search

lmbxmu / HRank

Pytorch implementation of our paper accepted by CVPR 2020 (Oral) -- HRank: Filter Pruning using High-Rank Feature Map
https://128.84.21.199/abs/2002.10179
252 stars 49 forks source link

How to determine the per-layer filter pruning rate for a given model #6

Open lzrvch opened 4 years ago

lzrvch commented 4 years ago

Hi,

Thanks for your awesome work. I was wondering - the per-layer pruning rates are hard-coded in the command line in your examples, but what is a non-manual way to determine these? Is there any code in your repo dedicated to searching for these per-layer pruning rates or did you tune them manually?

Thanks.

lmbxmu commented 4 years ago

Thanks for your interest.

Please refer to this repo(https://github.com/lmbxmu/ABCPruner).