/Rethinking_network_pruning

(KSC 2020) 학습률 감쇠 기법 관점에서의 네트워크 프루닝 기법에 대한 고찰

Primary LanguageJupyter NotebookMIT LicenseMIT

Rethinking network pruning from the perspective of learning rate decay methods

Thinet(option : greedy) paper : ThiNet: A Filter Level Pruning Method for Deep Neural Network Compression(ICCV,2017)

LASSO(option : lasso) paper: Channel Pruning for Accelerating Very Deep Neural Networks(ICCV,2017)

Prerequistes

  1. pytorch, python : pytorch 1.6 ↑, python 3.7 ↑
  2. package : numpy, os, torchsummaryX, tqdm

have to download VGG16_BN weight and then move model weights to directory './experiments/vgg16_exp_cifar100_0/checkpoints/'

https://drive.google.com/drive/folders/1D0zxgCMg3nDUGGxCt8n2PDtcU6FHVysP?usp=sharing

Experiments