卷积神经网络，手写识别训练

线性CCD测试代码+调试教程控制程序,实现线性CCD的采集循迹和处理,可用于飞思卡尔智能车,车模的调试和测试

matlab 例程-Matlab routines

对等间隔数据 ，使用快速傅里叶变换到频率域进行补零， 之后反变换回来实现数据插值的过程。程序调用了fftw库，所以在调试前请先准备fftw库。此算法目前只支持等间隔采样的数据，对于高维数据的插值可以通过，分别对每一个维度插值后最后整合起来，实现高维数据的插值。

计算力学编程,有限元-computational mechanics programming, finite element

二叉树的前序遍历-binary tree before preorder

DSP中输入信号的生成过程。 conio.cpp实现X(n)信号，其中有两个频率分量的正弦信号（正弦计算由sinwn.cpp实现），频率可变，这里取140Hz和70Hz。 考虑了高斯白噪声，由gauss.cpp实现。 最后该信号共产生2000个点，最后的信号点存储于 “x.txt”文本中。-DSP input signal generation process. Conio.cpp achieve X (n) signal, in which there are two components of the frequency sinusoidal signals (sine calculation sinwn.cpp achieved), variable frequency, here and frequencies from 140Hz. Consider a Gaussian white noise from gauss.cpp achieve. Finally, the signal generated 2,000 points, the final point signal storage in the "x.txt" text.

! 2 ! HOMPACK90 is a suite of FORTRAN 90 subroutines for solving nonlinear 3 ! systems of equations by homotopy methods. There are subroutines for 4 ! fixed point, zero finding, and general homotopy curve tracking problems, 5 ! utilizing both dense and sparse Jacobian matrices, and implementing 6 ! three different algorithms: ODE-based, normal flow, and augmented 7 ! Jacobian. The (driver) subroutines called by the user are given in the 8 ! table below, and are well documented internally. The user need not 9 ! be concerned with any other subroutines in HOMPACK90. -! 2 ! HOMPACK90 is a suite of FORTRAN 90 subroutines for solving nonlinear 3 ! systems of equations by homotopy methods. There are subroutines for 4 ! fixed point, zero finding, and general homotopy curve tracking problems, 5 ! utilizing both dense and sparse Jacobian matrices, and implementing 6 ! three different algorithms: ODE-based, normal flow, and augmented 7 ! Jac

快速排序是最受欢迎的因为它是不难实现，适用于各种不同类型的输入数据，并速度比任何其他在典型应用程序中的排序方法。它是的地方 （使用只有一个小的辅助堆栈），需要时间正比于 N log N 对平均的 N 对项目排序，并且有极短的内环。快速排序是分而治之的排序方法。它的工作原理数组分成两部分，然后独立排序部分。 

包含两个源码，Wavelet1D是一维小波变换；Wavelet2D，是二维小波变换。可直观的显示小波算法的演示效果。-includes two source, Wavelet1D Victoria is a wavelet transform; Wavelet2D. is a two-dimensional wavelet transform. Visual display may wavelet algorithm demonstration effect.

根据已知数据的权重对数据进行拟合，若不能给出数据的权重就默认为数据权重为1.这种拟合 方法是根据最小二乘进行的。-The basis of known data on the weight of the data fitting, if the data should not give the weight for data on the default weight is 1. This method is based on least-squares fitting carried out.