Valostats, A Valorant Match Predictor

UQ Computing Society 2024 People's Choice Award Winner

Maintainers

Zain Al-Saffi - Team Lead, Design, Frontend, Data Analysis and ML

Varun Singh - Plots and graphs

Aman Gupta - Partial frontend

Jasnoor Matharu - Boosting algorithms

Lehan Ling - ML, Frontend and Data processing

Abhishek Bhattacharjee - Pickle

Background

This project started as a joke, it was inspired by a friend's affinity to gamble Twitch channel points and constantly losing, hence we made a predictor to help with his predictions using machine learning

Pre requisites

For starters get NPM (Node.js) installed on your local machine, then cd into:

cd vlresports-1.0.4

npm install

npm start

Setup anaconda / miniconda on python 3.8.9 for environment manegement, then install the following:

conda install scikit-learn

conda install numpy

conda install matplotlib

conda install seaborn

conda install streamlit

conda install kaggle

conda install pandas

Kaggle dataset setup

To use the continuous dataset updater through kaggle API, make sure to do the below: Go to Kaggle.com and log in. Navigate to your account settings by clicking on your profile picture in the top-right corner and selecting Account. Scroll down to the API section and click Create New API Token. This will download a kaggle.json file to your computer.

Then run the following:

mkdir -p ~/.config/kaggle
mv /path/to/kaggle.json ~/.config/kaggle/kaggle.json
chmod 600 ~/.config/kaggle/kaggle.json  # Secure the file

This should setup your Kaggle API, Keep in mind the dataset is updated once a month and likely is the model is displaying 50/50 odds it means there is not enough data about either team, meaning they are likely new teams.

Credits go to Ryanluong1 on kaggle for the valorant dataset, and the Orloxx23 for the vlresports API.