Skip to content

emannix/comfe-component-features

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

ComFe: Interpretable Image Classifiers With Foundation Models

Installation

Create a new Python 3.9 environment and install the packages in the requirements.txt file.

Fitting ComFe

To fit a ComFe head to the Flowers-102 dataset using frozen DINOv2 embeddings, run

python main.py +run=base.yaml +R=comfe +model/dataset=torchvision_flowers +model/networks=dinov2_vits_14  seed=1

This should fit pretty quickly on a local GPU, and to visualize the results use

python main.py +run=base.yaml +R=comfe-view +model/dataset=torchvision_flowers +model/networks=dinov2_vits_14

Which will save the explanations for the test images in ../output/local/comfe-view_0/tensorboard/version_0/explanations.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages