tensorflow/models · error · ValueError

Unrecognized augmentation_type: %s

Error message

Unrecognized augmentation_type: %s

What it means

Error "Unrecognized augmentation_type: %s" thrown in tensorflow/models.

Source

Thrown at official/projects/videoglue/datasets/common/utils.py:436

          fn=random_square_crop_by_scale_fn,
          feature_name=output_feature_name,
          fn_name='{}_random_square_crop_by_scale'.format(output_feature_name),
          stateful=sync_random_state)
      def resize_and_pad_fn(x, state=None):
        return processors.resize_and_pad(
            frames=x,
            max_resize=crop_size,
            pad_size=crop_size,
            random=True,
            state=state)
      preprocessor_builder.add_fn(
          fn=resize_and_pad_fn,
          feature_name=output_feature_name,
          fn_name='{}_resize_random_pad'.format(output_feature_name),
          # Use state to keep coherence between modalities if requested.
          stateful=sync_random_state)
    else:
      raise ValueError('Unrecognized augmentation_type: %s' %
                       augmentation_type)

    if random_flip_image:
      def random_flip_left_right_fn(x, state=None):
        return processors.random_flip_left_right(
            x, seed=seed, is_flow=False, state=state)
      preprocessor_builder.add_fn(
          fn=random_flip_left_right_fn,
          feature_name=output_feature_name,
          fn_name='{}_random_flip'.format(output_feature_name),
          stateful=sync_random_state)
  else:
    # Crop images, either a 3-view crop or a central crop.
    if multi_crop:
      resize_smallest_fn = functools.partial(
          processors.resize_smallest,
          min_resize=min_resize,
          is_flow=False)

View on GitHub (pinned to e006f5f0d5)

When it happens

Trigger: Thrown at official/projects/videoglue/datasets/common/utils.py:436 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/81137850c9c87789. Report an issue: GitHub.