🦓 CEBRA - Figures and Data

Code and data for reproducing the figures in Learnable latent embeddings for joint behavioral and neural analysis (Schneider, Lee and Mathis, 2023).

This repo only contains plotting functions which can be applied to pre-computed results. Code for reproducing experiments and applying CEBRA to custom datasets are be available in the CEBRA github repository.

All notebooks can be viewed in the CEBRA online documentation.

Quickstart

Make sure you are in a python>=3.8 environment that supports the pip install command (e.g., a virtual environment or a conda environment). Install dependencies, then render of all figures using:

make -j8 all

Figures will be placed in ipynb format into the figures/ directory.

Dependencies

pip install -r requirements.txt

Repo organization

• src: Jupyter notebooks for reproducing the paper figures, in python format
• data: Folder to data files
• figures: Rendered paper figures in ipynb format