The official code base for "On the Level Sets and Invariance of Neural Tuning Landscapes" in NeurIPS Workshop on Symmetry and Geometry in Neural Representations

This repository contains code to conduct in silico experiments for finding elements from the level sets on hidden units, and code to analyze in vivo neuronal tuning map data from primates.

The experimental data is deposited in https://osf.io/gpzm5/ which is from the paper "Tuning Landscapes of the Ventral Stream" in Cell Reports Nov. 2022 .

Environment Setup

Install LPIPS package locally from my fork.

git clone https://github.com/Animadversio/PerceptualSimilarity
cd PerceptualSimilarity/
git checkout distmat_dev
pip install -e .

pip install git+https://github.com/Animadversio/PerceptualSimilarity.git@9e7d938b31be8daa76f9c349a0b872b4836f2edd

conda install scikit-image