adjoint 话题在 GitHub 上没有填写简介，这里按 Star 数汇总最相关的热门仓库。
The fastest automatic differentiation library for C++, with Python bindings. Forward, adjoint and higher-order derivatives, checkpointing, Eigen support and JIT. Reproducible benchmarks against CppAD, Adept and autodiff.
：shrimp：电磁仿真 + 自动微分
A component of the DiffEq ecosystem for enabling sensitivity analysis for scientific machine learning (SciML). Optimize-then-discretize, discretize-then-optimize, adjoint methods, and more for ODEs, SDEs, DDEs, DAEs, etc.
DAFoam：基于OpenFOAM的离散伴随，实现高精度多学科设计优化
TCLB - Templated MPI+CUDA/CPU Lattice Boltzmann code
线性和非线性器件的频域光子仿真与逆向设计优化
GPU accelerated differentiable finite elements for solid mechanics with PyTorch
一组用于拓扑优化基准测试的光子逆设计挑战问题
Differentiable interface to FEniCS/Firedrake for JAX using dolfin-adjoint/pyadjoint
Differentiable interface to FEniCS for JAX
Workshop materials for training in scientific computing and scientific machine learning
Differentiable ensemble solvers for differential equations on GPUs, in JAX
Julia interface to MITgcm
Differentiable interface to FEniCS for PyMC3
A Pytorch implementation of the radon operator and filtered backprojection with, except for a constant, adjoint radon operator and backprojection.
A library for high-level algorithmic differentiation
Reverse-mode automatic differentiation with delimited continuations
Adjoint-based optimization and inverse design of photonic devices.
Differentiable interface to Firedrake for JAX
Automatic differentiation of FEniCS and Firedrake models in Julia