Unused Field

[dikshaaswal]

Thanks for the code,
I have the issue :

Unused field: 'iou_normalizer = 0.5'
Unused field: 'cls_normalizer = 1.0'
Unused field: 'iou_loss = ciou'
Unused field: 'nms_kind = greedynms'
Unused field: 'beta1 = 0.6'
95 route 91
96 conv 128 1 x 1 / 1 38 x 38 x 256 -> 38 x 38 x 128 0.095 BF
97 upsample 2x 38 x 38 x 128 -> 76 x 76 x 128
98 route 97 36
99 conv 128 1 x 1 / 1 76 x 76 x 384 -> 76 x 76 x 128 0.568 BF
100 conv 256 3 x 3 / 1 76 x 76 x 128 -> 76 x 76 x 256 3.407 BF
101 conv 128 1 x 1 / 1 76 x 76 x 256 -> 76 x 76 x 128 0.379 BF
102 conv 256 3 x 3 / 1 76 x 76 x 128 -> 76 x 76 x 256 3.407 BF
103 conv 128 1 x 1 / 1 76 x 76 x 256 -> 76 x 76 x 128 0.379 BF
104 conv 256 3 x 3 / 1 76 x 76 x 128 -> 76 x 76 x 256 3.407 BF
105 conv 255 1 x 1 / 1 76 x 76 x 256 -> 76 x 76 x 255 0.754 BF
106 yolo
Unused field: 'iou_normalizer = 0.5'
Unused field: 'cls_normalizer = 1.0'
Unused field: 'iou_loss = ciou'
Unused field: 'nms_kind = greedynms'
Unused field: 'beta1 = 0.6'

Why is it there unused fields. Am I doing something wrong here.

and Also not showing IOU and CIOU values while training. Can you help me out in it?

[Zzh-tju]

you must have used some other repositories and just used the cfg file provided in this repo.

[dikshaaswal]

I am using this repo only. But for darknet have downloaded the darknet into this repo otherwise how run this instruction
./darknet detector train ./cfg/coco-ciou.data ./cfg/coco-ciou.cfg

[Zzh-tju]

Did you compile? make -j8, then
./darknet detector train cfg/voc-diou.data cfg/voc-diou.cfg darknet53.conv.74 -gpus 0,1

[dikshaaswal]

@Zzh-tju
Okay the steps i followed other than your repo were:
git clone https://github.com/kriyeng/darknet/
make
copied ./darknet file into the same folder DIoU darknet.

[ywx-hub]

@Zzh-tju
Okay the steps i followed other than your repo were:
git clone https://github.com/kriyeng/darknet/
make
copied ./darknet file into the same folder DIoU darknet.

Hello,I also meet this problem.
Have you already sloved this problem?

[dikshaaswal]

No, The problem is not solved yet.