通过模型决定系数完成支持向量机参数优化 SVR_fitting

本算法是基于数值计算中的SQP编写的MATLAB程序。非常实用的一个小程序。(This algorithm is based on the numerical calculation of SQP written MATLAB program. Very practical one small program.)

C#编写的三次样条插值函数，TSS法求解对角矩阵方程函数，插值点函数值计算函数。(prepared by the cubic spline interpolation function, TSS method diagonal matrix equation function, interpolation function point value function.)

这是一个用于一维等离子体粒子模拟的Fortran程序。(This is a one-dimensional plasma particle simulation for Fortran programs.)

一维热传导方程的求解，包括显式和隐式，及其对比(One-dimensional heat conduction equations, including explicit and implicit, and the contrast)

能够实现二维边界造波，且自带消波，能够较好地模拟波浪荷载。(with a prgram of eliminating wave,it can be used to achieve boundry wave.)

大气参数计算，含密度、湿度、粘性、热导率等(-Air parameters, density, humidity, viscosity, thermal conductivity)

变系数/非线性微分方程的求解，包括原式及改变的式子。 (Variable coefficients/non-linear differential equations to solve, including the original style and formula changes.)

求解最短路算法，，， 赋路径初值 更新 显示迭代步数 显示每步迭代后(solve the lest distance)

工业汽轮机设计计算程序，具有很好的参考价值，可以参考学习一下(Industrial steam turbine design calculation procedures, a good reference value, you can refer to learn about)

说明：  共轭梯度法（Conjugate Gradient）是介于最速下降法与牛顿法之间的一个方法，它仅需利用一阶导数信息，但克服了最速下降法收敛慢的缺点，又避免了牛顿法需要存储和计算Hesse矩阵并求逆的缺点，共轭梯度法不仅是解决大型线性方程组最有用的方法之一，也是解大型非线性最优化最有效的算法之一。 在各种优化算法中，共轭梯度法是非常重要的一种。其优点是所需存储量小，具有步收敛性，稳定性高，而且不需要任何外来参数。(Conjugate gradient method Gradient) is a method between the steepest descent method and Newton's method. It only uses the first derivative information, but overcomes the disadvantage of slow convergence of steepest descent method, and avoids the disadvantage of storing and calculating Hesse matrix and solving inverse of Newton's method. Conjugate gradient method is not only one of the most useful methods to solve large-scale linear equations, but also the most effective method to solve large-scale nonlinear optimization One of the algorithms of. Among all kinds of optimization algorithms, conjugate gradient method is very important. It has the advantages of small storage, step convergence, high stability and no external parameters.)