Adaptive Template Systems

by Luis Polanco Contreras and Jose A. Perea

This repository implements the tools developed in the paper Approximating Continuous Functions on Persistence Diagrams Using Template Functions. The main goal of this package is to provide a tool that extract features fro persistent diagrams in an adaptive manner.

Requirments:

• Python/3.3. or higher
• GNU/4.7.1 or higher to run RIPSER
• GDA Toolbox package: https://github.com/geomdata/gda-public/
• Ripser
• Keras
• Theano
• Numba
• hdbscan

Before running:

We are gonna create an isolated enviroment to work, to do so you need the file environment.yml in this repository.

1. Lets create a new enviroment

   conda env create -f environment.yml
   

2. Lets activate the enviroment

   conda activate ats_env
   

   or

   source activate ats_env
   

3. Clone the following repository: GDA Toolbax

4. cd ~ #Any folder different to the one where gda-public was cloned!

5. Execute

   pip install /path_of_cloned_repo/gda-public
   

6. Let us verify our instalation:

   1. Open a python worksheet by typing: ipython or python
   2. type: import multidim, ripser

Running and Examples

In the folder Examples are 3 different examples available: 6-class manifold classification problem, shape classification from the SHREC 2014 data set and protein classification problem from the SCOPe data base.