Gradient free reinforcement learning for PyTorch

pydemic offers reinforcement learning algorithms built with Python on top of the deep learning library PyTorch.

You can extend pydemic according to your own needs. You can implement custom algorithms by extending simple abstract classes.

Algorithms

As of today, the following algorithms have been implemented:

• Particle Swarm Optimization (PSO) [1]
• Grey Wolf Optimization (GWO) [3]

Installation

1. Install PyTorch. You can find it here: PyTorch
2. pip install pydemic

Examples

You can find examples in examples/ directory

You can also run examples: python examples/pso_cartpole.py

You might want to export PYTHONPATH=/path/to/this/directory

Contribute

1. Implement new algorithms
2. Improve code design
3. Improve comments and readme
4. Tests