Skip to content

DeepPINK: reproducible feature selection in deep neural networks

License

Notifications You must be signed in to change notification settings

younglululu/DeepPINK

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

11 Commits
 
 
 
 
 
 

Repository files navigation

DeepPINK

Yang Lu, Yingying Fan, Jinchi Lv, William Stafford Noble. ["DeepPINK: reproducible feature selection in deep neural networks"] (https://papers.nips.cc/paper/8085-deeppink-reproducible-feature-selection-in-deep-neural-networks) Advances in Neural Information Processing Systems 31 (NeurIPS), 2018.

This repository contains a Python implementation of DeepPINK. The input is an n x 2p matrix and n x 1 labels. The output contains n x 1 feature importance values, n x 1 feature knockoff statistics, and the set of features selected subjected to the specified FDR threshold.

To use DeepPINK, you must first generate knockoffs. Note that there are multiple ways to generate such knockoffs, such as using deep neural networks.

All datasets used in the DeepPINK paper are available at (https://noble.gs.washington.edu/proj/DeepPINK).

About

DeepPINK: reproducible feature selection in deep neural networks

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages