~mpesz/code

THIS SITE IS PERMANENTLY UNDER CONSTRUCTION and the software available for download can change or temporarily disappear so we can make it better.
The code(s) authors are listed and the copyright to Department of Mathematics has been indicated. When using the software for your needs, please acknowledge the authors, the copyright, and the funding listed below. You are also welcome to send me an email to let us know you've used the code.
The software available through these pages includes prototypes only. For example, it works in 1D only, or with kludges and commented-out code to support testing on some elementary examples, or to promote the use for some specific applications. [Hence, the code ain't always pretty]. The code should always work on examples indicated in the header but not necessarily on all the combinations of code you would un-comment. We have full research versions available but unfortunately do not have the resources to provide documentation and/or support, so we will likely not be able to answer questions on the code. If you would like to collaborate on full research versions, and are affiliated with an academic institution as a student or researchers, please contact me so we can work out a way to do it.

Please consider visiting for a different layout od CODE and CODE CAPSULES, with lionks to github
Links
General
Research software
Code for classes
Utilities and other
Software

The scripts and templates below are well adjusted to the Numerical Analysis courses at OSU. They may be useful for people in other institutions. However, be aware that many of the algorithms may have better alternatives, for example, native MATLAB routines.

Please README first.

  • Numerical BVPs and PDEs
    • Template for solving a two-point Boundary Value Problem using point-centered finite difference method fd1d.m (by M. Peszynska)
    • Template for solving a two-point Initial Boundary Value Problem using point-centered finite difference method and backward Euler discretization in time fd1d_heat.m (by M. Peszynska)
    • Template for solving a two-point Boundary Value Problem using finite elements fem1d.m (by M. Peszynska)
    • Utility for setting up quadrature points for purpose of estimating finite element error tri_quadcofs.m (by M. Peszynska, copied from standard texts)
  • Vector Calculus:
  • Numerical Linear Algebra:
    • Conjugate Gradient Method: mycgs.m (by M. Peszynska)
    • Conjugate Gradient versus Steepest Descent 3D demo: cgdemo.m (by D. Fettig)
    • Preconditioned Conjugate Gradient Method: mypcg.m (by M. Peszynska)
    • Demo of how the use of preconditioners affects the spectral properties of a matrix and the number of iterations in PCG: pcgdemo.m (by D. Fettig, M. Peszynska)
    • SVD (Singular Value Decomposition) demo of application to image processing: svd_demo.m (by M. Peszynska)
    • SVD (Singular Value Decomposition) demo ... actual image: svd_demo_with_image.m (by J. Kyle)
    • sample BMP image arcdbk.bmp (by J. Kyle)
    • Gram_Schmidt: classical and modified orthogonalization: Gram_Schmidt.m (by M. Peszynska)
    • demonstration of the loss and recapture of orthogonality in classical and modified Gram-Schmidt methods testgram.m (by D. Fettig, M. Peszynska)
  • Numerical ODEs
    • Introduction to MATLAB and Differential Equations worksheet 552-first.txt (by M. Peszynska)
    • Forward Euler method demonstration FE.m (by M. Peszynska)
    • Backward Euler method demonstration with Newton/Picard' iterative solvers BE.m (by M. Peszynska)
    • How to use step Richardson's halving/doubling to get error estimates error_estimate_demo.m (by M. Peszynska)