Invitation of contributing to MMSegmentation.

[MengzhangLI]

Hi, first congrats for acceptance of ICLR'21. Loss function and metric are very important for semantic segmentation especially those have too much foreground classes numbers.

We are members of OpenMMLab whose codebase MMSegmentation is related with your excellent work. Currently we do not have too much metrics (only mIoU, FScore etc), would you like to join us to make a pr about metric(here is one example of FScore made by one warm-heated community member)?

Looking forward to your reply!

Best,