CFDML (Cross-disciplinary Fluid Dynamics and Machine Learning) is an open, collaborative alliance of scientists and engineers (from CAS, USTC, PKU, etc.), dedicated to advancing research, software, and tools at the convergence of Fluid Dynamics and Machine Learning. Our mission is to support, develop, and share open-source solutions that enable next-generation fluid dynamic modeling and simulation.
Our core focus includes:
- ⚙️ Efficient CFD solvers and infrastructures integrating data-driven methods
- 🔬 Physics-informed ML models for multi-scale and multi-physics flows
- 📊 Benchmarks and datasets for evaluating unified CFDML approaches
- 📚 Educational resources, lectures, and tutorials for community learning
CFDML thrives on open collaboration—whether you’re a researcher, engineer, developer, or student, there are ways to get involved:
- ⭐ Explore our repositories: Check out our repositories.
- 🐛 Report issues: Open an issue in any repository to report bugs or request features.
- 🔀 Submit Pull Requests: Find issues labeled
good first issueorhelp wantedto begin contributing. - 📄 Share knowledge: Write or improve documentation, add examples, or contribute to our learning resources.
Before submitting a PR, please check for a CONTRIBUTING.md file in that repository for guidelines on coding standards, testing, and our development process.
Our work is generously supported by:
- German Science Foundation (DFG)
- Alexander von Humboldt Foundation
- Hong Kong Research Grants Council
- National Natural Science Foundation of China
- Chinese Academy of Sciences