• Stars
    star
    416
  • Rank 104,068 (Top 3 %)
  • Language
    Python
  • License
    Apache License 2.0
  • Created about 2 years ago
  • Updated almost 2 years ago

Reviews

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Repository Details

羊了个羊 + 深度强化学习(Deep Reinforcement Learning + 3 Tiles Game)

DI-sheep:深度强化学习 + 羊了个羊

当 AI 技术的明珠——深度强化学习,遇到号称“通关率只有0.01%”的游戏“羊了个羊”,会碰撞出哪些奇思妙想呢?

P.S. 路过记得点个 star stars - di-sheep ,持续更新ing。

P.S.S. 想了解更多深度强化学习相关知识?快来 DI-engine 训练自己的智能体。

News

使用指南

算法原理解析

disheep drawio

快速上手

  • 如果想在线试玩 --> 在线网页(改进中)
  • 如果想本地部署/测试
    • 服务端(Python)
      # 预先安装好 Python3
      cd service
      pip install -r requirement.txt
      FLASK_APP=app.py flask run  # 玩家试玩
      # FLASK_APP=agent_app.py flask run  # 玩家 + AI 试玩
    • 客户端(react)
      # 预先安装好 node.js 和 react
      cd ui
      npm run build
      npm run preview
      然后在网页中打开对应链接即可
  • 如果想进行完整的深度强化学习训练
    # 预先安装好 Python3
    cd service
    pip install -r requirement-train.txt
    python3 -u sheep_ppo_main.py
  • 如果想使用定义好的 gym 羊了个羊环境 --> 点个 star 之后直接暴力 CTRL C+V 拿走 service/sheep_env.py 尽情魔改
  • 如果想获得训练好的深度强化学习模型 --> 访问 OpenDILab官网下载链接 (目前提供了两种试玩模型,但智能体仍有很多进步空间)
  • 如果想了解更多深度强化学习相关知识 --> 欢迎参阅 DI-engine相关文档
  • 如果想了解未来的更新计划 --> 请参阅更新计划
  • 如果有其他问题或想法 --> 欢迎 github ISSUE 区讨论,或是贡献 Pull requests

项目结构

.
├── LICENSE
├── ui                       --> react 网页前端
└── service                  --> Python 核心模块(算法和服务端)
    ├── app.py                  --> flask 服务 app (仅人类操作)
    ├── agent_app.py                  --> flask 服务 app(人类+AI操作)
    ├── requirement.txt         --> Python 依赖库列表
    ├── sheep_env.py            --> gym 格式环境
    ├── sheep_model.py          --> 基于 PyTorch 的 Actor-Critic 神经网络模型
    ├── sheep_ppo_main.py       --> 基于 DI-engine 的深度强化学习训练主函数
    ├── test_sheep_env.py       --> gym 格式环境的单元测试
    └── test_sheep_model.py     --> 神经网络模型的单元测试

更新计划

算法

  • 强化学习训练参数调整和算法微调
  • 提供可供本地试玩的模型权重
  • 详细的神经网络和强化学习算法设计文档
  • model-based RL 和 planning 算法
  • 神经网络压缩(用于部署)

环境

  • 添加原类型游戏(比如3tiles)中的各种道具
  • 牌的层数问题如何定义
  • 结合 JAX 优化环境运行速度
  • 更多结合 AI 的玩法设计

应用

  • 更多自定义主题和 BGM
  • 在线网页端部署
  • 样例神经网络权重下载
  • 更多 AI 训练时的行为分析
  • 移动端应用(欢迎Android/IOS开发者支持)

致谢

License

DI-sheep is released under the Apache 2.0 license.

More Repositories

1

awesome-RLHF

A curated list of reinforcement learning with human feedback resources (continually updated)
3,262
star
2

DI-engine

OpenDILab Decision AI Engine. The Most Comprehensive Reinforcement Learning Framework B.P.
Python
3,041
star
3

PPOxFamily

PPO x Family DRL Tutorial Course(决策智能入门级公开课:8节课帮你盘清算法理论,理顺代码逻辑,玩转决策AI应用实践 )
Python
1,875
star
4

DI-star

An artificial intelligence platform for the StarCraft II with large-scale distributed training and grand-master agents.
Python
1,215
star
5

LightZero

[NeurIPS 2023 Spotlight] LightZero: A Unified Benchmark for Monte Carlo Tree Search in General Sequential Decision Scenarios (awesome MCTS)
Python
1,097
star
6

awesome-model-based-RL

A curated list of awesome model based RL resources (continually updated)
851
star
7

awesome-diffusion-model-in-rl

A curated list of Diffusion Model in RL resources (continually updated)
739
star
8

awesome-decision-transformer

A curated list of Decision Transformer resources (continually updated)
671
star
9

LMDrive

[CVPR 2024] LMDrive: Closed-Loop End-to-End Driving with Large Language Models
Jupyter Notebook
592
star
10

DI-drive

Decision Intelligence Platform for Autonomous Driving simulation.
Python
563
star
11

InterFuser

[CoRL 2022] InterFuser: Safety-Enhanced Autonomous Driving Using Interpretable Sensor Fusion Transformer
Python
522
star
12

LLMRiddles

Open-Source Reproduction/Demo of the LLM Riddles Game
Python
515
star
13

GoBigger

[ICLR 2023] Come & try Decision-Intelligence version of "Agar"! Gobigger could also help you with multi-agent decision intelligence study.
Python
459
star
14

awesome-end-to-end-autonomous-driving

A curated list of awesome End-to-End Autonomous Driving resources (continually updated)
371
star
15

awesome-multi-modal-reinforcement-learning

A curated list of Multi-Modal Reinforcement Learning resources (continually updated)
367
star
16

awesome-exploration-rl

A curated list of awesome exploration RL resources (continually updated)
365
star
17

SO2

[AAAI2024] A Perspective of Q-value Estimation on Offline-to-Online Reinforcement Learning
Python
285
star
18

DI-engine-docs

DI-engine docs (Chinese and English)
Python
281
star
19

DI-orchestrator

OpenDILab RL Kubernetes Custom Resource and Operator Lib
Go
240
star
20

DI-smartcross

Decision Intelligence platform for Traffic Crossing Signal Control
Python
230
star
21

treevalue

Here are the most awesome tree structure computing solutions, make your life easier. (这里有目前性能最优的树形结构计算解决方案)
Python
228
star
22

DI-hpc

OpenDILab RL HPC OP Lib, including CUDA and Triton kernel
Python
222
star
23

awesome-AI-based-protein-design

A collection of research papers for AI-based protein design
216
star
24

ACE

[AAAI 2023] Official PyTorch implementation of paper "ACE: Cooperative Multi-agent Q-learning with Bidirectional Action-Dependency".
Python
212
star
25

DI-treetensor

Let DI-treetensor help you simplify the structure processing!(树形运算一不小心就逻辑混乱?DI-treetensor快速帮你搞定)
Python
202
star
26

GoBigger-Challenge-2021

Interested in multi-agents? The 1st Go-Bigger Multi-Agent Decision Intelligence Challenge is coming and a big bonus is waiting for you!
Python
195
star
27

Gobigger-Explore

Still struggling with the high threshold or looking for the appropriate baseline? Come here and new starters can also play with your own multi-agents!
Python
185
star
28

DI-store

OpenDILab RL Object Store
Go
177
star
29

LightTuner

Python
173
star
30

DOS

[CVPR 2023] ReasonNet: End-to-End Driving with Temporal and Global Reasoning
Python
145
star
31

DI-toolkit

A simple toolkit package for opendilab
Python
113
star
32

DI-bioseq

Decision Intelligence platform for Biological Sequence Searching
Python
111
star
33

DI-1024

1024 + 深度强化学习(Deep Reinforcement Learning + 1024 Game/ 2048 Game)
Python
109
star
34

SmartRefine

[CVPR 2024] SmartRefine: A Scenario-Adaptive Refinement Framework for Efficient Motion Prediction
Python
107
star
35

DIgging

Decision Intelligence for digging best parameters in target environment.
Python
90
star
36

awesome-driving-behavior-prediction

A collection of research papers for Driving Behavior Prediction
77
star
37

PsyDI

PsyDI: Towards a Personalized and Progressively In-depth Chatbot for Psychological Measurements. (e.g. MBTI Measurement Agent)
TypeScript
70
star
38

DI-adventure

Decision Intelligence Adventure for Beginners
Python
68
star
39

GenerativeRL

Python library for solving reinforcement learning (RL) problems using generative models (e.g. Diffusion Models).
Python
48
star
40

huggingface_ding

Auxiliary code for pulling, loading reinforcement learning models based on DI-engine from the Huggingface Hub, or pushing them onto Huggingface Hub with auto-created model card.
Python
46
star
41

CodeMorpheus

CodeMorpheus: Generate code self-portraits with one click(一键生成代码自画像,决策型 AI + 生成式 AI)
Python
45
star
42

OpenPaL

Building open-ended embodied agent in battle royale FPS game
33
star
43

awesome-ui-agents

A curated list of of awesome UI agents resources, encompassing Web, App, OS, and beyond (continually updated)
31
star
44

.github

The first decision intelligence platform covering the most complete algorithms in academia and industry
19
star
45

CleanS2S

High-quality and streaming Speech-to-Speech interactive agent in a single file. 只用一个文件实现的流式全双工语音交互原型智能体!
1
star