MeanAveragePrecision.tm_to_coco example fails with IndexError: list index out of range

[robmarkcole]

🐛 Bug

There appears to be a bug, the example doesn't include masks but this is raised

To Reproduce

Run the example from the docs

from torch import tensor
from torchmetrics.detection import MeanAveragePrecision
preds = [
  dict(
    boxes=tensor([[258.0, 41.0, 606.0, 285.0]]),
    scores=tensor([0.536]),
    labels=tensor([0]),
  )
]
target = [
  dict(
    boxes=tensor([[214.0, 41.0, 562.0, 285.0]]),
    labels=tensor([0]),
  )
]
metric = MeanAveragePrecision()
metric.update(preds, target)
metric.tm_to_coco("tm_map_input")  

---------------------------------------------------------------------------
IndexError                                Traceback (most recent call last)
[<ipython-input-2-d5b2db9d9305>](https://localhost:8080/#) in <cell line: 18>()
     16 metric = MeanAveragePrecision()
     17 metric.update(preds, target)
---> 18 metric.tm_to_coco("tm_map_input")

1 frames
[/usr/local/lib/python3.10/dist-packages/torchmetrics/detection/mean_ap.py](https://localhost:8080/#) in _get_coco_format(self, labels, boxes, masks, scores, crowds, area)
    865                 image_boxes = image_boxes.cpu().tolist()
    866             if masks is not None:
--> 867                 image_masks = masks[image_id]
    868                 if len(image_masks) == 0 and boxes is None:
    869                     continue

IndexError: list index out of range

Expected behavior

A pair of json are written

Environment

• TorchMetrics version 1.4.0.post0
• Fresh Google colab

Additional context

NA

[robmarkcole]

Excellent, tested and the files are now generated