tensorflow/models · error · ValueError

batch_size cannot be None for panoptic segmentation model.

Error message

batch_size cannot be None for panoptic segmentation model.

What it means

Error "batch_size cannot be None for panoptic segmentation model." thrown in tensorflow/models.

Source

Thrown at official/projects/panoptic/serving/panoptic_deeplab.py:41

from official.projects.panoptic.modeling import panoptic_deeplab_model
from official.vision.serving import semantic_segmentation


class PanopticSegmentationModule(
    semantic_segmentation.SegmentationModule):
  """Panoptic Deeplab Segmentation Module."""

  def __init__(self,
               params: cfg.ExperimentConfig,
               *,
               model: tf_keras.Model,
               batch_size: int,
               input_image_size: List[int],
               num_channels: int = 3):
    """Initializes panoptic segmentation module for export."""

    if batch_size is None:
      raise ValueError('batch_size cannot be None for panoptic segmentation '
                       'model.')
    if not isinstance(model, panoptic_deeplab_model.PanopticDeeplabModel):
      raise ValueError('PanopticSegmentationModule module not '
                       'implemented for {} model.'.format(type(model)))
    params.task.train_data.preserve_aspect_ratio = True
    super(PanopticSegmentationModule, self).__init__(
        params=params,
        model=model,
        batch_size=batch_size,
        input_image_size=input_image_size,
        num_channels=num_channels)

  def _build_model(self):
    input_specs = tf_keras.layers.InputSpec(shape=[self._batch_size] +
                                            self._input_image_size + [3])

    return factory.build_panoptic_deeplab(
        input_specs=input_specs,

View on GitHub (pinned to e006f5f0d5)

When it happens

Trigger: Thrown at official/projects/panoptic/serving/panoptic_deeplab.py:41 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/b6b717981d42d9b2. Report an issue: GitHub.