[blog8]ubuntu18.04配置yolov3

参考https://blog.csdn.net/weixin_42747301/article/details/96636614#_76

前提：如果想用GPU，装好N卡驱动，cuda cudnn opencv再来

下载darknet，下载yolove.weights

git clone https://github.com/pjreddie/darknet
cd darknet
make
wget https://pjreddie.com/media/files/yolov3.weights

在cfg里面把yolov3.vfg改为test的batch和sub=1，train注释掉，测试

如果测试报错out of memory，就把cfg里面的图像大小改一改

./darknet detect cfg/yolov3.cfg yolov3.weights data/dog.jpg

出现结果：

layer     filters    size              input                output
    0 conv     32  3 x 3 / 1   416 x 416 x   3   ->   416 x 416 x  32  0.299 BFLOPs
    1 conv  

……

Loading weights from yolov3.weights...Done!
data/dog.jpg: Predicted in 0.142452 seconds.
dog: 99%
truck: 83%
ca