final_rl

强化学习选修课期末大作业代码

https://github.com/zheng-guangyuan/final_rl

Science Score: 44.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • CITATION.cff file
    Found CITATION.cff file
  • codemeta.json file
    Found codemeta.json file
  • .zenodo.json file
    Found .zenodo.json file
  • DOI references
  • Academic publication links
  • Academic email domains
  • Institutional organization owner
  • JOSS paper metadata
  • Scientific vocabulary similarity
    Low similarity (11.3%) to scientific vocabulary
Last synced: 10 months ago · JSON representation ·

Repository

强化学习选修课期末大作业代码

Basic Info
  • Host: GitHub
  • Owner: Zheng-guangyuan
  • Language: Python
  • Default Branch: main
  • Size: 304 KB
Statistics
  • Stars: 0
  • Watchers: 1
  • Forks: 0
  • Open Issues: 0
  • Releases: 0
Created over 1 year ago · Last pushed over 1 year ago
Metadata Files
Readme Citation

README.md

Final_RL

This repository contains the implementation and experiments for multi-agent reinforcement learning algorithms, including CPPO, IPPO, and MAPPO, tested on the VMAS balance scenario.

File Structure

  • docs/: Documentation files for the project.
  • vmas/: VMAS environment and simulation files.
  • CITATION.cff: Citation file for referencing this project.
  • README.md: This README file providing an overview of the project.
  • codecov.yml: Configuration file for code coverage tools.
  • requirements.txt: List of Python dependencies required for this project.
  • setup.cfg: Configuration file for Python packaging.
  • setup.py: Script for installing the project as a Python package.
  • test_CPPO.py: Implementation and tests for the CPPO algorithm.
  • test_IPPO.py: Implementation and tests for the IPPO algorithm.
  • test_MAPPO.py: Implementation and tests for the MAPPO algorithm.
  • test_PPO.py: Baseline implementation and tests for PPO.

Prerequisites

  • Python 3.8 or higher
  • Ray 2.1.0
  • VMAS environment (included in the repository)
  • Additional dependencies listed in requirements.txt

To install the required dependencies:

bash pip install -r requirements.txt

Usage

Running the Algorithms

Each of the test scripts (test_CPPO.py, test_IPPO.py, test_MAPPO.py, test_PPO.py) can be executed to train and evaluate the corresponding algorithm in the VMAS balance scenario.

For example, to run the CPPO algorithm:

bash python test_CPPO.py

Configurations

  • Modify the environment and training parameters directly in the respective test scripts.
  • Training logs and evaluation results will be displayed in the console or logged via tools like wandb (if configured).

Environment

The VMAS environment is a vectorized multi-agent simulation framework for reinforcement learning. This repository uses the balance scenario with customizable configurations.

Owner

  • Name: 郑广源
  • Login: Zheng-guangyuan
  • Kind: user
  • Location: Sun Yat-sen University/major in Computer Science
  • Company: Sun Yat-sen University

Citation (CITATION.cff)

cff-version: 1.2.0
message: If you use this software, please cite it as below.
title: VMAS
authors:
  - family-names: Bettini
    given-names: Matteo
preferred-citation:
  type: conference-paper
  title: "VMAS: A Vectorized Multi-Agent Simulator for Collective Robot Learning"
  authors:
    - family-names: Bettini
      given-names: Matteo
    - family-names: Kortvelesy
      given-names: Ryan
    - family-names: Blumenkamp
      given-names: Jan
    - family-names: Prorok
      given-names: Amanda
  collection-title: Proceedings of the 16th International Symposium on Distributed Autonomous Robotic Systems
  publisher: Springer
  year: 2023

GitHub Events

Total
  • Push event: 2
  • Create event: 2
Last Year
  • Push event: 2
  • Create event: 2

Dependencies

docs/requirements.txt pypi
  • gym *
  • numpy *
  • pyglet <=1.5.27
  • six *
  • torch *
requirements.txt pypi
  • gym *
  • numpy *
  • pyglet <=1.5.27
  • six *
  • torch *
setup.py pypi
  • numpy *