Description: 用MATLAB求解无约束的问题，主要有最速下降法，牛顿法，共轭梯度法，变尺度法（DFP和BFGS法），非线性最小二乘法。
用MATLAB求解有约束的问题，主要是外惩罚函数和广义乘子法。
以及一些对具体问题的分析，MATLAB的代码在文档里都有。
-Using MATLAB to solve the problem of non-binding, there are the steepest descent method, Newton method, conjugate gradient method, variable metric method (DFP and BFGS method), nonlinear least square method. Using MATLAB to solve the problem of binding, mainly outside the penalty function method and generalized multipliers. As well as some specific issues for analysis, MATLAB code in the document, are limitless. Platform: |
Size: 81920 |
Author:ljw |
Hits:
Description: 各种最小二乘算法总结（算例及Matlab程序）
1、一般最小二乘算法的一次计算法和递推算法
2、遗忘因子最小二乘的一次算法和递推算法
3、限定记忆最小二乘递推算法
4、偏差补偿最小二乘算法
5、增广最小二乘
6、广义最小二乘
7、辅助变量法
8、二步法
9、多级最小二乘
10、Yule-Walker辨识算法
（参考文献：过程辨识，方崇智，清华大学出版社，1988）-Summary of various least-squares algorithm (Matlab examples and procedures) a general least-squares algorithm is a calculation method and the recursive algorithm 2, a forgetting factor least squares algorithm and the recursive algorithm 3, limited memory least-squares recursive algorithm 4, a deviation of penalized least squares algorithms 5, 6 Extended Least Squares, Generalized Least Squares 7, supplementary variable method 8, 9, 2-step, multi-stage least-squares 10, Yule-Walker Identification Algorithm (see literature: the process of identification, Fang Chongzhi, Tsinghua University Press, 1988) Platform: |
Size: 1110016 |
Author: |
Hits:
Description: 为了更有效地提取图像的局部特征，提出了一种基于2维偏最小二乘法(two—dimensional partial least
square，2DPLS)的图像局部特征提取方法，并将其应用于面部表情识别中。该方法首先利用局部二元模式(1ocal
binary pattern，LBP)算子提取一幅图像中所有子块的纹理特征，并将其组合成局部纹理特征矩阵。由于样本图像
被转化为局部纹理特征矩阵，因此可将传统PLS方法推广为2DPLS方法，用来提取其中的判别信息。2DPLS方法
通过对类成员关系矩阵的构造进行相应的修改，使其适应样本的矩阵形式，并能体现出人脸局部信息重要性的差
异。同时，对于类成员关系协方差矩阵的奇异性问题，也推导出了其广义逆的解析解。基于JAFFE人脸表情库的
实验结果表明，该方法不但可以有效地提取图像局部特征，并能取得良好的表情识别效果。-To better the image of the local feature extraction, a partial least squares method based on 2D (two-dimensional partial least
square, 2DPLS) image local feature extraction method, and applied to facial expression recognition. In this method, use of local binary pattern (1ocal
binary pattern, LBP) operator extracts an image texture features of all sub-blocks, and their combination into the local texture feature matrix. As the sample image
Be translated into the local texture feature matrix, so the traditional PLS method can be generalized to 2DPLS method used to extract the identification information. 2DPLS method
Through the class membership matrix in the corresponding modifications to adapt the sample matrix, and can reflect the importance of face poor local information
Different. Meanwhile, members of the class covariance matrix of the singular relations issues, also derived the generalized inverse of the analytical solution. Based on the JAFFE facial expression database
Platform: |
Size: 315392 |
Author:MJ |
Hits:
Description: 此程序为用matlab编写的基于最小二乘法通过求解广义矩阵特征值及特征向量进行椭圆曲线拟合程序。
程序中x,y分别为采样点的横纵坐标。生成的结果矩阵为椭圆曲线aX^2+bXY+cY^2+dX+eY+f=0中的各项系数。-This program was written with matlab by solving the generalized least square method based on matrix values and eigen vectors of the elliptic curve fitting procedure. Process x, y, respectively transverse and longitudinal coordinates of sampling points. The results generated matrix of elliptic curve aX ^ 2+ bXY+ cY ^ 2+ dX+ eY+ f = 0 in the coefficients. Platform: |
Size: 1024 |
Author:kone_zhang |
Hits:
Description: 频域不同导频密度的信道估计在被研究，并且在高低时延信道其信道估计被固定了导频分配时域。频域信道估计是基于插补，多项式的广义线性模型（或最小二乘法）和维纳滤波（或线性最小均方误差法）的方法。时域的信道估计是用线性插补完成的。我们通过测量在基于克拉克散射模型的瑞利衰落信道上并带有QPSK调制的粗误码率，已经比较了所有这些方案的结果。-Different pilot density in frequency domain channel estimate to research, and its high and low delay channel estimation channel was fixed allocation of time-domain pilot. Frequency domain channel estimation is based on interpolation, polynomial generalized linear model (or least squares) and the Wiener filter (or linear minimum mean square error method) method. Time domain channel estimation is done using linear interpolation. We are by measuring the scattering model based on Clark and Rayleigh fading channels with QPSK modulation bit error rate of crude, has compared the results of all these programs. Platform: |
Size: 2505728 |
Author:雨儿 |
Hits:
Description: 此程序是基于广义最小二乘法，运用递推算法。-This procedure is based on generalized least square method, the use of recursive algorithms. Platform: |
Size: 1024 |
Author:侯宁 |
Hits:
Description: 地心地固（ECEF）坐标系下，基于广义最小二乘的雷达组网空间误差配准-Radar network based on generalized least square error of registration space Platform: |
Size: 2048 |
Author:梁苑 |
Hits:
Description: 频域SCOT自适应时延估计，实质上是对广义功率频函数平方得到的数的相位谱做最小二乘拟合，求得时延估计。-Frequency domain SCOT adaptive time delay estimation, is essentially the generalized power-frequency phase of the function of the square of the number spectrum of the least squares fitting obtained delay estimates. Platform: |
Size: 4096 |
Author:puLei |
Hits:
Description: 适当选择辅助变量，使之满足相应条件，参数估计值就可以是无偏一致。估计辅助变量法的计算量与最小二乘法相当，但辨识效果却比最小二乘法好的多。尤其当噪声是有色的，而噪声的模型结构又不好确定时，增广最小二乘法和广义最小二乘法一般都不好直接应用，因为他们需要选用特定的模型结构，而辅助变量法不需要确定噪声的模型结构，因此辅助变量法就显得更为灵活，但辅助变量法不能同时获得噪声模型的参数估计。-Choose appropriate secondary variables, meet the relevant conditions and parameter estimate can be unbiased consistent. Estimated auxiliary variable method calculation and least square method is quite, but the identification effect is much better than the least square method. Especially when the noise is colored, and noise model structure and not sure, augmented the least squares and the generalized least squares method is generally not used directly, because they need to choose specific model structure, and auxiliary variable method does not need to make sure that noise model structure, so the auxiliary variable method is more agile, but not at the same time auxiliary variable method for noise model parameter estimation. Platform: |
Size: 1024 |
Author:张鹏 |
Hits:
Description: 自己编写的数值逼近程序。包含三个文件，分别为三角拟合，最小二乘拟合，广义延拓逼近。-I have written a numerical approximation procedure. Contains three files, were triangular fitting, least square fitting, the generalized extended approximation Platform: |
Size: 2048 |
Author:单边 |
Hits:
Description: 用镜像矩阵求出矩阵的QR分解（正交三角分解），并求解线性方程组，该方法不必选主元，但其计算过程非常稳定。也可用于求矩阵的广义逆和求解线性最小二乘问题。子过程QRDCMP用镜像矩阵求m*n矩阵A的QR分解，A=Q^TR,其中Q是m*m正交矩阵，R为m*n上三角矩阵；子过程QRBKSB用矩阵的QR分解求解线性方程组Ax=b，其中A为n阶非奇异方阵。-Matrix obtained with the mirror matrix QR decomposition (orthogonal triangular decomposition), and solving linear equations, the method does not have to choose the main element, but its calculation process is very stable. Can also be used to find the generalized inverse matrix and solving linear least squares problems. Sub-process of the mirror matrix QRDCMP m* n matrix A QR decomposition, A = Q ^ TR, wherein Q is a m* m orthogonal matrix, R is m* n upper triangular matrix subprocess QRBKSB a QR decomposition to solve the matrix linear equations Ax = b, where A is non-singular square matrix of order n. Platform: |
Size: 1024 |
Author:changyee |
Hits:
Description: 利用广义逆求解无约束条件下的优化问题。求解线性方程组最小二乘解的广义逆法-Under the condition of using the generalized inverse solving unconstrained optimization problems.The least-square solutions to the linear system of equations of the generalized inverse method Platform: |
Size: 2048 |
Author:caoxiaoyue |
Hits:
Search - generalized least square