Adjoint Python Packages

ceviche

:shrimp: Electromagnetic Simulation + Automatic Differentiation

ceviche-challenges

A suite of photonic inverse design challenge problems for topology optimization benchmarking

torch-fem

GPU accelerated differentiable finite elements for solid mechanics with PyTorch

angler

Frequency-domain photonic simulation and inverse design optimization for linear and nonlinear devices

kirchcig

GPU Kirchhoff migration to common-image gathers (CIG), with an exact adjoint. CUDA/CuPy, offset and angle domain, PyTorch autograd, ready for least-squares migration.

gradsolve

Differentiable ensemble solvers for differential equations on GPUs, in JAX

conduction

Python package for solving implicit heat conduction