ShAM

Shapley Anything Model

Authors: Mike (Siyu) Wang, Kuai Yu, Ishan Kalburge

Overview

Documentation: contains steps for inpainting for generative inpainting pytorch and using Rockfish.

Quadrant-based Feature Reduction: contains files for running Quadrant-based approach for shapley value calculation.

The Shapley Anything Model: contains files for running SAM-based shapley value calculation

VAE_GAN_experiment: contains files for VAE and GAN for studying data distribution.

superpixel_experiment: contains files for using skimage superpixel

• Details for using quadrant-based Feature Reduction and The Shapley Anything Model is inside each folder
• We give credit to the authors of generative-inpainting model as well as h-shap prediction model based on BBBC041. The detailed link to those directory could be found in Inpainting Model and h-shap