https://github.com/aserhiychuk/pyreinforce

Deep Reinforcement Learning library for Python
https://github.com/aserhiychuk/pyreinforce

ai atari machine-learning neural-networks openai-gym python reinforcement-learning tensorflow

Last synced: 7 months ago
JSON representation

Deep Reinforcement Learning library for Python

• Host: GitHub
• URL: https://github.com/aserhiychuk/pyreinforce
• Owner: aserhiychuk
• License: mit
• Created: 2019-05-14T23:25:23.000Z (almost 7 years ago)
• Default Branch: master
• Last Pushed: 2022-06-09T23:25:21.000Z (almost 4 years ago)
• Last Synced: 2025-07-24T04:01:40.835Z (8 months ago)
• Topics: ai, atari, machine-learning, neural-networks, openai-gym, python, reinforcement-learning, tensorflow
• Language: Python
• Homepage:
• Size: 3.76 MB
• Stars: 10
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# PyReinforce

**PyReinforce** is a collection of algorithms that can be used to solve a variety of different reinforcement learning problems.

| Breakout | Pong | Space Invaders |

|:-:|:-:|:-:|

| [![Breakout](assets/Breakout_160x210.gif?raw=true)](examples/Atari.ipynb) | [![Pong](assets/Pong_160x210.gif?raw=true)](examples/Atari.ipynb) | [![Space Invaders](assets/SpaceInvaders_160x210.gif?raw=true)](examples/Atari.ipynb) |

| Cart Pole | Lunar Lander | Pendulum |

|:-:|:-:|:-:|

| [![Cart Pole](assets/CartPole_160x160.gif?raw=true)](examples/MonteCarlo.ipynb) | [![Lunar Lander](assets/LunarLander_160x160.gif?raw=true)](examples/TemporalDifference.ipynb) | [![Pendulum](assets/Pendulum_160x160.gif?raw=true)](examples/DDPG.ipynb) |

## Basics

This is how [OpenAI Gym](https://gym.openai.com/) describes the reinforcement learning process:

>There are two basic concepts in reinforcement learning: the environment (namely, the outside world) and the agent (namely, >the algorithm you are writing). The agent sends actions to the environment, and the environment replies with observations and >rewards (that is, a score).

PyReinforce is fully compatible with [OpenAI Gym](https://gym.openai.com/). In fact, it will work with **any** environment as long as it implements these methods:

* `reset()`

* `step(action)`

Your part is to implement a `Brain`: a neural network that agents use to decide which actions to pick for given states. Agents get better over time by performing training steps on their brains. See [examples](examples) for more details.

## Installation

Install PyReinforce from PyPI:

```bash

pip install PyReinforce

```

or from source:

```bash

git clone https://github.com/aserhiychuk/pyreinforce.git

cd pyreinforce

pip install -e .

```

## Examples

* [Monte Carlo](examples/MonteCarlo.ipynb)

* [Temporal Difference](examples/TemporalDifference.ipynb)

* [Policy Gradient](examples/PolicyGradient.ipynb)

* [Deep Deterministic Policy Gradient](examples/DDPG.ipynb)

* [Deep Q-Learning](examples/Atari.ipynb)

In order to run the examples you need to install dependencies:

```bash

pip install -r examples/requirements.txt

```