Issues

• Why multiple 0.25?

  strongwolf/DW#42 · Filter by author yuanqianguang opened on May 6, 2024

• How to apply DW to two-stage detectors？

  strongwolf/DW#40 · Filter by author seacloud-0420 opened on Aug 24, 2023
  1

• Are mean and sigma in CenterPrior learnable?

  strongwolf/DW#39 · Filter by author Yuanyang-Zhu opened on Jul 3, 2023
  1

• About the version of mmcv

  strongwolf/DW#38 · Filter by author liangabcd opened on May 20, 2023
  1

• oom问题

  strongwolf/DW#37 · Filter by author Yangjie0610 opened on May 11, 2023

• weight design

  strongwolf/DW#36 · Filter by author gnftau opened on Apr 15, 2023
  3

• how to understand this

  strongwolf/DW#35 · Filter by author sqdcbbj opened on Mar 23, 2023

• Welcome update to OpenMMLab 2.0

  strongwolf/DW#34 · Filter by author vansin opened on Mar 14, 2023

• At inference, a neg prediction in the ranking list will not affect the recall but decrease the precision.

  strongwolf/DW#31 · Filter by author ycyinchao opened on Dec 2, 2022
  1

• Why use neg_avg_factor = (1 - p_neg_weight).sum() ?

  strongwolf/DW#30 · Filter by author Icecream-blue-sky opened on Nov 28, 2022
  1

• Expect dw to bring more gain in yolov6！

  strongwolf/DW#16 · Filter by author meituan-gengyifei opened on Jun 29, 2022
  2

• 运行出现错误

  strongwolf/DW#13 · Filter by author sharp-xiang opened on May 28, 2022
  3

• Top-k method for selecting candidate bags

  strongwolf/DW#6 · Filter by author soHardToHaveAName opened on Apr 24, 2022

• how dose DW perform on Yolov5

  strongwolf/DW#2 · Filter by author CangHaiQingYue opened on Apr 2, 2022
  1