superran1
superran1
I meet same problem with you ,keras version is 2.4.3.
这个是部署到C++上面的,目前只支持静态模型,直接将模型转换为静态结构应该怎么做呢
 我更换了unet情况是一样的,模型其余部分都能正常显示shape,只有当进行bilinear_interp时会出现-1的问题,这个能帮忙看一下吗 链接:https://pan.baidu.com/s/1dkVF6yM9PA1op_cVCL4s6g 提取码:zf73 --来自百度网盘超级会员V4的分享
I have the same question. YOLOv8 has better precision and recall than YOLOv5 in the training dataset and test dataset, but YOLOv8 performs poorly than yolov5 in the test data.