SOFTWARE FOR
Computational Mathematics
Software for Computational Mathematics is used to find solutions to large-scale computational problems arising in science, engineering, industry and commerce. It allows for the design, comparison and improvement of new and existing solvers.
CUTEst – a versatile testing environment for optimisation and linear algebra solvers
CAPABILITIES
CUTEst is a versatile testing environment for optimisation and linear algebra solvers. It provides access to over 1150 example test problems and interfaces to over 50 optimisation solvers. The software helps developers design, compare and improve new and existing solvers.
TRAINING INFORMATION
STFC does not currently provide training for this software. Full documentation is available.
ACCESS
This software is available via GitHub.
FitBenchmarking – software for comparing different minimizers/fitting frameworks
CAPABILITIES
Funded by ALC, FitBenchmarking is an open source tool for comparing different minimizers/fitting frameworks. It forms a bridge between domain specific software, and the mathematical algorithms that analyze the data, ensuring that:
- Scientists, know what is the best algorithm for fitting their model to data they might encounter, on their specific hardware;
- Scientific software developers know what is the state-of-the-art in fitting algorithms and implementations, what they should recommend as their default solver, and if they should implement a new method in their software; and
- Mathematicians and numerical software developers understand the types of problems on which current algorithms do not perform well, and to have a route to expose newly developed methods to users.
TRAINING INFORMATION
STFC does not currently provide training for this software. Full documentation is available.
ACCESS
This software is available via GitHub.
GALAHAD – a thread safe library of modern Fortran packages for solving nonlinear optimisation problems
CAPABILITIES
GALAHAD is a thread-safe library of modern Fortran packages for solving nonlinear optimisation problems.
It offers unconstrained and bound-constrained optimisation, linear, nonlinear and quadratic programming, systems of nonlinear equations and inequalities, as well as the capability for global optimisation.
TRAINING INFORMATION
STFC does not currently provide training for this software. Full documentation is available.
ACCESS
This software is available via GitHub
GOFit – software designed for the global optimization of parameters in curve-fitting problems
CAPABILITIES
Funded by ALC, GOFit is a package of C++ algorithms with python interfaces designed for the global optimization of parameters in curve fitting, i.e. for nonlinear least-squares problems arising from curve fitting. GOFit was developed with scientific curve fitting problems in mind but is also applicable to general curve fitting problems provided they can be formulated as nonlinear least-squares problems.
TRAINING INFORMATION
STFC does not currently provide training for this software. Full documentation is available.
ACCESS
This software is available via GitHub.
HSL – a collection of state-of-the-art packages for large-scale scientific computation
CAPABILITIES
HSL is a collection of state-of-the-art packages for large-scale scientific computation. It has an international reputation as a source of robust and efficient numerical software, which can be used for eigenvalues and eigenvectors, solving sparse linear systems of equations and nonlinear equations.
TRAINING INFORMATION
STFC does not currently provide training for this software. Full documentation is available.
ACCESS
This software is available free for academic purposes, and to purchase here.
RALFit – software package implementing algorithms and regularization techniques for small to medium-sized nonlinear-least squares problems
CAPABILITIES
RALFit implements various algorithms and regularization techniques for small to medium-sized nonlinear-least squares problems. Access to the algorithms is provided through C/C++/Fortran or Python interfaces.
TRAINING INFORMATION
STFC does not currently provide training for this software. Full documentation is available.
ACCESS
This software is available via GitHub.
SPRAL – an open-source library for sparse linear algebra algorithms
CAPABILITIES
Funded by CoSeC, the Sparse Parallel Robust Algorithms Library, SPRAL, is an open-source library for sparse linear algebra algorithms. Some of the major packages include:
- SSIDS: a GPU-enabled sparse solver for symmetric indefinite linear systems;
- SSMFE: a package that uses the Jacobi-conjugate preconditioned gradients method to solve sparse symmetric eigenvalue problems matrix-free; and
- LSMR: an iterative solver for space least-squares problems using the LSMR algorithm.
TRAINING INFORMATION
STFC does not currently provide training for this software. Full documentation is available.
ACCESS
This software is available via GitHub.
Sorry, nothing found.
There are no options that match your search.
Please try another option, or click here to reset the filters.
