Skip to content

RBuck8339/DOTA2-Ranked-Predictions

Repository files navigation

How To Use It

Git clone this project, run "app.py" and open it using local host in your browser. Then, either enter your own data, or click "Generate Data" and "Predict" to see your results.

Project Goal

Create my own dataset using REST API and GraphQL calls that I can use in machine learning tasks. Focused around the popular videogame DOTA 2. I am using this as a way to reinforce my machine learning and neural network knowledge, while adding the unique aspect of building my own dataset. I am also using this as a way to learn full-stack development.

Data Processing

Using the Open DOTA REST API and Stratz API GraphQL Application, creating my own dataset to use Handling missing data, preprocessing, and feature engineering

Storing data in an SQLite database for convenience and lightweight use

Feel free to use this dataset, and my provided methods in DataPreprocessing.py to use for your own purposes.

The Model

Testing different types of machine learning models and neural networks for the predictor. For now, using a random forest model. This project will also serve as a way for me to get comfortable with different models and tasks. I created a large dataset for the purpose of being able to fit regression, binary classification, and multiclass classification tasks. However, much of it will not be implemented in the website, and will exist purely in the "prediction.ipynb" jupyter notebook.

The Website

Given that I have never designed a website before, I wanted to take my time and learn HTML, CSS, and JS. My primary goal in designing a website is to have a functional and responsive website, but I also want it to look professional and engaging, while fitting the darker color pallete of DOTA 2.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published