Sensitivity Analysis Tutorial

This tutorial was prepared for the BIG-MAP AI school January 2022 by Jonas Busk (jbusk@dtu.dk).

The repository contains a series of exercises demonstrating a sensitivity analysis of a battery degradation model in a step-by-step manner. After completing these exercises, you will be able to perform a similar analysis of your own dataset and you will have some code to get started. The exercises are provided as a series of Python notebooks that let you write and run code. Some basic understanding of Python programming and machine learning is assumed.

The easiest way to get started with the exercises is to open them using Google Colab, but you can also download and run them locally if you prefer.

Run exercises using Google Colab

You can run these exercises using Google Colab without the need to install anything on your computer. This requires a Google account.

• Go to https://colab.research.google.com/.
• Select 'File' > 'Open notebook'.
• Select 'GitHub'.
• Enter the GitHub URL to this repository and hit return (a list of the exercise notebooks should show up).
• Select the exercise notebook you would like to open.

Run exercises locally

You can also download the exercises and run them locally on your own computer. This requires that you have Python 3 installed and know how to run Python notebooks.

• Download this repository.
• Create and activate a new virtual environment.
• Install requirements listed in requirements.txt.
• Run the exercises.

Citation

We are currently preparing a publication based on the methods and data presented in this tutorial which will be available in the near future.