HoNNeT-web

Website for my HoNNeT deep learning side-project.

This app will load the trained keras deep learning model trained from the main HoNNeT repository and predicts which team would win, how long the match is going to last, and whether the team would lose by conceding or by the main tower being destroyed.

It contains the files required to run the web inside Heroku. I use Vue.js, Bulma.io and Flask as web technology stack.

The only file for web interface is templates/index.html. It contains heavy use of Jinja2 templating, Bulma.io CSS classes, and Vue.js scripting.

Role of each file

• honnet_brain.h5 is the Keras model representing the knowledge of the neural network (about 1 MB in size)
• honnet_brain_architecture.* are text files generated to show the architecture of the network in human-readable format only, they are not used for anything else
• app.py is the main application to run
• heroes_name.pkl is a pickle generated file containing HoN hero names as a python dictionary, it's used to convert from hero_id to hero_name and vice versa
• honnet.py is a module containing related functions to use the brain
• requirements.txt is containing all modules required to run app.py, (this file is generated by calling pip freeze > requirements.txt)
• Procfile is for Heroku, it will read and run the command inside this file to start the web
• runtime.txt is for Heroku, it will read this file to figure out which Python version to use

Requirements

• Python 3
• virtualenv (pip install virtualenv)

Running on local host (your machine)

1. Clone this repository and cd into it.
2. Create a development environment dedicated for this repository because we are going to install something here: run virtualenv venv
3. Tell python to use the environment recently created by running source activate venv (or activate venv if you are on Windows).
4. Install dependencies: pip install -r requirements.txt (gunicorn is unnecessary, don't worry if there is an error installing it, it's required for the real Linux web server only) This will install some libraries into that venv directory.
5. Enter python app.py to run the application
6. Open web browser and go to http://127.0.0.1:5000/
7. Play with the application.

Disclaimer

The data used for training the model needs to be updated regularly because heroes mechanisms are updated often. Look at the date of the honnet_brain.h5 to see when it was last updated.

Have fun!