IncompressibleNavierStokes.jl is a differentiable open-source solver for incompressible Navier-Stokes equations that supports GPU acceleration, high-resolution simulations, and a-posteriori training of neural network closure models for turbulence.
Don’t Unroll Adjoint: Differentiating SSA-FormPrograms,March2019
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
math.NA 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
A differentiable software suite for accelerated simulation of turbulent flows
IncompressibleNavierStokes.jl is a differentiable open-source solver for incompressible Navier-Stokes equations that supports GPU acceleration, high-resolution simulations, and a-posteriori training of neural network closure models for turbulence.