lib_math

This library is a mathematics extension library based on Eigen. Some useful tiny functions that were created in some of my projects are rearranged and templated in this project to facilitate my work.

The library will be updated continuously.

Requirement

• CMake tool (version should be no less than 3.10).
• Eigen library (version should be no less that 3.4.0) .

How to use?

Download this library from github.

git clone https://github.com/wlfrii/lib_math.git

I. Configure the library in your project

I.1 For Linux / Mac OS (recommended)

After unpackaged the repository, do the <make&install> in your terminal as follows.

cd lib_math
mkdir build
cd build
cmake ..
make -j
sudo make install

Then you can use this library in your project by adding the following commands in your CMakeLists.txt.

# Find lib_math
find_package(lib_math REQUIRED)

# Inlcude headers path of lib_math to your project
target_include_directories(${PROJECT_NAME}  PUBLIC
    $<BUILD_INTERFACE:${lib_math_INCLUDE_DIRS}>
)

# Link lib_math to your project
target_link_libraries(${PROJECT_NAME} PUBLIC
    ${lib_math_LIBRARIES}
)

I.2 For Windows

Open your terminal and check your cmake supported Visual Studio version first by type the command

cmake -G

Then find the generator name that has an asterisk * in front of the name, which is the support generator. For example, * Visual Studio 16 2019.

After unpackaged the repository, do the <make&install> in your terminal as follows.

cd lib_math
mkdir build
cd build
cmake "Visual Studio 16 2019" ..
make -j
sudo make install

Finally, open the lib_math.sln in your build/ folder by your Visual Studio and compile the solution. After compile done, you can include this library into you project by add the path of headers lib_math/export/ and static library lib_math.lib.

Possible problem:

• Cannot find Eigen3. Fix: Comment the Line:44-Line:49 of the CMakeLists.txt, and replace ${EIGEN3_INCLUDE_DIRS} in Line:58 of the CMakeLists.txt by your source code path of Eigen3.

II. Use the library

The library could then be used by only including the following header file in your source file. Note, all the interfaces are defined in mmath{} namespace. A tiny example is given as follows.

#include <lib_math/lib_math>
#include <vector>

int main()
{
    // Your 2D coordinates
    std::vector<Eigen::Vector2f> pts = {...}; 
    // Fit GaussianCurve
    mmath::GaussianCurve gauss = mmath::fitGuassianCurve(pts);
    // Show results
    printf("Gaussian paramters: a:%f, mu:%f, sigma:%f\n",
        gauss.a, gauss.mu, gauss.sigma);

    return 0;
}

The specific application examples can be found in lib_math/examples folder. More detailed functions about this library is suggested to referring Intergfaces Mannul.

NOTE: The precision for calculation of kinemaices related functions is change to float, which is generally enough for most of application. To enable double precision, define LIB_MATH_USE_DOUBLE before include this library, as follows.

#define LIB_MATH_USE_DOUBLE
#include <lib_math/lib_math>

Or, a more straight forward way is to define LIB_MATH_USE_DOUBLE when compile the project, as follws.

...
cmake -DLIB_MATH_USE_DOUBLE=ON ..
...

In addition, the Eigen library is very sensitive to precision (such as Eigen::Matrixf can not be assigned to Eigen::Matrixd directly). Thus, to avoid compiling errors in your projects, the arguments type for the functions in this library, kinemaices related functions is suggested to use mmath::kfloat.

Interfaces Manual

Download this repository and see the interfaces manual by opening ./doc/lib_math_doc.html via your browser.