• MRPT学习笔记Matrices and Vectors


    头文件:

    image

    Matrices

    Matrices are implemented as class templates in MRPT, but the followingtwo types are provided for making programs more readable:


    typedef CMatrixTemplateNumeric<f loat> CMatrixFloat ;
    typedef CMatrixTemplateNumeric<double> CMatrixDouble ;

    A matrix with any given size can be created by passing it at construction time, or otherwise it can be resized later as shown in this example:


    CMatrixDouble M( 2 , 3 ) ; // Create a 2x3 matrix
    cout << M( 0 , 0 ) << endl ; // Pr int out t he l e f t −top element
    CMatrixDouble A; // Another way of c r e a t i ng
    A. s e t S i z e ( 3 , 4 ) ; // a 2x3 matrix
    A( 2 , 3 ) = 1 . 0 ; // Change t he bottom−r i g h t element


    A matrix can be resized at any time, and the contents are preserved ifpossible. Notice also in the example how the element at the r’th row and c’th column can be accessed through M(r, c). Sometimes, predefined values must be loaded into a matrix, and writing all the assignments element by element can be tedious and error prone. In those cases, better use this constructor:


    const double numbers [ ] = {
    1 , 2 , 3 ,
    4 ,5 ,6 } ;
    CMatrixDouble N(2 , 3 , numbers ) ;
    cout << ” I n i t i a l i z e d matr ix : ” << endl << N << endl ;

    固定大小矩阵:

    image

    文件读写:

    CMatrixDouble H,Z ,D;
    H. loadFromTextFi le ( ”H. txt ” ) ; // H <− ’H. t x t ’
    H. e i genVe c tor s (Z ,D) ; // Z: e i g env e c t or s , D: e i g e nv a l u e s
    Z . saveToTextFi le ( ”Z . txt ” ) ; // Save Z in ’Z. t x t ’

    Vectors

    The base class for vectors is the standard STL container std::vector,

    typedef s td : : vector<f loat> v e c t o r f l o a t ;
    typedef s td : : vector<double> ve c tor doubl e ;

    Resize

    ve c tor doubl e V( 5 , 1 ) ; // Create a v e c t or wi th 5 ones .
    V. r e s i z e ( 1 0 ) ;
    cout << V << endl ; // Pr int out t he v e c t or to c onsol e

    文件读写:

    ve c tor doubl e v ;
    loadVector ( CFi leInputStream ( ” in . txt ” ) , v )

    ve c tor doubl e v ( 4 , 0 ) ; // [ 0 0 0 0]
    vectorToTextFi l e ( v , ”o1 . txt ” ) ; // Save as row
    vectorToTextFi l e ( v , ”o2 . txt ” , true ) ; // Append a new row
    vectorToTextFi l e ( v , ”o3 . txt ” , fal se , true ) ; // Save as a column

    Serializing
    If you prefer to serialize the vectors in binary form (see chapter 10), that can be done as simply as:


    ve c tor doubl e v = l i n s p a c e ( 0 , 1 , 1 0 0 ) ; // [ 0 . . . 1]
    CFi leOutputStream( ”dump. bin ” ) << v ;

    基本矩阵(向量)运算

    image

    image

    一些优化的矩阵(向量)运算

    image

    提取子阵

    image

    image

    矩阵合并

    image


    注意:在使用CMatrixFixedNumeric时候编译可通过,但会出现Debug未知错误,原因是CMatrixFixedNumeric.h这个头文件文件名小写了,改成首字母大写格式即可。

  • 相关阅读:
    ZOJ 3349 Special Subsequence
    ZOJ 3684 Destroy
    ZOJ 3820 Building Fire Stations
    HDU 5291 Candy Distribution
    HDU 3639 Hawk-and-Chicken
    HDU 4780 Candy Factory
    HDU 4276 The Ghost Blows Light
    ZOJ 3556 How Many Sets I
    技术人员的眼界问题
    神经网络和深度学习
  • 原文地址:https://www.cnblogs.com/feisky/p/1609761.html
Copyright © 2020-2023  润新知