PDEBench and CFDBench

These are complements that address different scopes of the same problem: PDEBench provides a broad benchmark across multiple PDE types including fluid dynamics, while CFDBench specializes in large-scale fluid dynamics benchmarks with potentially deeper domain-specific scenarios, allowing practitioners to validate ML models across both general PDE solving and fluid-specific challenges.

PDEBench
70
Verified
CFDBench
41
Emerging
Maintenance 13/25
Adoption 10/25
Maturity 25/25
Community 22/25
Maintenance 6/25
Adoption 10/25
Maturity 8/25
Community 17/25
Stars: 1,082
Forks: 141
Downloads:
Commits (30d): 1
Language: Python
License:
Stars: 261
Forks: 33
Downloads:
Commits (30d): 0
Language: Python
License:
No risk flags
No License No Package No Dependents

About PDEBench

pdebench/PDEBench

PDEBench: An Extensive Benchmark for Scientific Machine Learning

This project provides a comprehensive benchmark for evaluating machine learning models designed to solve Partial Differential Equations (PDEs). It offers a wide range of realistic physical problems, along with ready-to-use datasets containing various initial/boundary conditions and PDE parameters. Scientists, engineers, and researchers working with scientific machine learning can use this to compare and develop methods for simulating complex physical phenomena.

scientific-machine-learning computational-physics numerical-simulation differential-equations model-benchmarking

About CFDBench

luo-yining/CFDBench

A large-scale benchmark for machine learning methods in fluid dynamics

This project provides a comprehensive collection of fluid dynamics simulations to help researchers and engineers evaluate how well machine learning models can predict fluid behavior. It includes various scenarios like fluid flowing through tubes or around objects, with different conditions and geometries. If you're developing or testing machine learning models for computational fluid dynamics (CFD), this benchmark helps you assess their generalizability across diverse real-world problems.

fluid-dynamics computational-fluid-dynamics machine-learning-evaluation simulation-benchmarking engineering-design

Scores updated daily from GitHub, PyPI, and npm data. How scores work