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.
- 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
We are gonna create an isolated enviroment to work, to do so you need the file environment.yml
in this repository.
-
Lets create a new enviroment
conda env create -f environment.yml
-
Lets activate the enviroment
conda activate ats_env
or
source activate ats_env
-
Clone the following repository: GDA Toolbax
-
cd ~ #Any folder different to the one where gda-public was cloned!
-
Execute
pip install /path_of_cloned_repo/gda-public
-
Let us verify our instalation:
- Open a python worksheet by typing:
ipython
orpython
- type:
import multidim, ripser
- Open a python worksheet by typing:
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.