zzk@2020
zzk@2020
填充到5x5里面，计算比较麻烦
5x5就不快了，而且浮点数乘法估计误差有点大 从 Windows 版邮件发送 发件人: ***@***.***> 发送时间: 2023年6月3日 23:31 收件人: ***@***.***> 抄送: ***@***.******@***.***>; ***@***.***> 主题: Re: [DingXiaoH/DiverseBranchBlock] 3X3卷积串联3X3卷积 (Issue #30) 可以具体说说吗，怎么实现？ ― Reply to this email directly, view it on...
Not helpful, It seems a little lower Rank1 and mAP compared to Resnet. same parameters and epoch in my REID task. I gusses face recognnition will be same, if you...
Yes, I also confused about it.
same question . no use for variable ``` float tot_diou_loss = 0; float tot_ciou_loss = 0; ``` in function forward_yolo_layer and there is no diou ciou loss added in l.cost...
> @WilburZjh It is seemed that there is no more people that concern the question. I think it doesn't implement the CIoU and DIoU loss.Altough it is defined in the...
It is important for me cause I use own iou loss, I want to know the reason do that
l.delta is used for cost，not l.cost