Deformable-ConvNets-v1-v2-caffe icon indicating copy to clipboard operation
Deformable-ConvNets-v1-v2-caffe copied to clipboard

Published 20 hours ago • verified publisher icon zhanglonghao1992

something wrong in my code

Open lucinyaLi opened this issue 6 years ago • 3 comments
trafficstars

hello, I have met some trouble in my code when I use the deformable convolution in SSD, the loss is always high and the mAP is very very low. I tried to change the deformable_group, but the solution has no change. So is there some trick in using the deformable convolution? And in the code, is the BN layer indispensable?

lucinyaLi avatar Apr 30 '19 13:04 lucinyaLi

I dont think I used any tricks training the deformable convnet. I suggest you print the “offsets” of every iter and see if they are reasonable.

zhanglonghao1992 avatar May 01 '19 12:05 zhanglonghao1992

I dont think I used any tricks training the deformable convnet. I suggest you print the “offsets” of every iter and see if they are reasonable.

Have you tried to train a net from random Initialization ？

lucinyaLi avatar May 03 '19 05:05 lucinyaLi

I dont think I used any tricks training the deformable convnet. I suggest you print the “offsets” of every iter and see if they are reasonable.

Have you tried to train a net from random Initialization ？

No

zhanglonghao1992 avatar May 03 '19 07:05 zhanglonghao1992