Running test_RFB.py with your provided model, RFBNet300_VOC_80_7.pth, get a mAP of 0.8056 not 80.7% posted in your github.

Question

ZhiqiJiang opened this issue 6 years ago · 6 comments

The command is python3 test_RFB.py --trained_model RFBNet300_VOC_80_7.pth. And the result is showed as below:

Answer 1 · 2019-05-18T12:27:38.000Z

hellow,I run the command same with yours, but my APs of 20 classes and mAP are 0. Did you meet the same problem before the current results.

Answer 2 · 2019-06-11T08:16:59.000Z

@ love1314pds No I didn't. Maybe I modified something correspondingly.

Answer 3 · 2019-06-11T08:32:06.000Z

@love1314pds 我已经不止一次遇到跨框架引起精度损失这个问题了,就算是利用作者提供的pretrained model.我一直还没有耐下性子去定位分析乃至解决这个问题.

Answer 4 · 2019-06-12T06:18:43.000Z

嗯，好的，谢谢

Answer 5 · 2020-03-13T16:14:22.000Z

I use the same commad but ge't a map=0.459 But I didnot change anything~

Answer 6 · 2020-08-28T07:09:59.000Z

Ps o

I have the same problem. Did you solve it?