Skip to main content

PyTorch implementations of Deep reinforcement learning algorithms.

Project description

Deepair is a Deep Reinforcement Learning library

PyPI version Documentation Status

Deepair implementations of reinforcement learning algorithms. It focus on DRL algorithms and implementing the latest advancements in DRL. Highly customizable support for training processes. Suitable for the research and application of the latest technologies in reinforcement learning.

Features

Documentation

Documentation is available: https://deepair.readthedocs.io/

Installation

pip install deepair

or

pip install git+https://github.com/sonnhfit/deepair.git

Example

import gym
from deepair.dqn import Rainbow

env = gym.make('LunarLander-v2')

rain = Rainbow(env=env, memory_size=10000, batch_size=32, target_update=256)

rain.train(timesteps=200000)

# test
state = env.reset()
done = False
score = 0

while not done:
    action = rain.select_action(state, deterministic=True)
    next_state, reward, done = env.step(action)

    state = next_state
    score += reward

print("score: ", score)

rainbow lunalander env

Implemented Algorithms

Tutorial

How To Contribute

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

deepair-0.2.0.tar.gz (11.5 kB view hashes)

Uploaded Source

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page