tensorflow/models · error · ValueError

`maskrcnn.Parser` only supports `pad = True`.

Error message

`maskrcnn.Parser` only supports `pad = True`.

What it means

Error "`maskrcnn.Parser` only supports `pad = True`." thrown in tensorflow/models.

Source

Thrown at official/vision/configs/maskrcnn.py:55

  aug_rand_vflip: bool = False
  aug_scale_min: float = 1.0
  aug_scale_max: float = 1.0
  aug_type: Optional[
      common.Augmentation] = None  # Choose from AutoAugment and RandAugment.
  skip_crowd_during_training: bool = True
  max_num_instances: int = 100
  rpn_match_threshold: float = 0.7
  rpn_unmatched_threshold: float = 0.3
  rpn_batch_size_per_im: int = 256
  rpn_fg_fraction: float = 0.5
  mask_crop_size: int = 112
  pad: bool = True  # Only support `pad = True`.
  keep_aspect_ratio: bool = True  # Only support `keep_aspect_ratio = True`.

  def __post_init__(self, *args, **kwargs):
    """Validates the configuration."""
    if not self.pad:
      raise ValueError('`maskrcnn.Parser` only supports `pad = True`.')
    if not self.keep_aspect_ratio:
      raise ValueError(
          '`maskrcnn.Parser` only supports `keep_aspect_ratio = True`.'
      )
    super().__post_init__(*args, **kwargs)


@dataclasses.dataclass
class DataConfig(cfg.DataConfig):
  """Input config for training."""
  input_path: Union[Sequence[str], str, hyperparams.Config] = ''
  weights: Optional[hyperparams.Config] = None
  global_batch_size: int = 0
  is_training: bool = False
  dtype: str = 'bfloat16'
  decoder: common.DataDecoder = dataclasses.field(
      default_factory=common.DataDecoder
  )

View on GitHub (pinned to e006f5f0d5)

When it happens

Trigger: Thrown at official/vision/configs/maskrcnn.py:55 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/ddfc7530e4a1ad8e. Report an issue: GitHub.