基于yolov3的知识蒸馏+剪枝+剪植2：代码实现

简单明了说下怎么复现吧
代码地址： SpursLipu /
YOLOv3v4-ModelCompression-MultidatasetTraining-Multibackbone

1. 环境安装

    #1.conda create -n yolov3 python=3.8
    Cython      #2.pip install Cython
    numpy==1.17	#3.pip install numpy
    opencv-python     #4.pip install opencv-python
    
    torch>=1.5.1	
    #5.conda install pytorch-1.6.0-py3.7_cuda10.2.89_cudnn7.6.5_0.tar.bz2
    
    matplotlib   #6.pip install matplotlib
    pillow      #7.pip install pillow
    tensorboard    #8.pip install tensorboard
    torchvision    #9.pip install torchvision
    scipy    #10.pip install scipy
    tqdm   #11.pip install tqdm
    
    git+https://github.com/cocodataset/cocoapi.git#subdirectory=PythonAPI
    #12.pip install "git+https://gitee.com/wsyin/cocoapi.git#subdirectory=PythonAPI"
    
    #13.pip install terminaltables
   

   直接按照上面的步骤来就行，第五步的安装包链接