Individual in-situ GPS-derived acceleration-speed profiling

Introduction

This project accompanies the research paper titled Individual in-situ GPS-derived acceleration-speed profiling: towards automatization and refinement published in the journal Sports Medecine-Open by N. Miguens.

Overview

Recently a proof-of-concept was proposed to derive the soccer players’ individual in-situ acceleration-speed (AS) profile from global positioning system (GPS) data collected over several sessions. The present study aimed to validate an automatized method of individual GPS-derived in-situ AS profiling in professional rugby union setting.

Project Contents

1. /data: This directory contains a dataset for testing the code and providing an example of the expected file.

2. /code: This directory contains the source code associated with the research. The code is divided into separate modules for easier understanding: outliers to identify outliers and regression to calculate AS profiles.

3. /results: This directory contains the results produced by the code: a csv file containing the characteristic values of the AS profiles by player and possibly by training exercise; point cloud images for each player and training exercise.

Requirements

Python 3 is required to run the code for this project.

Installation Instructions

1. Clone this repository to your local machine.
2. Install the required dependencies using the command: pip install -r requirements.txt
3. Run the main code using the command: python main.py -s -k

A dockerfile is also available.

Usage

Arguments

• The -f or --filename argument is used to select a csv file in data folder.

  python main.py --filename Session_example

• The -s argument is used to convert speed in km/h to m/s.

  python main.py -s 

• The -k argument is used to keep acceleration values from csv file.

  python main.py -k 

• The --dv argument is used to define the small speed range in max intensity identification.

  python main.py --dv 0.3

• The --n_max argument is used to define the numbers of points by small speed range in max intensity identification.

  python main.py --n_max 2

Input file - Requirements

Player	Speed	Timestamp
Adrien	4.27	12:45:59.0
Adrien	4.38	12:45:59.1
Adrien	4.57	12:45:59.2

'Acceleration' and 'Date' can also be added.

Results

CSV results

Player	a0 : Regression quantile	std_a0	s0 : Regression quantile	std_s0
Adrien	5.57	0.31	8.20	1.00
Franck	7.01	0.35	9.32	1.20
Loic	6.42	0.19	7.52	0.54

Image results

Contributing

If you wish to contribute to this project, follow these steps:

1. Fork the project
2. Create a new branch (git checkout -b feature/add-feature)
3. Commit the changes (git commit -am 'Add feature')
4. Push the branch (git push origin feature/add-feature)
5. Open a pull request

License

This project is licensed under MIT License. See the LICENSE.md file for more details.

Contacts

For any questions or concerns regarding this project, please contact N. Miguens or open an issue.