Skip to content

Latest commit

 

History

History
32 lines (20 loc) · 1.61 KB

README.md

File metadata and controls

32 lines (20 loc) · 1.61 KB

MachineLearning--DOTA2

This project uses a flask-based web application(Python 2.7 built on the Anaconda framework) to predict the outcome of a DOTA 2 match based on Radiant and Dire team hero selections using a Machine Learning Model (Suppor Vector Machine) trained on past matches. This repository contains the all the code and datasets that we used for this project. If you have any questions about the contents included, feel free to email me at [email protected]. You will need a valid DOTA2 API key for the API calls made in the code. Replace your API Key as indicated in the comments.

Files included :

  1. The output.csv file that holds data about previous match records
  2. The flask web application(FlaskWebApp.zip) that contains all the templates and code files

NOTE:

In order to execute this project, you need to have Python 2.7 installed, with the scientific packages numpy, scipy, sklearn. The easiest way to do this would probably be through an installation of the latest version of Anaconda, that would come with the scientific packages built in. You can install anaconda here:

https://docs.continuum.io/anaconda/install

After completing the above step, you will need to install the Dota2 Api package for python (pip install is the easiet method). Find the steps here:

https://dota2api.readthedocs.io/en/latest/installation.html

You will also need to have Flask, a web framework for Python, installed. You can find the install instructions here:

http://flask.pocoo.org/docs/0.12/installation/

Change the path to the output.csv file as required(indicated by the comments).

License:

MIT License