implementing the finite difference method in C++
C++ Python Makefile Shell
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
doc
examples
src
tests
.gitignore
Doxyfile
README.md
branches_description.md
changelog.md
deploy.sh
makefile
testing.sh

README.md

Finite Differences

I'm trying to implement the finite difference method for solving differential equations.

  • You can read the documentation by Doxygen (work in progress).
  • The mathematics is developed here as I'm learning it.

Purpose

The official aim of this program is to implement the finite difference method for solving differential equations.

Non-official purposes include

  • A first experiment of implementing numerical methods.
  • I (believe to) know that the finite difference method requires solving system of equations with sparse matrices. However I'm implementing the PLU decomposition for solving the system while I (believe to) know that there are more efficient ways to solve for sparse matrices.
  • Learning the cppunit framework for my unit tests.
  • Programming in C++ with no raw pointers.

Latest news

  • The PLU decomposition is done.
  • Next step : solving linear systems. See the branch alpha.

deploy.sh

The script deploy.sh does the following things :

  • show git status
  • make clean
  • make all
  • launch the unit tests

The test are covering more or less everything, but the most interesting ones are in plu_unit_tests.cpp. There are many matrices for which the PLU decomposition are tested against the one provided by sage.