Eric-mingjie / rethinking-network-pruning

Rethinking the Value of Network Pruning (Pytorch) (ICLR 2019)
MIT License
1.51k stars 293 forks source link

Is it reasonable to get a threshold for all bn layers? #35

Open zihaozhang9 opened 4 years ago

zihaozhang9 commented 4 years ago

When calculating the threshold, the weight ordering of all bn layers is used. Is this reasonable?

Is there such a phenomenon: ① The first value of the network is closer to the image pixel value, and the last layer is closer to the category probability. bn's weight is not necessarily the same. ② There is a shortcut in the middle of the network. After the two convolution pixel values are superimposed, the weight parameter becomes larger. May affect bn's weight.

在计算阈值时,将使用所有bn层的权重排序。 这合理吗? 是否存在这样的现象: ①网络最前面的数值,更靠近图像像素值,最后一层更靠近类别概率。bn的weight不一定分布相同。 ②在网络中间有shortcut,两个卷积像素值叠加后,weight参数变大。可能会影响bn的weight。

issues 期待您的回复。十分感谢。

Eric-mingjie commented 4 years ago

我们使用的就是原始的network slimming的方法,code实现了slimming的原始论文 https://arxiv.org/abs/1708.06519

slimming中全局排序bn在实验上被证明是简捷并有效的方法,对于你提的问题, 我们没有比较过每个bn层的scalng fcator,不过很有意义和值得研究,更精细的pruning可能会有更好的结果。