https://github.com/luke100000/async-gym-agents

https://github.com/luke100000/async-gym-agents

Last synced: about 1 month ago
JSON representation

• Host: GitHub
• URL: https://github.com/luke100000/async-gym-agents
• Owner: Luke100000
• Created: 2024-06-27T14:22:27.000Z (5 months ago)
• Default Branch: main
• Last Pushed: 2024-08-14T08:11:25.000Z (3 months ago)
• Last Synced: 2024-08-14T09:33:25.198Z (3 months ago)
• Language: Python
• Size: 98.6 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Async Gym Agents

Wrapper environments and agent injectors to allow for drop-in async training.

```py

import gymnasium as gym

from stable_baselines3 import TD3

from async_gym_agents.agents.async_agent import get_injected_agent

from async_gym_agents.envs.multi_env import IndexableMultiEnv

# Create env with 8 parallel envs

env = IndexableMultiEnv([lambda: gym.make("Pendulum-v1") for i in range(8)])

# Create the model, injected with async capabilities

model = get_injected_agent(TD3)("MlpPolicy", env)

# Train the model

model.learn(total_timesteps=10)

# Shut down workers

model.shutdown()

```