/QuantizedGromovWasserstein

Scalable framework for comparing metric measure spaces with up to 1M points.

Primary LanguageJupyter NotebookMIT LicenseMIT

color-transfer

lobby

Quantized Gromov-Wasserstein

This repository contains the implementations in the paper Quantized Gromov-Wasserstein by Samir Chowdhury, David Miller, and Tom Needham.

Installation

We recommend creating a virtual environment to test this code:

conda create -n qgw python=3.7
conda activate qgw
pip install -r requirements.txt

This installs Jupyter Notebook and Jupyter Lab, either of which can be used to test out the included notebooks.