B_ADI

计算按频率抽取的基4快速傅里叶变换算法，与基2FFT相比，基4FFT的乘法量约减少25 。(Calculating frequency taken by fast Fourier transform algorithm base 4, compared with the base 2FFT the base 4FFT multiplication about 25 less.)

加权总体最小二乘matlab工具箱，包含各种总体最小二乘算法的源代码(weighted total least squares toolbox)

通过卷积神经网络回归进行的图像配准，相比传统的图像配准算法，能更快地获得配准参数，也能更好地减小配准误差(Image registration based on convolution neural network regression can get registration parameters faster and reduce registration error better than traditional image registration algorithm.)

由美国SANDIA实验室一名研究员编写的用于计算半导体量子纳米线的一系列matlab文件。(SANDIA laboratory by the American researcher prepared a semiconductor quantum nanowire used to calculate a series of matlab file.)

说明：  模拟电场中的等势线，利用计算机数学软件MATLAB模拟仿真真空中自由电荷，电偶极子，带电细棒，尖端导体等附近静电场电场线及等势线分布图型(Equipotential line in simulated electric field)

焊接模拟，fluent，电弧udf，钨极氩弧焊，源项(Welding simulation, fluent, arc udf, TIG welding, the source term)

说明：  算法知识讲解，是一本不错的算法入门级教材的pdf版本(This is a good introduction to algorithm)

全局最优化方法的程序，是段老师团队原版程序，大家放心使用。(Global optimization method of procedure is the original section of the teacher team program, we ease of use.)

概念格的构造问题是形式概念分析应用的前提。由于概念格的时空复杂度随着形式背景的增大而可能指数性的增大，有关概念格的生成问题一直是形式概念分析应用研究的一个重点。国内外的学者和研究人员对此进行了深入的研究，提出了一些有效的算法来生成概念格，这些算法一般可分为两类：批生成算法(Batch Algorithm)和渐进式生成算法(Incremental Algorithm )。。 (Concept lattice structure problem is the premise formal concept analysis applications. Since the concept of space-time lattice with the increasing complexity of the form and context may increase exponentially, generating questions about the concept lattice has been a focus of formal concept analysis applied research. Domestic and foreign scholars and researchers conducted a thorough study made a number of efficient algorithms to generate the concept lattice, these algorithms can be divided into two categories: batch generation algorithm (Batch Algorithm) and progressive generation algorithm (Incremental Algorithm ). .)

最小二乘法拟合形如f(x)=(a+bx)/(1+cx)的函数（9.2.2(1))，并解决人口预测问题(9.2.2(2))(Least squares fit of the form f (x) = (a+bx)/(1+cx) function (9.2.2 (1)), and to solve the population prediction problem (9.2.2 (2)))