
    ):
      raise ValueError(
          'segmentation_groundtruth_padded_size ([height, width]) needs to be'
          'specified when segmentation_resize_eval_groundtruth is False.'
      )
    self._segmentation_groundtruth_padded_size = (
        segmentation_groundtruth_padded_size
    )
    self._dtype = dtype
    self._max_num_instances = max_num_instances
    self._output_size = output_size
    self._min_level = min_level
    self._max_level = max_level
    self._aug_scale_min = aug_scale_min
    self._aug_scale_max = aug_scale_max
    self._segmentation_ignore_label = segmentation_ignore_label
    self._panoptic_ignore_label = panoptic_ignore_label
    self.level = level
    self._mask_target_level = mask_target_level
    self._num_panoptic_categories = num_panoptic_categories

View on GitHub (pinned to e006f5f0d5)

When it happens

Trigger: Thrown at official/projects/maskconver/dataloaders/multiscale_maskconver_input.py:145 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/15b4a32d11e501b7. Report an issue: GitHub.