/ripsnet

RipsNet: a general architecture for fast and robust estimation of the persistent homology of point clouds

Primary LanguageJupyter Notebook

RipsNet: a general architecture for fast and robust estimation of the persistent homology of point clouds

This repository contains the code asscoiated to this arXiv article.

  • tutorial.ipynb is a jupyter-notebook to run a small version of our experiments on a synthetic data set. It has been tested with Tensorflow version 2.5.0.
  • expes_synth_ucr.zip contains code and instructions for reproducing the full experiments on synthetic and UCR time series data sets.
  • expes_3Dshape.zip contains code and instructions for reproducing the full experiments on 3D shape data sets.