UMAT beginning

Levenberg Marquardt算法的matlab代码,简洁无误,结构清晰,便于快速学习。( LevenbergMarquardt algorithm matlab code, simple and correct, clear structure, easy to learn quickly.)

利用emd算法解决经验模式分解的一些存在的端点效应问题。 有效的解决了问题。(Using emd algorithm to solve some of the empirical mode decomposition end effects exist. Effective solution to the problem.)

针对传统的算法如遗传算法、粒子群算法等在TSP问题上求解精确性和求解规模上都还有一定的不足，本文提出了一种基于动态规划思想的粒子群优化算法。该算法用动态规划的方法实现粒子间的信息交互和粒子的进化，并且将粒子群中的粒子按无标度信息指导网络拓扑图的方式进行连接。仿真结果表明该方法能有效地减小误差率，提高解的精确，同时还保持了较低的计算复杂度，具有良好的稳健性。(TSP problem solving for the traditional algorithms such as genetic algorithms, particle swarm optimization accuracy and solving the scale, there is a certain lack of a dynamic programming algorithm-based particle swarm optimization algorithm. The algorithm using dynamic programming method to achieve the evolution of information exchange between the particles and the particles, and particle swarm particle scale-free information guide network topology map to connect. The simulation results show that this method can effectively reduce the error rate and improve the solution precision, while maintaining a low computational complexity, good soundness.)

算法优化非常好，几乎没有循环，Relief计算分类权重，包括四元数的各种计算。( Algorithm optimization is very good, almost no circulation, Relief computing classification weight, Including quaternion various calculations.)

基于Laguerre多项式的拟合程序，可以控制多项式的结束，以及变量alpha。(Based on Laguerre polynomial fitting procedure, you can control the end of polynomials, and variable alpha.)

Hansen最新（2008）修订后的关于反演正则化方法的电子书与相应Matlab程序(Hansen the latest (2008) as amended on the inversion method is the e-book with the corresponding Matlab program)

This directory contains code implementing the K-means algorithm. Source code may be found in KMEANS.CPP. Sample data isfound in KM2.DAT. The KMEANS program accepts input consisting of vectors and calculates the given number of cluster centers using the K-means algorithm. Output is directed to the screen. Usage for KMEANS is: KMEANS SOURCEFILE <enter> The format of the source file is: NPat - Number of patterns (int) SizeVect - Size of vector (int) NClust - Number of cluster centers(int) vect[1,1] ... vect[1,SizeVect] - vector 1 (real) vect[2,1] ... vect[2,SizeVect] - vector 2 (real) . . . . . . vect[NPat,1] ... vect[NClust,SizeVector] - vector N (real) To compile: ICC KMEANS.CPP <enter> (This directory contains code implementing the K-means algorithm. Source code may be found in KMEANS.CPP. Sample data isfound in KM2.DAT. The KMEANS program accepts input consisting of vectors and calculates the given number of cluster centers using the K-means algorithm. Output is directed to the screen. Usage for KMEANS is: KMEANS SOURCEFILE <enter> The format of the source file is: NPat - Number of patterns (int) SizeVect - Size of vector (int) NClust - Number of cluster centers(int) vect[1,1] ... vect[1,SizeVect] - vector 1 (real) vect[2,1] ... vect[2,SizeVect] - vector 2 (real) . . . . . . vect[NPat,1] ... vect[NClust,SizeVector]- vector N (real) To compile: ICC KMEANS.CPP <enter> )

说明：  从别的网上下来的，是一个简单又经典的分子动力学模拟程序，很适合初级使用。(for molecular simulation)

利用C语言实现MOORE-PENROSE广义逆矩阵的计算，经试验OK。(The realization of MOORE-PENROSE generalized inverse matrix is calculated by using the C language, through the test of OK.)

惩罚函数法C语言相关程序 它将有约束最优化问题转化为求解无约束最优化问题：其中M为足够大的正数， 起"惩罚"作用， 称之为罚因子，F(x, M )称为罚函数。(Penalty Function Method C language-related program it would have constrained optimization problem is transformed into solving unconstrained optimization problem: where M is a sufficiently large positive number, from the " punishment" effect, called the penalty factor, F (x, M) called the penalty function.)