Skip to content

Replace FC2, LeNet-5, VGG, Resnet, Densenet's full-connected layers with MPO

License

Notifications You must be signed in to change notification settings

zfgao66/deeplearning-mpo

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

48 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

MPO-code

This repository contains the code for using MPO structure to replace FC2, LeNet-5, VGG, Resnet, Densenet's full-connected layers introduced in the following paper
"Compressing deep neural networks by matrix product operators"
Ze-Feng Gao, Song Cheng, Rong-Qiang He, Hui-Hai Zhao, Z.Y.Xie, Zhong-Yi Lu, and Tao Xiang
The code is built on Tensorflow

Requirement

Python >=3.4.3
TensorFlow >=1.3.0

How to run

$ pip install --user -r requirements.txt $ python train.py --args=args-you-want-change

Note

About

Replace FC2, LeNet-5, VGG, Resnet, Densenet's full-connected layers with MPO

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages