Skip to main content

AI-powered symbolic regression engine that discovers governing equations from data.

Project description

SymbolicMind

一个能从数据中自主发现物理定律的 AI 引擎。

License: MIT

SymbolicMind 是一个因果符号回归引擎,基于 BIC 精拟合P0 可证伪性边界 构建。它不仅拟合数据,还能自我修正、自我发现、自我进化——让机器从原始观测数据中提炼出人类可读的数学控制方程。

核心能力

  • 方程发现:自动从时间序列或静态数据中识别非线性动力学方程。
  • 可证伪性:内置 P0 探测器,通过残差自相关判别模型的可靠性,绝不捏造物理。
  • 自演化算子库:在需要时自主生成时变、分数阶、嵌套积分核等高级数学结构。
  • 恒等式识别:自动识别 dx/dt = y 等简单恒等式,避免过度建模。
  • AI 智能体就绪:提供符合 OpenAI Function Calling 规范的工具接口,可被任何 AI Agent 调用。
  • 记忆增强:本地经验库记录成功发现,越用越聪明。

快速开始

pip install numpy scipy scikit-learn
python quickstart.py

你将看到引擎发现简谐振动方程:d²x/dt² = -x,并输出详细的诊断报告。

基准测试成绩
SymbolicMind  Feynman、Strogatz、ODEBench  Blackbox 等基准上取得了领先成绩。详见 BENCHMARKS.md。

项目结构
text
SymbolicMind/
├── symbolimind/          # 核心引擎包   ├── engine.py         # BIC 精拟合器 + P0 探测器   ├── extensions.py     # 恒等式检测 + 自适应断层扫描   ├── skill.py          # AI 智能体工具接口   ├── brain.py          # 自然语言大脑   └── memory.py         # 本地经验记忆库
├── tests/                # 测试
├── BENCHMARKS.md         # 基准测试成绩单
├── CONTRIBUTING.md       # 贡献指南
├── LICENSE               # MIT 许可证
├── quickstart.py         # 5 分钟快速体验
└── README.md             # 本文件
引用
如果 SymbolicMind 对你的研究有帮助,请引用本项目。

贡献
我们欢迎社区贡献!无论是新的物理算子、基准测试还是文档改进,请参见 CONTRIBUTING.md。

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

symbolimind-0.3.0.tar.gz (38.1 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

symbolimind-0.3.0-py3-none-any.whl (41.0 kB view details)

Uploaded Python 3

File details

Details for the file symbolimind-0.3.0.tar.gz.

File metadata

  • Download URL: symbolimind-0.3.0.tar.gz
  • Upload date:
  • Size: 38.1 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.11.9

File hashes

Hashes for symbolimind-0.3.0.tar.gz
Algorithm Hash digest
SHA256 cc2cb90205ff8511b308ee700e0916c370bd83f70b67921b3b1728999561caad
MD5 211e6e265ab85c341370187f513ea5f9
BLAKE2b-256 fcd30b116d9b4ad470bd994c23efd28d56f69361027c8c90d57ccd82d8a3e32a

See more details on using hashes here.

File details

Details for the file symbolimind-0.3.0-py3-none-any.whl.

File metadata

  • Download URL: symbolimind-0.3.0-py3-none-any.whl
  • Upload date:
  • Size: 41.0 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.11.9

File hashes

Hashes for symbolimind-0.3.0-py3-none-any.whl
Algorithm Hash digest
SHA256 35ad536420caff80a4398c321ebb595de0ef725cd80e9352662ad1cb8b342220
MD5 6b142548020fc5bc1fc2deab93bc9eb4
BLAKE2b-256 1f4a1aa6592967a6c1f7c163f329bbf1f318525b6dc81f30fcdf58e48631c416

See more details on using hashes here.

Supported by

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