JasperKirk

Prune的ignore_idx指的是哪几层

2

您好，感谢您开源您的工作，请问您在剪枝部分ignore_idx=[230, 260, 290],这个[230, 260, 290]分别指的是哪些层，或者您是否有一个权重可以共享

yolov4-tiny和yolov4-tiny-3l的结果一样

2

请问你在yolov4-tiny-3l上怎么训练的，其中loss的balance怎么设置，anchor呢？ 我用YOLOv4-tiny和yolov4-tiny-3l训练出来的结果是一样的。应该yolov4-tiny-3l要比yolov4-tiny高才对，你能提供一份yolov4-tiny-3l的yaml吗？

In the paper,for MS COCO dataset, the learning rate is change at 90, 120 and 140 epoch, respectively. But in the yaml, the learning rate is change at 90, 110...

Compute MBR Diagonal

5

This is a really great repo and I'm enjoying reading it.But I have a question about Compute MBR Diagonal. Why Am,Bm,Cm,Dm are all use **torch.max()**