Fix gt priority tiny bug

[simonJJJ]

The original method calculates gt priority has a tiny bug, it actually outputs the rank, not the priority.
E.g. areas=tensor([200., 300., 500., 400., 100.])
The sort_idx would be tensor([2, 3, 1, 0, 4]). But the priority is actually tensor([3, 2, 0, 1, 4]) (Due to the gt areas). So I use an torch.argsort() method to get it.

[CLAassistant]

All committers have signed the CLA.

[Johnson-Wang]

Hi, thanks for the contribution. Have you run the baseline after fixing this?

[simonJJJ]

Hi, thanks for the contribution. Have you run the baseline after fixing this?

Nope. I also want to see the performance gap due to the fix.

[Johnson-Wang]

I have tried to re-benchmark the performance with the fix. The AP of Resnet50 does not change (still 37.4), while that of Resnet50 with 0.2-0.5 treated as ignored decreases a bit (from 37.0 to 36.0). Performance of other backbone is still pending.

[Johnson-Wang]

An update on the performance.

	R50-ignore range 0.2-0.5	R50	R101
Before fix	37.0	37.4	39.3
After fix	36.0	37.4	39.4

[Johnson-Wang]

Thanks for pointing this bug out. Most of configs are quite robust to this modification while the performance with 0.2-0.5 as ignored decreases a bit. I will update this log afterwards.