https://github.com/177arc/fpl-advisor

Jupyter notebook for interactively analysing Fantasy Premier League (FPL) players and optimising your team's performance
https://github.com/177arc/fpl-advisor

fantasy-premier-league football fpl jupyter-notebook python

Last synced: about 2 months ago
JSON representation

Jupyter notebook for interactively analysing Fantasy Premier League (FPL) players and optimising your team's performance

• Host: GitHub
• URL: https://github.com/177arc/fpl-advisor
• Owner: 177arc
• License: mit
• Created: 2019-11-03T18:39:53.000Z (about 5 years ago)
• Default Branch: master
• Last Pushed: 2021-07-24T09:44:33.000Z (over 3 years ago)
• Last Synced: 2023-02-28T19:41:43.298Z (almost 2 years ago)
• Topics: fantasy-premier-league, football, fpl, jupyter-notebook, python
• Language: Jupyter Notebook
• Size: 2.18 MB
• Stars: 13
• Watchers: 4
• Forks: 6
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
[![Binder](https://mybinder.org/badge_logo.svg)](https://mybinder.org/v2/gh/177arc/fpl-advisor/master?filepath=advisor.ipynb)

[![Python 3.8](https://img.shields.io/badge/python-3.8-blue.svg)](https://www.python.org/downloads/release/python-380/)

# Fantasy Premier League (FPL) Advisor

**WARNING: Playing FPL can be highly adictive.**

## Purpose

The purpose of the Advisor Jupyter notebook is to help with the selection of team members for the [Fantasy Premier League](https://fantasy.premierleague.com/) (FPL) by attempting to forecast how many points players will earn.  

It provides visual analysis and uses linear optimisation to recommend a team with the maximum expected points to improve the performance of your current team.

The underlying data comes the [fpl-data project](https://github.com/177arc/fpl-data) which in turn gets it from the FPL API. The data is updated on an hourly basis.

If you are not familiar with the Fantasy Premier League, you can watch this introduction:



## Usage

To use the FPL Advisor Jupyter notebook interactively, simply open the [advisor.ipynb](advisor.ipynb) notebook on [![Binder](https://mybinder.org/badge_logo.svg)](https://mybinder.org/v2/gh/177arc/fpl-advisor/master?filepath=advisor.ipynb) (it may take a bit of time to deploy the notebook).

Alternatively, simply clone the repository and open advisor.ipynb locally.

Here is a screenshot of the interactive chart for analysing players:

[![FPL Advisor Visualisation](fpl_advisor.png)](https://mybinder.org/v2/gh/177arc/fpl-advisor/master?filepath=advisor.ipynb)

And you can use the optimiser for selecting the best players for a wildcard/free hit or recommending transfers for your team:  

[![FPL Advisor Optimiser](optimiser.png)](https://mybinder.org/v2/gh/177arc/fpl-advisor/master?filepath=advisor.ipynb)

To explore the FPL data using a neural network, [train_nn_model.ipynb](train_nn_model.ipynb) notebook on [![Binder](https://mybinder.org/badge_logo.svg)](https://mybinder.org/v2/gh/177arc/fpl-advisor/master?filepath=train_nn_model.ipynb).

## Contributing

1. Fork the repository on GitHub.

2. Run the tests with `python -m unittest discover -s tests/unit` to confirm they all pass on your system.

   If the tests fail, then try and find out why this is happening. If you aren't

   able to do this yourself, then don't hesitate to either create an issue on

   GitHub,  send an email to [[email protected]](mailto:[email protected]>).

3. Either create your feature and then write tests for it, or do this the other

   way around.

4. Run all tests again with with `python -m unittest discover -s tests/unit` to confirm that everything

   still passes, including your newly added test(s).

5. Create a pull request for the main repository's ``master`` branch.