还是不太明白bottle neck 和 transition layer具体实现方式,只get到是降维。。。
mark : https://blog.csdn.net/u014380165/article/details/75142664
论文:Densely Connected Convolutional Networks
论文链接:https://arxiv.org/pdf/1608.06993.pdf
代码的github链接:https://github.com/liuzhuang13/DenseNet
MXNet版本代码(有ImageNet预训练模型): https://github.com/miraclewkf/DenseNet