Skip to content

Latest commit

 

History

History
74 lines (51 loc) · 3.07 KB

README.md

File metadata and controls

74 lines (51 loc) · 3.07 KB
Raw

CMake

BSpline Interpolation Lib

This library use B-spline to interpolate given data of any dimension on a Cartesian mesh grid.

Perquisite

This library requires C++11. CMake is required only for build and run tests.

How to use

Since the library is header only, include the file include/Interpolation.hpp and one can use all its functionality under the namespace intp. Check test/interpolation-test.cpp and test/interpolation-template-test.cpp for more info.

Packaging using CMake

This library can be installed using CMake by executing the following commands:

git clone https://github.com/12ff54e/BSplineInterpolation.git
cd BSplineInterpolation

cmake -B build -DCMAKE_INSTALL_PREFIX=/folder/you/want/to/install/in
cmake --install build

If you do not add CMAKE_INSTALL_PREFIX in configuration, you may need sudo to perform the installation. Then in the CMakeLists.txt file you can do:

find_package(BSplineInterpolation) # can also specify version here

add_executable(main main.cpp)
target_link_libraries(main BSplineInterpolation)

and in the source file:

#include <BSplineInterpolation/Interpolation.hpp>

using namespace intp;
InterpolationFunction<double, 2> func(
    3, // interpolation function order
    z_mesh, // mesh storing z values, an object of type intp::Mesh<double, 2>
    std::make_pair(x_min, x_max), // x range
    std::make_pair(y_min, y_max)); // y range

Note: this project follows Semantic Version 2.0.0 so the interface will be compatible within one major version.

Modify Library's Behavior

You can control the bahavior of this library by defining the following macros:

  • INTP_TRACE, INTP_DEBUG: Defines levels of logging, with decreasing verbosity.
  • INTP_MULTITHREAD: Multi-threading in solving control points.
  • INTP_ENABLE_ASSERTION: Add assertion to ensure data to be interplolated is valid.
  • INTP_PERIODIC_NO_DUMMY_POINT: Whether to accept dummy point when interpolating periodic data. If this macro is defined, number of data point is one less than the specified dimension, since the last point is implicitly set as the same of the first one.
  • INTP_STACK_ALLOCATOR: Use stack allocator in some small dynamic memory allocation scenario.
  • INTP_CELL_LAYOUT: Store weights in cell layout redundantly to speed up evaluation.

By default all of the above macros is not defined.

Matlab Interface

Check matlab/Example.m for instructions. Please notice that matlab/bspline.mexw64 is compiled for windows platform, you may need to compile your own MEX file from bspline.cpp (basically a wrapper) on other platforms.

Note

  • Limited compiler support: only tested using g++ 11.3 and clang++ 14.0 (on Ubuntu).
  • The Interpolation Template feature is not efficient in dimension 2 or higher.

Known Issues

  • Not found yet

Future Plan

  • Support multithreading.
  • Interpolation with boundary derivative specified.