In this repository we describe the python interface to R package TDAvec, which can be installed from CRAN
First, you need to create a clean python environment. From the project root directory run:
> python3 -m venv venv
> source venv/bin/activate
> pip install numpy==1.26.4 ripser==0.6.8
Now compile and install the package:
> python3 setup.py build_ext --inplace
> pip install .
after that you should have tdavec package installed in your environment. To check this run python and try the following commands:
> from tdavec.TDAvectorizer import createEllipse, TDAvectorizer
> ee = createEllipse()
> v = TDAvectorizer()
> v.fit([ee])
> len(v.diags) # ==> 1 since there is only one diagram
> len(v.diags[0]) # ==> 1 since there is two dimensions
> len(v.diags[0][0]) # ==> 99 since there are 99 hom0 features
or run the provided unit test:
> python tdavec/unit_test.py
Alternatively you can install the package directly from GitHub:
> pip install git+https://github.com/ALuchinsky/tdavect