AttributeError: module 'supervision' has no attribute 'create_tiles'

[MeedocTheMedic]

Search before asking

• I have searched the Supervision issues and found no similar bug report.

Bug

AttributeError Traceback (most recent call last)
/tmp/ipython-input-1680823999.py in <cell line: 0>()
24 annotated_images.append(annotated_image)
25
---> 26 grid = sv.create_tiles(
27 annotated_images,
28 grid_size=(GRID_SIZE, GRID_SIZE),

AttributeError: module 'supervision' has no attribute 'create_tiles'

Environment

Google Colab Notebook:

• Supervision: 0.27.0rc1
• Python: 3.12.11
  Also Tested in Kaggle:
• Supervision: 0.27.0rc1
• Python: 3.11.13
  Have also tried uninstalling and reinstalling Supervision: 0.25.0 but didn't work

Minimal Reproducible Example

I was following the train-rt-detr-on-custom-dataset-with-transformers.ipynb tutorial from Roboflow from this website: https://blog.roboflow.com/train-rt-detr-custom-dataset-transformers/, which has a link to the following colab notebook https://colab.research.google.com/github/roboflow-ai/notebooks/blob/main/notebooks/train-rt-detr-on-custom-dataset-with-transformers.ipynb?ref=blog.roboflow.com. It was using an older version of supervision which still had BoundingBoxAnnotator instead of BoxAnnotator, but I was able to run it after making those changes. Despite being able to get it working, it now suddenly no longer works and after checking the changelogs I cannot find anything on sv.create_tiles being removed, but I know it definitely used to exist because this code used to work. Unfortunately, after reaching the "Display dataset sample" section it broke through the following code:

  GRID_SIZE = 5
  
  def annotate(image, annotations, classes):
      labels = [
          classes[class_id]
          for class_id
          in annotations.class_id
      ]

    bounding_box_annotator = sv.BoxAnnotator()
    label_annotator = sv.LabelAnnotator(text_scale=1, text_thickness=2)

    annotated_image = image.copy()
    annotated_image = bounding_box_annotator.annotate(annotated_image, annotations)
    annotated_image = label_annotator.annotate(annotated_image, annotations, labels=labels)
    return annotated_image

  annotated_images = []
  for i in range(GRID_SIZE * GRID_SIZE):
      _, image, annotations = ds_train[i]
      annotated_image = annotate(image, annotations, ds_train.classes)
      annotated_images.append(annotated_image)
  
  grid = sv.create_tiles(
      annotated_images,
      grid_size=(GRID_SIZE, GRID_SIZE),
      single_tile_size=(400, 400),
      tile_padding_color=sv.Color.WHITE,
      tile_margin_color=sv.Color.WHITE
  )
  sv.plot_image(grid, size=(10, 10))`

`AttributeError Traceback (most recent call last)
/tmp/ipython-input-1680823999.py in <cell line: 0>()
24 annotated_images.append(annotated_image)
25
---> 26 grid = sv.create_tiles(
27 annotated_images,
28 grid_size=(GRID_SIZE, GRID_SIZE),

AttributeError: module 'supervision' has no attribute 'create_tiles'`

Additional

sv.create_tiles() used to work but now has disappeared.