tensorflow/models · error · ValueError

Expected the input masks (..., height, width) has rank >= 2,

Error message

Expected the input masks (..., height, width) has rank >= 2, was: %s

What it means

Error "Expected the input masks (..., height, width) has rank >= 2, was: %s" thrown in tensorflow/models.

Source

Thrown at official/projects/panoptic/modeling/layers/panoptic_segmentation_generator.py:40

from official.vision.ops import spatial_transform_ops


def _batch_count_ones(masks: tf.Tensor,
                      dtype: tf.dtypes.DType = tf.int32) -> tf.Tensor:
  """Counts the ones/trues for each mask in the batch.

  Args:
    masks: A tensor in shape (..., height, width) with arbitrary numbers of
      batch dimensions.
    dtype: DType of the resulting tensor. Default is tf.int32.

  Returns:
    A tensor which contains the count of non-zero elements for each mask in the
    batch. The rank of the resulting tensor is equal to rank(masks) - 2.
  """
  masks_shape = masks.get_shape().as_list()
  if len(masks_shape) < 2:
    raise ValueError(
        'Expected the input masks (..., height, width) has rank >= 2, was: %s' %
        masks_shape)
  return tf.reduce_sum(tf.cast(masks, dtype), axis=[-2, -1])


class PanopticSegmentationGenerator(tf_keras.layers.Layer):
  """Panoptic segmentation generator layer."""

  def __init__(
      self,
      output_size: List[int],
      max_num_detections: int,
      stuff_classes_offset: int,
      mask_binarize_threshold: float = 0.5,
      score_threshold: float = 0.5,
      things_overlap_threshold: float = 0.5,
      stuff_area_threshold: float = 4096,
      things_class_label: int = 1,

View on GitHub (pinned to e006f5f0d5)

When it happens

Trigger: Thrown at official/projects/panoptic/modeling/layers/panoptic_segmentation_generator.py:40 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/f75cd11b6fbc5350. Report an issue: GitHub.