README
¶
Gosl - Go scientific library
Gosl is a set of tools for developing scientific simulations using the Go language. We mainly consider the development of numerical methods and solvers for differential equations but also present some functions for fast Fourier transforms, the generation of random numbers, probability distributions, and computational geometry.
This library contains essential functions for linear algebra computations (operations between all combinations of vectors and matrices, eigenvalues and eigenvectors, linear solvers) and the development of numerical methods (e.g. numerical quadrature).
We link Gosl with existent libraries written in C and Fortran, such as OpenBLAS, LAPACK, UMFPACK, MUMPS, QUADPACK and FFTW3. These existing libraries have been fundamental for the development of high-performant simulations over many years. We believe that it is nearly impossible to rewrite these libraries in native Go and at the same time achieve the same speed delivered by them. Just for reference, a naive implementation of matrix-matrix multiplication in Go is more than 100 times slower than OpenBLAS.
Installation
Because of the other libraries, the easiest way to work with Gosl is via Docker. Having Docker and VS Code installed, you can start developing powerful numerical simulations using Gosl in a matter of minutes. Furthermore, it works on Windows, Linux, and macOS out of the box.
Containerized
- Install Docker
- Install Visual Studio Code
- Install the Remote Development extension for VS Code
- Clone https://github.com/cpmech/hello-gosl
- Create your application within a container (see gif below)
Done. And your system will "remain clean."
Debian/Ubuntu GNU Linux
First, install Go as explained in https://golang.org/doc/install
Second, install some libraries:
sudo apt-get install \
gcc \
gfortran \
libfftw3-dev \
liblapacke-dev \
libmetis-dev \
libmumps-seq-dev \
libopenblas-dev \
libsuitesparse-dev
Finally, download and compile Gosl:
git clone https://github.com/cpmech/gosl.git
cd gosl
./all.bash
Done. Installation completed.
Documentation
Gosl includes the following essential packages:
- chk. To check numerical results and for unit testing
- io. Input/output including printing to the terminal and handling files
- utl. To generate series (e.g. linspace) and other functions as in pylab/matlab/octave
- la. Linear Algebra: vector, matrix, efficient sparse solvers, eigenvalues, decompositions
Gosl includes the following main packages:
- fun. Special functions, DFT, FFT, Bessel, elliptical integrals, orthogonal polynomials, interpolators
- gm. Geometry algorithms and structures
- hb. Pseudo hierarchical binary (hb) data file format
- num. Fundamental numerical methods such as root solvers, non-linear solvers, numerical derivatives and quadrature
- ode. Solvers for ordinary differential equations
- opt. Numerical optimization: Interior Point, Conjugate Gradients, Powell, Grad Descent
- pde. Solvers for partial differential equations (FDM, Spectral, FEM)
- rnd. Random numbers and probability distributions
(see each subdirectory for more information)
For the sake of maintenance (see next section), we have removed the previous mpi sub-package. However, we recommend the external library gompi if you plan to use MPI.
Previous version
The previous version, including more packages, is available here and can be used with the Docker image 1.1.3 as in this hello gosl example.
These other packages, such as machine learning, plotting, etc., have been removed because they do not depend on CGO and may be developed independently. We can now maintain the core of Gosl more efficiently, which has a focus on the foundation for other scientific code.
Directories
¶
| Path | Synopsis |
|---|---|
|
Package chk contains functions for checking and testing computations
|
Package chk contains functions for checking and testing computations |
|
Package fun (functions) implements special functions such as elliptical, orthogonal polynomials, Bessel, discrete Fourier transform, polynomial interpolators, and more.
|
Package fun (functions) implements special functions such as elliptical, orthogonal polynomials, Bessel, discrete Fourier transform, polynomial interpolators, and more. |
|
fftw
Package fftw wraps the FFTW library to perform Fourier Transforms using the "fast" method by Cooley and Tukey
|
Package fftw wraps the FFTW library to perform Fourier Transforms using the "fast" method by Cooley and Tukey |
|
Package gm (geometry and meshes) implements auxiliary functions for handling geometry and mesh structures
|
Package gm (geometry and meshes) implements auxiliary functions for handling geometry and mesh structures |
|
msh
Package msh defines mesh data structures and implements interpolation functions for finite element analyses (FEA)
|
Package msh defines mesh data structures and implements interpolation functions for finite element analyses (FEA) |
|
tri
Package tri wraps Triangle to perform mesh generation a Delaunay triangulation
|
Package tri wraps Triangle to perform mesh generation a Delaunay triangulation |
|
Package hb implements a pseudo hierarchical binary (hb) data file format
|
Package hb implements a pseudo hierarchical binary (hb) data file format |
|
Package io (input/output) implements auxiliary functions for printing, parsing, handling files, directories, etc.
|
Package io (input/output) implements auxiliary functions for printing, parsing, handling files, directories, etc. |
|
Package la implements functions and structure for Linear Algebra computations.
|
Package la implements functions and structure for Linear Algebra computations. |
|
oblas
Package oblas implements lower-level linear algebra routines using OpenBLAS for maximum efficiency.
|
Package oblas implements lower-level linear algebra routines using OpenBLAS for maximum efficiency. |
|
Package num implements fundamental numerical methods such as numerical derivative and quadrature, root finding solvers (Brent's and Newton's methods), among others.
|
Package num implements fundamental numerical methods such as numerical derivative and quadrature, root finding solvers (Brent's and Newton's methods), among others. |
|
qpck
Package qpck wraps the QUADPACK library for numerical integration
|
Package qpck wraps the QUADPACK library for numerical integration |
|
Package ode implements solvers for ordinary differential equations, including explicit and implicit Runge-Kutta methods; e.g.
|
Package ode implements solvers for ordinary differential equations, including explicit and implicit Runge-Kutta methods; e.g. |
|
Package opt implements routines for solving optimization problems
|
Package opt implements routines for solving optimization problems |
|
Package pde implements numerical methods for the solution of Partial Differential Equations.
|
Package pde implements numerical methods for the solution of Partial Differential Equations. |
|
Package rnd implements random numbers generators (wrapping the standard functions or the Mersenne Twister library).
|
Package rnd implements random numbers generators (wrapping the standard functions or the Mersenne Twister library). |
|
dsfmt
Package dsfmt wraps the dSFMT Double precision SIMD-oriented Fast Mersenne Twister
|
Package dsfmt wraps the dSFMT Double precision SIMD-oriented Fast Mersenne Twister |
|
sfmt
Package sfmt wraps the SFMT SIMD-oriented Fast Mersenne Twister
|
Package sfmt wraps the SFMT SIMD-oriented Fast Mersenne Twister |
|
Package utl implements functions for simplifying calculations and allocation of structures such as slices and slices of slices.
|
Package utl implements functions for simplifying calculations and allocation of structures such as slices and slices of slices. |