Vectorize box decoding in FCOS model

[jdsgomes]

🚀 The feature

The for loop in torchvision/models/detection/fcos.py:80 can probably be vetorized.

        pred_boxes = [
            self.box_coder.decode_single(bbox_regression_per_image, anchors_per_image)
            for anchors_per_image, bbox_regression_per_image in zip(anchors, bbox_regression)
        ]

Motivation, pitch

Simplify/speedup the execution

Alternatives

No response

Additional context

No response

[abhi-glitchhg]

Can I try this?

[datumbox]

@abhi-glitchhg absolutely! Thanks for offering to help.