https://github.com/xperianer/crm20_how_does_software_break

Executing Data Analysis of Mutation Testing Data from popular python projects.
https://github.com/xperianer/crm20_how_does_software_break

data-science mutation-testing testing

Last synced: 3 months ago
JSON representation

Executing Data Analysis of Mutation Testing Data from popular python projects.

• Host: GitHub
• URL: https://github.com/xperianer/crm20_how_does_software_break
• Owner: XPerianer
• License: mit
• Created: 2020-07-26T16:01:51.000Z (almost 6 years ago)
• Default Branch: master
• Last Pushed: 2024-11-24T00:15:30.000Z (over 1 year ago)
• Last Synced: 2025-04-05T21:11:48.515Z (about 1 year ago)
• Topics: data-science, mutation-testing, testing
• Language: Jupyter Notebook
• Homepage:
• Size: 13.4 MB
• Stars: 3
• Watchers: 2
• Forks: 0
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# CRM20 How does software break ![Sanity Checks](https://github.com/XPerianer/CRM20_How_does_software_break/workflows/Sanity%20Checks/badge.svg)

This repository is to document my work in the SS 2020 Seminar Code Repository Mining.

It contains analysis code, and also documented analysis in the jupyter notebook.

# Installation

The requirements are documented in the `requirements.txt` file, so if you are using `pip`, you can call

```

pip install -r requirements.txt

```

I recommend to create a virtual environment beforehand.

# Structure

The mutation testing datasets were generated with [Mutester](https://github.com/XPerianer/Mutester), a tool I developed to simplify generation of mutation testing data. Four of these datasets are also available in the `data` folder via git lfs. In the [folder specific readme](data/README.md), you can also find some details about the datasets.

A good starting point is the [Jupyter Notebook](./How_Does_Software_Break.ipynb). There you finde a walkthrough of the different aspects that were highlighted during the seminar.

The Notebook makes heavy use of code imported from the `src` folder, that is partitioned in diverse helper modules.

Especially the metric analysis code from the `src` folder is also tested (`test` folder).