Skip to content

Testing procedures for Super-Resolution, i.e. testing spikes from low frequency measurements.

Notifications You must be signed in to change notification settings

ydecastro/super-resolution-testing

Repository files navigation

Testing Procedure for Super Resolution

This repository contains an illustration of the numerical experiments performed in the paper entitled "Testing Gaussian Process with Applications to Super-Resolution" by Jean-Marc Azaïs, Yohann De Castro and Stéphane Mourareau.

Download file on the ArXiv (arXiv:1706.00679v3)

Chapter 6 (Pages 25-29) of version 3.

Python source code

The Python code can be downloaded at

Github repository super-resolution-testing ZIP file

The file aux.py contains all the auxiliary functions. The file testing_superresolution.py is illustrated in the notebook 'testing_super_resolution.ipynb'.

It requires at most: python>=3.6.4, numpy>=1.14.0, scipy>=1.0.0, seaborn>=0.8.1

Getting started

Once you have downloaded the two files testing_superresolution.py and aux.py, you can execute testing_superresolution.py to make the numerical experiments of the paper. The notebook testing_super_resolution.ipynb is an illustration of this code. It should take you between 15 to 30min to run all the experiments below (for less than iterations<=500 montecarlo simulations).

Have a look at the Jupiter notebook on blob

Thank you for your time!