Skip to content

A quantization method for neural networks based on quantiles.

Notifications You must be signed in to change notification settings

rkhkim/QuantQuant

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 
 
 

Repository files navigation

QuantQuant

A quantization method for neural networks based on quantiles in empirical distributions.

A final project for Harvard's CS 242, "Computing at Scale," taught by Prof. HT Kung.

This project was a collaboration with Ralph Estanboulieh and Ryan Kim. My main contributions involve a significant portion of the code for application of quantization to neural networks, as well a large share of writing the paper.

Check out the report, as well as the relevant code.

About

A quantization method for neural networks based on quantiles.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 100.0%