Welcome to phlearn’ documentation

phlearn is a python package for modelling pseudo-Hamiltonian systems with neural networks as decribed in (Eidnes et al. 2022).

Installation

The phlearn package is available via PyPi:

$ pip install phlearn

Alternatively, to get the latest updates not yet available on PyPi, you can clone the repositrory from Github and install directly:

$ git clone https://github.com/SINTEF/pseudo-Hamiltonian-neural-networks.git
$ cd pseudo-Hamiltonian-neural-networks
$ pip install -e phlearn

Example use

Example scripts and notebooks can be found in the Github repo.

phlearn API

phlearn is a package for simulating, controlling and learning dynamical systems in general, and pseudo-Hamiltonian systems in particular. phlearn contains three subpackages:

  • phsystems with tools for simulation

  • phnns with tools for constructing and training pseudo-Hamiltonian neural networks

  • control for control functionality

  • utils for convenience functions.