tensorflow/models · error · ValueError

Cannot finalize with {finalize_method[i]}.

Error message

Cannot finalize with {finalize_method[i]}.

What it means

Error "Cannot finalize with {finalize_method[i]}." thrown in tensorflow/models.

Source

Thrown at official/projects/edgetpu/vision/serving/export_util.py:139

    return output_layer
  discrete = False
  for i in range(len(finalize_method)):
    if finalize_method[i] == 'argmax':
      discrete = True
      is_argmax_last = (i + 1) == len(finalize_method)
      if is_argmax_last:
        output_layer = tf.argmax(
            output_layer, axis=3, output_type=tf.dtypes.int32)
      else:
        # TODO(tohaspiridonov): add first_match=False when cl/383951533 submited
        output_layer = custom_layers.argmax(
            output_layer, keepdims=True, epsilon=1e-3)
    elif finalize_method[i] == 'squeeze':
      output_layer = tf.squeeze(output_layer, axis=3)
    else:
      resize_params = finalize_method[i].split('resize')
      if len(resize_params) != 2 or resize_params[0]:
        raise ValueError('Cannot finalize with ' + finalize_method[i] + '.')
      resize_to_size = int(resize_params[1])
      if discrete:
        output_layer = tf.image.resize(
            output_layer, [resize_to_size, resize_to_size],
            method=tf.image.ResizeMethod.NEAREST_NEIGHBOR)
      else:
        output_layer = tf.image.resize(
            output_layer, [resize_to_size, resize_to_size],
            method=tf.image.ResizeMethod.BILINEAR)
  return output_layer


def preprocess_for_quantization(image_data, image_size, crop_padding=32):
  """Crops to center of image with padding then scales, normalizes image_size.

  Args:
    image_data: A 3D Tensor representing the RGB image data. Image can be of
      arbitrary height and width.

View on GitHub (pinned to e006f5f0d5)

When it happens

Trigger: Thrown at official/projects/edgetpu/vision/serving/export_util.py:139 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/6fbb02b016b1efa7. Report an issue: GitHub.