search

lucaslie / torchprune

A research library for pytorch-based neural network pruning, compression, and more.
https://people.csail.mit.edu/lucasl/
MIT License
161 stars 24 forks source link

Reproduce the informative features analysis? #5

Closed ohaijen closed 2 years ago

ohaijen commented 2 years ago

Dear authors,

Thanks for the nice paper and code repo. Is there a way to use this code to reproduce the informative features analysis (Lost in pruning / what is preserved)?

Thanks in advance!

lucaslie commented 2 years ago

Hi, Thanks for your interest in our paper and code repo. :)

Unfortunately, I never got around to clean up the code for these experiments. But I added what I have to the repo here:

https://github.com/lucaslie/torchprune/tree/main/paper/lost/script/sis_experimental

I don't know if it runs bug-free since I haven't tested it for a while but hopefully, you should be able to modify it and get it to run one way or the other.

ohaijen commented 2 years ago

Hi Lucas, no worries at all :) Thanks a lot for sharing what you have!