-
Notifications
You must be signed in to change notification settings - Fork 2
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add 24 compressor #167
base: main
Are you sure you want to change the base?
Add 24 compressor #167
Conversation
2015e71
to
dea129e
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Very clean.
lgtm after tests
!!
2f69d16
to
fc4b23c
Compare
dea129e
to
68ca6c3
Compare
fc4b23c
to
dd16499
Compare
Move tests Remove unused import
dd16499
to
7155e61
Compare
68ca6c3
to
6636872
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Overall code looks simple. I'd like to reformulate the scope, though. Specifically, I'm not following why we are restricting to just 2:4 right now when we could easily expand this to handle all sparsity cases and detect whether it is 2:4 format, some type of structured pruning, and if not any then set as unstructured. cc @dsikka
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
testing?
This PR adds Sparse24Compressor, for 2:4 sparse models
The code is based off #129
Depends on:
This implements Part 3 of the Design doc: https://www.notion.so/Design-Document-24-Compressor-25ac643aee604c298f2bb12a6c220861?pvs=4
Class Hierarchy:
File Structure: