tensorflow/models · error · ValueError

{} is not unique across image directories

Error message

{} is not unique across image directories

What it means

Error "{} is not unique across image directories" thrown in tensorflow/models.

Source

Thrown at official/vision/data/create_coco_tf_record.py:311

    num_annotations_skipped: Number of (invalid) annotations that were ignored.

  Raises:
    ValueError: if the image pointed to by data['filename'] is not a valid JPEG,
      does not exist, or is not unique across image directories.
  """
  image_height = image['height']
  image_width = image['width']
  filename = image['file_name']
  image_id = image['id']

  if len(image_dirs) > 1:
    full_paths = [os.path.join(image_dir, filename) for image_dir in image_dirs]
    full_existing_paths = [p for p in full_paths if tf.io.gfile.exists(p)]
    if not full_existing_paths:
      raise ValueError(
          '{} does not exist across image directories.'.format(filename))
    if len(full_existing_paths) > 1:
      raise ValueError(
          '{} is not unique across image directories'.format(filename))
    full_path, = full_existing_paths
  # If there is only one image directory, it's not worth checking for existence,
  # since trying to open the file will raise an informative error message if it
  # does not exist.
  else:
    image_dir, = image_dirs
    full_path = os.path.join(image_dir, filename)

  with tf.io.gfile.GFile(full_path, 'rb') as fid:
    encoded_jpg = fid.read()

  feature_dict = tfrecord_lib.image_info_to_feature_dict(
      image_height, image_width, filename, image_id, encoded_jpg, 'jpg')

  num_annotations_skipped = 0
  if bbox_annotations:
    box_feature_dict, num_skipped = bbox_annotations_to_feature_dict(

View on GitHub (pinned to e006f5f0d5)

When it happens

Trigger: Thrown at official/vision/data/create_coco_tf_record.py:311 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/48a4f45c20c7b715. Report an issue: GitHub.