MaskRCNN: Inconsistency between documentation and output

[Adityav2410]

As per mask RCNN documentation, the predicted masks have the shape:
masks (Tensor[N, H, W]):

however as per the tutorial, https://colab.research.google.com/github/pytorch/vision/blob/temp-tutorial/tutorials/torchvision_finetuning_instance_segmentation.ipynb#scrollTo=M58J3O9OtT1G,
and the results/output i see from the model, the masks have the shape: [N, 1, H, W]

The masks are predicted as [N, 1, H, W]
i.e. i output masks have 4 dim, whereas documentation has 3dim..

[fmassa]

Yes, this is a problem with the documentation.

Could you send a PR fixing it?