bipedal-walker-deep-ga

The project solves Bipedal-Walker-v3 using deep neural networks trained by genetic algorithm. We used simplified version of Deep Neuroevolution: Genetic Algorithms Are a Competitive Alternative for Training Deep Neural Networks for Reinforcement Learning. Our best model reached average return of ~310 over 100 episodes.

How to install project

1. Clone the repository and change current directory

git clone https://github.com/standa42/bipedal-walker-deep-ga
cd "bipedal-walker-deep-ga"

2. Create virtual environment

• you can also reuse an existing one but this tutorial acts as you would create a new one

/usr/bin/python3 -m venv "venv"

3. Install pip requirements

venv/bin/python3 -m pip install -r "requirements.txt"

How to train a model

Training is simple using following script:

venv/bin/python3 train.py

Default parameters are already set for the best reached model. Logs are stored into logs/train{TIMESTAMP}_{UUID4_CODE} directory.

How to evaluate a model

Evaluation of model is done using following script:

venv/bin/python3 evaluate.py

Evaluation can be visualized using render_each parameter.