https://github.com/ikostrikov/pytorch-ddpg-naf

Implementation of algorithms for continuous control (DDPG and NAF).
https://github.com/ikostrikov/pytorch-ddpg-naf

ddpg deep-deterministic-policy-gradient deep-learning pytorch reinforcement-learning

Last synced: 2 days ago
JSON representation

Implementation of algorithms for continuous control (DDPG and NAF).

• Host: GitHub
• URL: https://github.com/ikostrikov/pytorch-ddpg-naf
• Owner: ikostrikov
• License: mit
• Created: 2017-01-28T21:45:50.000Z (almost 8 years ago)
• Default Branch: master
• Last Pushed: 2021-02-16T03:42:25.000Z (over 3 years ago)
• Last Synced: 2024-05-23T06:50:59.316Z (6 months ago)
• Topics: ddpg, deep-deterministic-policy-gradient, deep-learning, pytorch, reinforcement-learning
• Language: Python
• Homepage:
• Size: 36.1 KB
• Stars: 304
• Watchers: 10
• Forks: 72
• Open Issues: 7
• Metadata Files:
  • Readme: README.md
  • License: LICENSE.md

Awesome Lists containing this project

README

        
### DEPRECATION

This repository is deprecated and is no longer maintaned. Please see a more recent implementation of RL for continuous control at [jax-sac](https://github.com/ikostrikov/jax-sac/).

### Description

Reimplementation of [Continuous Deep Q-Learning with Model-based Acceleration](https://arxiv.org/pdf/1603.00748v1.pdf) and [Continuous control with deep reinforcement learning](https://arxiv.org/pdf/1509.02971.pdf).

Contributions are welcome. If you know how to make it more stable, don't hesitate to send a pull request.

### Run

Use the default hyperparameters.

#### For NAF:

```

python main.py --algo NAF --env-name HalfCheetah-v2

```

#### For DDPG

```

python main.py --algo DDPG --env-name HalfCheetah-v2

```