这个问价夹里包含了压缩感知领域中炙手可热的GAMP仿真代码，内同丰富所以在此推荐给大家！

Pade逼近概论--学习数值逼近的人必学的一本书，本书经典、权威

是一个最佳逼近算法的编程，程序运行出来 结果精确度挺高的，大家可以看看！

该压缩包内含有多个文件,其中Approximation.m是住程序文件,只需将该文件放入相应路径内,以调用函数的形式便可以使用.程序内有注释.

多个自变量输入的非线性函数、逼近误差<5%

里面的网址现在找不到了，但是资源又删不了，请大家不要下载了。 一个日本人的3D曲面逼近相关的科研成果，包括原始数据、代码及其发表的文章。内有详细列表及网址链接，有兴趣的可以下载看看。附其中一篇文章的题目及摘要： Sparse Low-degree Implicit Surfaces with Applications to High Quality Rendering, Feature Extraction, and Smoothing Abstract We propose a new surface representation delivering an accurate approximation to a set of points scattered over a smooth surface by Sparse Low-degree IMplicits (SLIM). The SLIM surface representation consists of a sparse multi-scale set of nonconforming surface primitives which are blended along view rays during the rendering phase. This new representation leads to an interactive real-time visualization of large-size models and delivers a better rendering quality than standard splatting techniques based on linear primitives. Further, SLIM allows us to achieve a fast and accurate estimation of surface curvature and curvature derivatives and, therefore, is very suitable for many non-photorealistic rendering tasks. Applications to ray-tracing and surface smoothing are also considered.

chebyshev三次最佳逼近代码 clear clc format long a=0;b=1;n=3; for j=0:n x(j+1)=0.5*(a+b+(b-a)*cos((j+0.5)*pi/(n+1))); end for i=1:n+1 c(i)=exp(x(i)); end…………

很好的帕德逼近程序，用于拟合计算，数值逼近，个人感觉很实用。

这是矩阵的广义低秩逼近的matlab源代码。

基于BP神经网络的函数逼近（不使用任何工具函数）