search

Eric-mingjie / rethinking-network-pruning

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

Pruning strategy #37

Open JHC2019 opened 4 years ago

JHC2019 commented 4 years ago

Dear author: If I want to prune VGG-16 model using L1-norm based channel pruning method in ImageNet.

Should I prune the shallow or deep convolutional layer first? I read some related papers, but it don't seem to mention this one in the article. Or is it simply based on sensitivity of layer and experience?

Best Regard.

Eric-mingjie commented 4 years ago

Our code for l1-norm implements the proposed method in this paper (https://arxiv.org/abs/1608.08710).