练习代码：
/**
* @file main.cpp
* @author your name (you@domain.com)
* @brief 学习eigen: 数据类型的基本操作
* @version 0.1
* @date 2021-10-12
*
* @copyright Copyright (c) 2021
*
*/
#include <iostream>
#include "Eigen/Dense" //稠密矩阵相关定义
void PrintTestPlatform()
{
std::cout << "------------------------test platform info----------------------------"<<std::endl;
std::cout << "Eigen version : "<< EIGEN_WORLD_VERSION <<"."<< EIGEN_MAJOR_VERSION << "."<< EIGEN_MINOR_VERSION << std::endl;
std::string system_name =
#if defined(__Apple__)
"Mac Os"
#elif defined(__linux__)
"Linux"
#else
"Windows";
#endif
std::cout<< "System : "<< system_name << std::endl;
std::string compiler =
#if defined(__clang__)
"clang " + ver_string(__clang_major__, __clang_minor__, __clang_patchlevel__);
#elif defined(__GNUC__)
"gcc " + ver_string(__GNUC__, __GNUC_MINOR__, __GNUC_PATCHLEVEL__);
#else
"msvc " + std::to_string(_MSC_VER);
#endif
std::cout<< "Compiler : " << compiler << std::endl;
std::cout << "------------------------test platform info----------------------------"<<std::endl;
}
int main(int argc, char** argv)
{
PrintTestPlatform();
//-----------------------构造函数、初始化--------------------------
{
//数组列向量形式vector
/*
[1]
[2]
...
*/
//列向量由Matrix固定列为1，定义行数所得，如
//typedef Matrix<float, 2, 1> Vector2f;
Eigen::Vector2f v2f(1.1f, 2.2f); //构造函数小括号初始化
Eigen::Vector2d v2d{3.0, 5.0}; //c++11统一初始化格式大括号初始化
Eigen::Vector2i v2i = {1, 3}; //大括号赋值初始化
std::cout<< "Vector2f: \n"<<v2f << "\n"
<< "Vector2d: \n"<<v2d << "\n"
<< "Vector2i: \n"<<v2i << "\n";
Eigen::Vector3i v3i; //未初始化则为随机数值，Eigen不会默认初始化
std::cout<<" Vector3i before initialize: \n" << v3i << "\n";
v3i << 1, 2, 3; //<<运算符初始化
Eigen::VectorXi vxi(5); //未初始化则为随机数值，Eigen不会默认初始化
std::cout<<" VectorXi before initialize: \n" << vxi << "\n";
vxi <<1, 2, 3, 4, 5;
std::cout<< "Vector3i: \n"<<v3i <<"\n"
<< "VectorXi: \n"<<vxi << "\n";
Eigen::Vector3f v3fz = Eigen::Vector3f::Zero(); //初始化为0
std::cout<< "Eigen::Vector3f::Zero(): "<< v3fz << "\n";
//数组行向量形式
// [1] [2]
Eigen::RowVector2f rv2f(2.0f, 4.0f);//typedef Matrix<float, 2, 1> Vector2f;
Eigen::RowVector2d rv2d{1.0, 3.0};
Eigen::RowVector2i rv2i = {1, 2};
Eigen::RowVectorXi rvxi(5);
rvxi << 1, 2, 3, 4, 5;
std::cout<< "RowVector2f: "<<rv2f << "\n"
<< "RowVector2d: "<<rv2d << "\n"
<< "RowVector2i: "<<rv2i << "\n"
<< "RowVectorXi: "<<rvxi << "\n";
//稠密矩阵
//Eigen::Matrix2i m2i(1, 2, 3, 4); 错误，无此构造函数
//Eigen::Matrix2i m2i = {1, 2, 3, 4}; 错误，无此初始化方法
//Eigen::Matrix2i m2i{1, 2, 3, 4}; 错误，无此初始化方法
Eigen::Matrix2i m2i; //<< 运算符初始化
std::cout<< "Matrix2i before initialize: \n" <<m2i <<"\n";
m2i << 1, 2, 3, 4;
//1 2
//3 4
std::cout<< "Matrix2i: \n" <<m2i <<"\n";
Eigen::Matrix3i m3i;
std::cout<< "Matrix3i before initialize: \n" <<m3i <<"\n";
m3i << 1, 2, 3,
4, 5, 6,
7, 8, 9;
std::cout<< "Matrix3i: \n" <<m3i <<"\n";
Eigen::MatrixXi mat_xi(2, 4); //构造2x4静态矩阵
std::cout<< "MatrixXi before initialize: \n" <<mat_xi <<"\n";
mat_xi << 1, 2, 3, 4,
5, 6, 7, 8;
std::cout<< "MatrixXi: \n" <<mat_xi <<"\n";
//初始化为0
Eigen::Matrix3i m3iz = Eigen::Matrix3i::Zero();
std::cout<<"Eigen::Matrix3i::Zero():\n"<<m3iz << "\n";
//获取最大静态矩阵尺寸
//int max_index = (std::size_t(1) << (8 * sizeof(int) - 1)) - 1;
//size_t max_size = std::sqrt(max_index) - 1;
//std::cout<<"max size: "<< max_size << std::endl;
//Eigen::MatrixXi max_mat(max_size, max_size);
//单位矩阵
Eigen::Matrix4f mat_identity = Eigen::Matrix4f::Identity();
std::cout<<"Eigen::Matrix4f::Identity(): \n" << mat_identity << std::endl;
}
//------------------------赋值、取值---------------------------------
{
const size_t size = 100;
Eigen::MatrixXi mat(size, size);
//单个元素取值、赋值，矩阵索引从左上角(0, 0)开始
mat(0, 0) = 1;
std::cout << "mat(0,0): "<< mat(0, 0) << std::endl;
//set方法赋值
Eigen::Matrix4i mat4i = Eigen::Matrix4i::Identity();
std::cout<< "mat4i before set ones:\n" << mat4i << std::endl;
//全部元素置为1
mat4i.setOnes();
std::cout<< "mat4i after set ones:\n" << mat4i << std::endl;
//全部元素置为0
mat4i.setZero();
std::cout<< "mat4i after set zeros:\n" << mat4i << std::endl;
//随机数
mat4i.setRandom();
std::cout<< "mat4i after set random:\n" << mat4i << std::endl;
//区域块取值、赋值
//从位置（10,10)起，往右下角方向取6x8区域块元素
std::cout << mat.block(10, 10, 6, 8) << std::endl;
Eigen::MatrixXi tmp_mat(6, 8);
tmp_mat.setOnes();
//现有matrix赋值给区块
mat.block(10, 10, 6, 8) = tmp_mat;
std::cout << mat.block(10, 10, 6, 8) << std::endl;
//直接操作设置区块
mat.block(10, 10, 6, 8).setZero();
std::cout << mat.block(10, 10, 6, 8) << std::endl;
//左上角为起点
mat.topLeftCorner(4, 4) = Eigen::Matrix4i::Zero();
std::cout<< "mat.topLeftCorner(4, 4):\n"<< mat.topLeftCorner(4, 4) << std::endl;
//右上角为起点
mat.topRightCorner(4, 4) = Eigen::Matrix4i::Zero();
std::cout<< " mat.topRightCorner(4, 4):\n" << mat.topRightCorner(4, 4) << std::endl;
//左下角为起点
mat.bottomLeftCorner(4, 4) = Eigen::Matrix4i::Ones();
std::cout<<" mat.bottomLeftCorner(4, 4):\n"<< mat.bottomLeftCorner(4, 4) << std::endl;
//右下角为起点
mat.bottomRightCorner(4, 4) = Eigen::Matrix4i::Ones();
//右下角为起点
std::cout<<" mat.bottomRightCorner(4, 4)\n"<< mat.bottomRightCorner(4, 4)<< std::endl;
}
//-----------------------基本运算-----------------------------------
{
//数乘
Eigen::Vector3f v3f = Eigen::Vector3f::Ones();
auto v2 = 3.0f * v3f;
std::cout<<v2 << std::endl;
Eigen::Matrix4i mat4i = Eigen::Matrix4i::Ones();
auto m1 = 5 * mat4i; //或者 auto r1 = mat4i * 5;
std::cout << m1 << std::endl;
//加减乘
Eigen::Matrix3i m3i_1 = Eigen::Matrix3i::Ones();
Eigen::Matrix3i m3i_2 = Eigen::Matrix3i::Identity();
auto add = m3i_1 + m3i_2;
std::cout<< "add:\n" << add << std::endl;
auto sub = m3i_1 - m3i_2;
std::cout<< "sub:\n" << sub << std::endl;
auto mul = m3i_1 * m3i_2;
std::cout<< "mul:\n" << mul << std::endl;
//求逆
//Eigen::Matrix3i m3i; 非浮点类型数据不可逆，编译不通过
Eigen::Matrix3f m3f;
m3f << 1, 0, 2,
2, 3, 0,
0, 2, 4;
std::cout << "inv:\n"<<m3f.inverse() << std::endl;
//转置
std::cout<< m3f.transpose() << std::endl;
//元素和
std::cout<< m3i_2.sum() << std::endl;
//元素均值
std::cout<< m3f.mean() << std::endl;
//行列式
std::cout<< m3i_1.determinant() << std::endl;
}
return 0;
}运行结果
------------------------test platform info----------------------------
Eigen version : 3.3.9
System : Windows
Compiler : msvc 1929
------------------------test platform info----------------------------
Vector2f:
1.1
2.2
Vector2d:
3
5
Vector2i:
1
3
Vector3i before initialize:
-858993460
-858993460
-858993460
VectorXi before initialize:
-842150451
-842150451
-842150451
-842150451
-842150451
Vector3i:
1
2
3
VectorXi:
1
2
3
4
5
Eigen::Vector3f::Zero(): 0
0
0
RowVector2f: 2 4
RowVector2d: 1 3
RowVector2i: 1 2
RowVectorXi: 1 2 3 4 5
Matrix2i before initialize:
-858993460 -858993460
-858993460 -858993460
Matrix2i:
1 2
3 4
Matrix3i before initialize:
-858993460 -858993460 -858993460
-858993460 -858993460 -858993460
-858993460 -858993460 -858993460
Matrix3i:
1 2 3
4 5 6
7 8 9
MatrixXi before initialize:
-842150451 -842150451 -842150451 -842150451
-842150451 -842150451 -842150451 -842150451
MatrixXi:
1 2 3 4
5 6 7 8
Eigen::Matrix3i::Zero():
0 0 0
0 0 0
0 0 0
Eigen::Matrix4f::Identity():
1 0 0 0
0 1 0 0
0 0 1 0
0 0 0 1
mat(0,0): 1
mat4i before set ones:
1 0 0 0
0 1 0 0
0 0 1 0
0 0 0 1
mat4i after set ones:
1 1 1 1
1 1 1 1
1 1 1 1
1 1 1 1
mat4i after set zeros:
0 0 0 0
0 0 0 0
0 0 0 0
0 0 0 0
mat4i after set random:
-16343 2785 10578 6897
2083 -660 8080 443
-10050 -4906 -10679 -6423
10116 12974 11761 -15893
-842150451 -842150451 -842150451 -842150451 -842150451 -842150451 -842150451 -842150451
-842150451 -842150451 -842150451 -842150451 -842150451 -842150451 -842150451 -842150451
-842150451 -842150451 -842150451 -842150451 -842150451 -842150451 -842150451 -842150451
-842150451 -842150451 -842150451 -842150451 -842150451 -842150451 -842150451 -842150451
-842150451 -842150451 -842150451 -842150451 -842150451 -842150451 -842150451 -842150451
-842150451 -842150451 -842150451 -842150451 -842150451 -842150451 -842150451 -842150451
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
mat.topLeftCorner(4, 4):
0 0 0 0
0 0 0 0
0 0 0 0
0 0 0 0
mat.topRightCorner(4, 4):
0 0 0 0
0 0 0 0
0 0 0 0
0 0 0 0
mat.bottomLeftCorner(4, 4):
1 1 1 1
1 1 1 1
1 1 1 1
1 1 1 1
mat.bottomRightCorner(4, 4)
1 1 1 1
1 1 1 1
1 1 1 1
1 1 1 1
3
3
3
5 5 5 5
5 5 5 5
5 5 5 5
5 5 5 5
add:
2 1 1
1 2 1
1 1 2
sub:
0 1 1
1 0 1
1 1 0
mul:
1 1 1
1 1 1
1 1 1
inv:
0.6 0.2 -0.3
-0.4 0.2 0.2
0.2 -0.1 0.15
1 2 0
0 3 2
2 0 4
3
1.55556
0本文由芒果浩明发布，转载请注明出处。
本文链接：https://mangoroom.cn/eigen/data-type-manipulation-in-eigen.html
