[数学计算] libogg
首先对输入音频PCM信号进行时频分析，决定MDCT的长度，即加窗，然后进行MDCT变换；同时对原始音频信号要进行FFT分析。两种变换的频谱系输入给心理声学模型单元，MDCT系数用于噪声掩蔽计算，H可结果用于音调掩蔽特性计算，共同构造总的掩蔽曲线。然后根据MDCT系数及掩蔽曲线，对频谱系数进行线性预测分析用LPC(Linear Prediction Coefficience，线性预测系数)表示频谱包络，即基底曲线(Floor Curve)；或通过线性分段逼近方式获得基底曲线。从MDCT系数中去掉频谱包络则得到白化的残差频谱(Residue)，由于残差频谱波动范围明显变小，从而降低量化误差。之后可以选择是否采用声道耦合(Channel Coupling)技术进一步降低冗余度，耦合主要是将左右声道数据从直角坐标映射到平方极坐标；最后对白化的残差信号有效地以矢量量化表示。最后将要传输的各种信息数据按Vorbis定义的包格式组装，形成Vorbis压缩码流。
First, the input audio PCM signal time-frequency analysis to determine the length of the MDCT, that windowing, then MDCT transform while the original audio signal to FFT analysis. Two Transformations spectral line input to the psychoacoustic model unit, MDCT coefficients used to calculate the noise masking, H can be used to tone masking characteristic calculation result, the overall structure of the common masking curve. Then according to the MDCT coefficient and the masking curve, spectrum analysis using linear prediction coefficients LPC (Linear Prediction Coefficience, linear prediction coefficients) represent the spectral envelope, ie the base curve (Floor Curve) or approaching the base curve obtained by linear segments. Removed the MDCT coefficients in the spectral envelope of the residual spectrum obtained albino (Residue), since the fluctuation range of the residual spectrum significantly smaller, thereby reducing the quantization error. Then you can choose whether to use cha (2016-06-27, Visual C++, 494KB, 下载5次)
[数学计算] queenList1
八皇后问题最早是由国际西洋棋棋手马克斯·贝瑟尔于1848年提出。之后陆续有数学家对其进行研究，其中包括高斯和康托，并且将其推广为更一般的n皇后摆放问题。八皇后问题的第一个解是在1850年由弗朗兹·诺克给出的。诺克也是首先将问题推广到更一般的n皇后摆放问题的人之一。1874年，S.冈德尔提出了一个通过行列式来求解的方法，这个方法后来又被J.W.L.格莱舍加以改进。
Eight queens problem was first proposed by the International Chess Grandmaster Mark Searle Spey in 1848. After its research mathematician after another, including Gauss and Cantor, and its extension to more general n-queens problem placing. The first eight queens problem solution in 1850 was given by the Franz Roanoke. Knox is also one of the first questions will be extended to the more general problem of placing n queens people. 1874, S. Gandre propose a method to solve through the determinant, this method later was JWL Gelaishejia to improve. (2014-04-10, Visual C++, 1KB, 下载3次)
[数学计算] cPP-Calculator
（1）具备整型数据、浮点型数据的算术（加、减、乘、除）运算功能。依次输入第一个运算数、运算符（+，-，*，/）、第二个运算数，然后输出结果。结果可以作为下一个运算的第一运算数。按‘C’清屏，按‘R’返回 菜单。
例如：输入：2
+
5
输出：7
（2）实现单运算符表达式计算的功能。输入的操作数可以包含整数或浮点数。输入表达式如下：
例如：输入：2+5
输出：7
(1) with integer data, floating point arithmetic (addition, subtraction, multiplication, division) arithmetic functions. Followed by input of the first operand, operator (+,-,*, /), the second operand, and then output the results. The results can be used as the first operand of the next operation. Clear the screen, press the ' C' Press ' R' to return to the menu. Example: Input: 2+ 5 output: 7 (2) single operator expression. Input operands can contain integer or floating-point number. Enter the expression as follows: For example: Input: 2+5 output: (2013-05-19, Visual C++, 452KB, 下载2次)
[数学计算] linear-algebra
李超+线性代数在OI中的应用与题目讲解
线性代数是数学的一个分支，它的研究对象是向量，向量空间（或称线性空间），线性变换和有限维的线性方程组。向量空间是现代数学的一个重要课题；因而，线性代数被广泛地应用于抽象代数和泛函分析中；通过解析几何，线性代数得以被具体表示。线性代数的理论已被泛化为算子理论。由于科学研究中的非线性模型通常可以被近似为线性模型，使得线性代数被广泛地应用于自然科学和社会科学中。
Lie+ linear algebra applications with the subject in the OI explain
Linear algebra is a branch of mathematics, it is a vector, the vector space (or linear space), the linear transformation and the finite-dimensional linear equations. The vector space is an important issue in modern mathematics Thus, linear algebra is widely used in abstract algebra and functional analysis able to be specifically expressed through analytic geometry, linear algebra. The theory of linear algebra has been generalized in operator theory. As scientific research in nonlinear models can often be approximated by linear models, linear algebra is widely used in the natural sciences and social sciences. (2013-03-15, Visual C++, 105KB, 下载2次)
[数学计算] matrix
3、 矩阵（三）
编写C++程序完成以下功能：
（1） 用类来实现矩阵，定义一个矩阵的类，属性包括：
矩阵大小，用 lines, rows（行、列来表示）；
存贮矩阵的数组指针，根据矩阵大小动态申请（new）。
（2） 矩阵类的方法包括：
构造函数，参数是矩阵大小，需要动态申请存贮矩阵的数组；
析构函数，需要释放矩阵的数组指针；
拷贝构造函数，需要申请和复制数组；
输入，可以从cin中输入矩阵元素；
输出，将矩阵格式化输出到cout；
矩阵相加的函数，实现两个矩阵相加的功能,结果保存在另一个矩阵类，但必须矩阵大小相同；
矩阵相减的函数，实现两个矩阵相减的功能,结果保存在另一个矩阵类，但必须矩阵大小相同。
（3） 定义三个矩阵：A1、A2、A3；
（4） 初始化A1、A2；
（5） 计算并输出A3 = A1加A2，A3=A1减A2；
（6） 用new动态创建三个矩阵类的对象：pA1、pA1、pA3；
（7） 初始化pA1、pA2；
（8） 计算并输出pA3=pA1加pA2，pA3=pA1减pA2；
（9） 释放pA1、pA1、pA3。
3, matrix, (c)
Write a C++ program to perform the following functions:
(1) matrix class, define a matrix of class attributes include:
matrix size, lines, rows (rows and columns to represent)
storage matrix array pointer, dynamic application (new) according to the size of the matrix.
(2) matrix class method comprising:
constructor parameter is the size of the matrix, dynamic application memory matrix array
destructor release matrix array pointer
copy constructor, and the need to apply for and copy an array
The input from cin input matrix elements
output, the the matrix formatted output to cout
Function matrix sum, the sum of two matrices, and the results are stored in another matrix class, but must be the same as the size of the matrix
matrix function subtraction, subtract two matrices, and save the results in another matrix class, but it must be the same as the size of the matrix.
(3) d (2012-10-07, Visual C++, 1KB, 下载4次)
[数学计算] matrix
此包包含了众多矩阵处理程序，能够满足矩阵处理的一般要求，由于将各功能分开到不同的“.cpp”文件中，故使用时需要用户自行选取更换合适自己使用的“.cpp”文件。其中，矩阵功能有：输出矩阵、矩阵转置、矩阵归一化、判断矩阵对称、判断矩阵对称正定、全选主元法求矩阵行列式、全选主元高斯(Gauss)消去法求一般矩阵的秩、用全选主元高斯-约当(Gauss-Jordan)消去法计算实(复)矩阵的逆矩阵、用“变量循环重新编号法”法求对称正定矩阵逆、特兰持(Trench)法求托伯利兹(Toeplitz)矩阵逆、实矩阵LU分解、用豪斯荷尔德(Householder)变换对一般m*n阶的实矩阵进行QR分解、对称正定阵的乔里斯基(Cholesky)分解及求其行列式值、一般实矩阵的奇异值分解、广义逆的奇异值分解。
This package contains a number of matrix processing, matrix processing to meet the general requirements, due to the separation of functions to different ". Cpp" file, so the need to use their own user-selected replacement suitable to use ". Cpp" files. Among them, the matrix function: output matrix, the matrix transpose, matrix normalized to determine the matrix symmetric, symmetric positive definite matrix to determine the whole pivoting method for the matrix determinant, full pivoting Gauss (Gauss) elimination method for the general matrix of rank, with full pivoting Gauss- Jordan (Gauss-Jordan) elimination method to calculate the real (complex) inverse of a matrix, with a "variable cycle re-number method" method for the symmetric positive definite matrix inverse, Portland hold (Trench) method for the Tuobo Leeds (Toeplitz) matrix inverse real matrix LU factorization, with high Siheerde (Householder) transformations of the general order of m* n matrix QR decomposition of real symmetr (2012-01-05, Visual C++, 1052KB, 下载36次)
[数学计算] fraction
定义和使用分数类fraction：
1.自定义分数类fraction，使用该类可以完成分数的输入、分数的加、减、乘、除二目运算和一目减运算、分数的约分操作、分数的倒数运算、对两个分数进
行六种比较运算、以及对分数的输出等操作。
2.尽量用运算符重载函数。如，可重载+、-、*、/、>、>=、<、<=、==、!=，以实现分数的各种运算。并且实现分数与整数的混合运算。
Class definition and use of scores fraction:
1 Custom score class fraction, the use of such input can be completed scores, scores of add, subtract, multiply and divide two operations and a project by project operations, scores some points operation, scores of reciprocal operation of the two scores into the
Line six kinds of comparison operators, as well as scores of output and other operations.
(2) the best use of operator overloading function. For example, scores can be overloaded+,-,*,/,>,>=,<,<=,==,!=, to achieve a variety of operations. And to achieve a mixture fraction and integer operations.
(2011-12-17, Visual C++, 2KB, 下载10次)
[数学计算] CPPsimplex
单纯形法，求解线性规划问题的通用方法。单纯形是美国数学家G.B.丹齐克于1947年首先提出来的。它的理论根据是：线性规划问题的可行域是 n维向量空间Rn中的多面凸集，其最优值如果存在必在该凸集的某顶点处达到。顶点所对应的可行解称为基本可行解。单纯形法的基本思想是：先找出一个基本可行解，对它进行鉴别，看是否是最优解；若不是，则按照一定法则转换到另一改进的基本可行解,再鉴别；若仍不是，则再转换,按此重复进行。因基本可行解的个数有限，故经有限次转换必能得出问题的最优解。如果问题无最优解也可用此法判别。
Simplex method for the minimum (2011-09-18, Visual C++, 1531KB, 下载14次)
[数学计算] FFT2D
2dFFT的奈奎斯特频率为 ，（ 为采样间隔）实际上的FFT变换点数要求为2的n次幂，各点所对应的频率是从 到 ，求出 后，可以用 ，这里N为FFT变换的点数。如变换点数为2048，由于对称性，实际有用的点数为1024个。
在该程序中，做完富氏变换后，已经将数据按照频率从0到 到0的顺序排好。
在运行结果中，只需存储左下角的1/4区域的数据即可。
2dFFT of the Nyquist frequency, (for the sampling interval) is actually the FFT, the n number of points required for the power of 2, the frequency corresponding to each point is to, find, you can use the FFT, where N points. Such as the transformation point to 2048, due to the symmetry, the actual useful points to 1024. In the process, done after the Fourier transform has the data in accordance with the frequency range from 0 to 0 in order to line up. In the run results, only the lower left corner of the store 1/4 of regional data can be. (2011-05-23, Visual C++, 513KB, 下载13次)
[数学计算] FFT1D
FFT的奈奎斯特频率为 ，（ 为采样间隔）实际上的FFT变换点数要求为2的n次幂，各点所对应的频率是从 到 ，求出 后，可以用 ，这里N为FFT变换的点数。如变换点数为2048，由于对称性，实际有用的点数为1024个。
在该程序中，做完富氏变换后，已经将数据按照频率从0到 到0的顺序排好。
在运行结果中，只需存储左下角的1/4区域的数据即可。
FFT of the Nyquist frequency, (for the sampling interval) is actually the FFT, the n number of points required for the power of 2, the frequency corresponding to each point is to, find, you can use the FFT, where N points. Such as the transformation point to 2048, due to the symmetry, the actual useful points to 1024. In the process, done after the Fourier transform has the data in accordance with the frequency range from 0 to 0 in order to line up. In the run results, only the lower left corner of the store 1/4 of regional data can be. (2011-05-23, Visual C++, 150KB, 下载7次)
[数学计算] Simplex_method
单纯形法，求解线性规划问题的通用方法。单纯形是美国数学家G.B.丹齐克于1947年首先提出来的。它的理论根据是：线性规划问题的可行域是 n维向量空间Rn中的多面凸集，其最优值如果存在必在该凸集的某顶点处达到。顶点所对应的可行解称为基本可行解。单纯形法的基本思想是：先找出一个基本可行解，对它进行鉴别，看是否是最优解；若不是，则按照一定法则转换到另一改进的基本可行解,再鉴别；若仍不是，则再转换,按此重复进行。因基本可行解的个数有限，故经有限次转换必能得出问题的最优解。如果问题无最优解也可用此法判别。
Simplex method (2010-08-07, Visual C++, 288KB, 下载21次)
[数学计算] blitz-0.9.tar
blitz++库的相应源码
Blitz++提供了一个N维（1—10）的Array类,这个Array类以reference counting技术实现，支持任意的存储序(row-major的C-style数组，column-major的Fortran-style数组)，数组的切割(slicing),子数组的提取(subarray),灵活的Array相关表达式处理。另外提供了可以产生不同分布的随机数(F,Beta,Chi-Square，正态，均匀分布等)的类也是很有特色的。
blitz++ library of the corresponding source Blitz++ provides an N-dimensional (1-10) of the Array class, the Array class with reference counting technology in support of any storage sequence (row-major of the C-style array, column-major in Fortran- style array), an array of cutting (slicing), sub-array extraction (subarray), and flexible expression Array-related processing. Also provided that can produce different distributions of random numbers (F, Beta, Chi-Square, normal, uniform distribution) of the class is very unique. (2010-07-15, Visual C++, 3051KB, 下载73次)
[数学计算] juzhengsuoyouyunxuan
详细说明：矩阵的转置、行列式、秩，逆矩阵求法，矩阵的三角分解、qr分解，对称正定矩阵的乔里斯基分解及行列式值，奇异值分解，广义逆的奇异值分解，矩阵特征值与特征向量的各种计算方法-(all kinds of computational method of transposition of matrix矩阵的转置, determinant行列式, rank秩, inverse of matrix逆矩阵求法,triangle decomposition矩阵的三角分解, qr decomposition qr分解, cholesky decomposition and determinant value based on symmetry positive definite matrix对称正定矩阵的乔里斯基分解及行列式值, decomposition of singular value奇异值分解, decomposition of generalized inversal singular value广义逆的奇异值分解, eigenvalue and eigenvector矩阵特征值与特征向量)
Details: matrix transpose, determinant, rank, inverse matrix, matrix of the triangular decomposition, qr decomposition, Cholesky decomposition of symmetric positive definite matrix and determinant, singular value decomposition, generalized inverse of singular value decomposition, matrix eigenvalue and eigenvector of the various calculation methods- (all kinds of computational method of transposition of matrix matrix transpose, determinant determinant, rank order, inverse of matrix inverse matrix method, triangle decomposition matrix of the triangular decomposition, qr decomposition qr decomposition, cholesky decomposition and determinant value based on symmetry positive definite matrix of the Cholesky decomposition of symmetric positive definite matrix and determinant, decomposition of singular value singular value decomposition, decomposition of generalized inversal singular value generalized inverse of singular value decomposition, eigenvalue and eigenvector Matrix Eigenvalues and Ei (2010-04-27, Visual C++, 84KB, 下载88次)
[数学计算] xx
该程序实现对一维数据轴上的最临近点的求解问题
采用方法：分治方法1（该方法在递归时由于使用数组的值，使得
递归进栈的数据很多，消耗系统空间很大，所以最多
处理的个数经测试不超过70个，可见在设计程序时，考虑
空间的使用也是很必要的！）
The realization of the program of one-dimensional data point near the axis of the most used method of solving the problem: separation of Method 1 (recursive when the method is the use of an array of value, making recursive stack data into a lot of consumption of the system is much room for Therefore, the largest number of processing have been tested not more than 70 shows in the design process, consider the use of space is also very necessary!) (2008-12-15, Visual C++, 2KB, 下载2次)
[数学计算] LE_GaussSeidelIteration
vc++实现线性方程组求解
1全选主元高斯消元法
2全选主元高斯-约当消元法
3三对角方程组的追赶法
4一般带型方程组求解
5对称方程组的分解法
6对称正定方程组的平方根法
7大型稀疏方程组全选主元高斯-约当法
8托伯利兹方程组的列文逊法
9高斯-赛德尔迭代法
10对称正定方程组的共轭梯度法
11线性最小二乘问题的豪斯荷尔德变换法
12线性最小二乘问题的广义逆法
13病态方程组求解
最后注意，在VC++ 6.0中设置好路径，特别是include目录(文件夹)的路径，否则在编译时会出现找不到头文
件的错误，使编译无法正常进行。
err (2008-12-02, Visual C++, 2531KB, 下载164次)
[数学计算] dadizhutijiesuan
白塞尔大地主题解算
包括正算与反算。可由经纬度计算距离，及根据一点经纬度、大地方位角、大地线长度计算另一点经纬度。包括克拉索夫斯基椭球体、1975国际椭球体、WGS-84椭球体，也可以在代码中自己添加椭球体参数计算。
Serge white earth solvers, including the theme and counter-balance is operator. Calculating the distance by latitude and longitude, and latitude and longitude in accordance with that, the earth azimuth, calculate the length of geodesics another point latitude and longitude. Including his克拉索ellipsoid, 1975 International ellipsoid, WGS-84 ellipsoid can also be added in the code itself ellipsoid parameters. (2008-11-29, Visual C++, 52KB, 下载290次)
[数学计算] 5-1
简单有限元计算程序。
NJF---结点自由度数
E---弹性模量
NE---单元数
T---板厚度
VM---泊松比
z---实型二维数组，用来存放结点坐标
ZJ---实型二维数组，用来存放每个约束自由度方向和该方向的约束值，
ie---实型二维数组，存放布尔矩阵
p---实型一维数组，先存放等效结点载荷，解完方程后存放输出的结点位移
a,b,a---实型数组，存放各个单元的面积及由结点坐标计算的bi,bj,ci,cj
g---实型数组,存放单刚子块
zk---实型二维数组，存放整刚矩阵
Simple finite element program. NJF--- the number of nodes degree of freedom E--- elastic modulus NE--- units T--- plate thickness VM--- Poisson (2008-10-14, Visual C++, 1KB, 下载13次)
[数学计算] whitenoise
1．产生白噪声程序
编程产生一组正态分布的白噪声信号,它的均值和方差以及长度可随意调整。将产生的白噪声信号存入数据文件。
本程序算法用C++语言编写。首先用乘同余法产生均匀分布白噪声，再用变换抽样法转换为高斯分布白噪声。算法及程序实现叙述如下。
1) 设定x初值为11,A＝179，长度WNlength,均值Average,方差Serror为用户输入的变量；
2) M =235，ζi= x/M；
3) 取ζi的小数部分再赋值给ζi＋1，这就是均匀分布白噪声；
4) 利用公式 η1=Serror×(–2*logζ1)0.5×cos(2pζ2) ＋Average
η2= Serror×(–2*logζ1)0.5×sin(2pζ2) +Average
计算得到均值和方差可任意调整的白噪声序列。式中 为均匀分布白噪声。
err (2008-09-12, Visual C++, 160KB, 下载166次)
[数学计算] goodjsq
设计一个简单的计算器类（类型），要求：
（1） 从键盘读入算式。
（2） 可以进行加、减、乘、除运算。
（3） 运算要有优先级。
（4） 用户可以按任何的运算符出现顺序进行输入。
（5） 不限定用户输入的计算式的长度。
（6） 有排错功能，当用户输入错误的算式时提示用户。
例如，如果用户输入：3 + 4 * 5 - 7，计算结果应为16。
（提示：可以使用本节所定义的栈类来帮助进行处理）。
Design a simple calculator-type (type), requirements: (1) Reading from the keyboard into the formula. (2) can add, subtract, multiply, with the exception of computing. (3) The operator must have priority. (4) the user can press any of the emergence of the order of operator input. (5) does not limit the user to enter the length of the calculation formula. (6) have debugging function, when the user enters the wrong formula when prompts the user. For example, if a user enters: 3+ 4* 5- 7, the calculation results should be 16. (Hint: You can use this section of the stack type definition to help to deal with). (2007-12-12, Visual C++, 2KB, 下载19次)
[数学计算] 最短路径的求解
本实验中的问题是一个逻辑问题，而实验的目的是运用邻接矩阵解决图存储并解决在此结构情况下对矩阵中元素的扫描计算。其关键是怎样储存和计算。储存用的是邻接矩阵，而计算是用迪杰斯特拉算法。
experiment of the problem is a problem of logic, and The purpose of this experiment is to use the adjacency matrix storage solutions and plans to solve the structure of the matrix elements of scanning calculation. The key is how storage and computation. The store is adjacent matrix, which is calculated using Floyd algorithm. (2005-12-25, Visual C++, 9KB, 下载62次)