/sphericov

Companion Jupyter notebook of the paper "Functional Estimation of Anisotropic Covariance and Autocovariance Operators on the Sphere"

Primary LanguageJupyter NotebookMIT LicenseMIT

Instructions

This repository hosts the companion Jupyter notebook for the paper "Functional Estimation of Anisotropic Covariance and Autocovariance Operators on the Sphere", from Alessia Caponera, Julien Fageot, Matthieu Simeoni and Victor Panaretos.

The code was written by Matthieu Simeoni.

Scope

This Jupyter notebook allows to reproduce the simulations, numerical experiments and plots of Section 6 of the paper.

Installation

The notebook requires Python 3.9 or greater. It is developed and tested on x86_64 systems running MacOS and Linux.

Dependencies

The notebook extra dependencies are listed in the file requirements.txt. It is recommended to install those extra dependencies in an Anaconda environment Miniconda or Anaconda.

>> conda create -n sphericov python=3.9
>> conda activate sphericov
>> pip install -r requirements.txt

Run the code

Once the dependencies installed, you can run the companion notebook by executing the following commands:

>> git clone https://github.com/matthieumeo/sphericov
>> cd <repository_dir>/
>> conda activate sphericov
>> jupyter lab companion_nb.ipynb