project design

bot-detector-ML

this repository is responsible for the machine learning model. We are currently using two models, a binary classifier and multi label classifier, the binary classifier is responsible for the Real_Player & Unkown_bot classification, if the Real_Player classification is less then 50% the predictions of the multi class classifier are used.

design

building this repository

Install:

• Docker
• Github desktop
  • Git windows, Git unix will also work.
• An integrated development environment (IDE).
  • We recommend VSCode, but any IDE will work.

Setup:

1. Open a terminal cmd
2. Navigate cd to where you want to save our code.
3. The command below will Create a folder bot-detector with two sub folders remote & local & download the remote repositories in the remote folder.
   • To add the repositories in github desktop, select File on the top left than click Add local repository, and navigate to the cloned repositories.

Windows

mkdir bot-detector\remote bot-detector\local && cd bot-detector\remote
git clone https://github.com/Bot-detector/Bot-Detector-Core-Files.git
git clone https://github.com/Bot-detector/bot-detector-mysql.git
git clone https://github.com/Bot-detector/bot-detector-ML.git

Linux

mkdir -p bot-detector/{remote,local}
git clone https://github.com/Bot-detector/Bot-Detector-Core-Files.git
git clone https://github.com/Bot-detector/bot-detector-mysql.git
git clone https://github.com/Bot-detector/bot-detector-ML.git

4. Now you can start the project, the command below will create the necessary docker containers, the first time might take a couple minutes. Make sure docker desktop is running!

cd Bot-Detector-Core-Files
docker-compose up -d

5. In the terminal you will now see /usr/sbin/mysqld: ready for connections. this means the database is ready.
6. Test the api's:
   • Core api: http://localhost:5000/
   • Machine learning: http://localhost:8000/

adding /docs at the end will give return the swagger documentation for the components /docs