NonlinearIntegrators.jl is a affliated package in the GeometricIntegrators.jl community. This package aims to generalize continuous Galerkin variational integrators from linear basis to nonlinear basis for achieving large time-step integration.
Till now, several options for nonlinear basis are available, but only neural network basis with one hidden layer is frequently used and well maintained.
- One Hidden Layer Neural Network (Shallow Neural Network)
- Implemented with Lux
- Implemented with GeometricMachineLearning.jl and SymbolicNeuralNetwork.jl
- Deep Neural Networks.
- Nested Sindy
GeometricIntegrators.jl and all of its dependencies can be installed via the Julia REPL by typing
]add NonlinearIntegrators.jl
We are using git hooks, e.g., to enforce that all tests pass before pushing. In order to activate these hooks, the following command must be executed once:
git config core.hooksPath .githooks