Skip to content

The fast transformation algorithm for transposed convolutional layers

Notifications You must be signed in to change notification settings

WendongMao/DeConv-FTA

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 

Repository files navigation

A Fast Transformation Algorithm for Deconvolution

It provides detailed transformation parameters of the paper, which proposes a fast transformation algorithm (FTA) to accelerate deconvolution computation for deep nerual networks, eg. Generative Adversarial Network, Fully Convolutional Networks.

Wendong Mao, Peixiang Yang and Zhongfeng Wang, "FTA-GAN: A Computation-Efficient Accelerator for GANs With Fast Transformation Algorithm" in IEEE Transactions on Neural Networks and Learning Systems: Regular Paper, 2021.

[Paper] | Detailed Parameters of FTA for various deconvolutions(/transposed convolutions) will be available soon.

Please consider citing our paper if you find it useful for your work.

Contact

Wendong Mao

wdmao@smail.nju.edu.cn

About

The fast transformation algorithm for transposed convolutional layers

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages