tetris-rnn

Joseph Hines and John Zlotek

Usage

Requires python3 and virtualenv, make preferred

To install dependencies

make venv or bash install.sh

Windows: python3 -m venv venv && source venv/Scripts/activate && pip install -r requirements.txt

To run the game

make run-main

Windows: source venv/Scripts/activate && python src/main.py

Put any args after the main.py you want to run -h is for help

At the end of the game, either press ENTER to play another, or q to quit. Data with be gathered into several JSON files, which can be aggregated with:

make data Windows: source venv/Scripts/activate && python src/datatools.py --combine default

To train the model

Given there is a file called default.json at the root of the project (generated from the previous steps):

make run-rnn Windows: source venv/Scripts/activate && python src/rnn.py Put any args after the rnn.py you want to run -h is for help

To see the model play

Given there is a model.h5 file (generated from training):

make model-play Windows: source venv/Scripts/activate && python src/main.py -m model.h5

To clean

make clean