tuoyuannihe2

KNN with configurable label prior, chuncking calculation for large data, and estimation of posterior

用改进遗传算法求解TSP问题,并编制了完整的Matlab程序予以仿真实现.程序中选择算子采用最佳个体保存与赌轮选择相结合的策略,最后分析了最佳个体保存比例对寻优效果的影响. 更多还原 (Improved genetic algorithm for TSP, and the preparation of a complete Matlab simulation program to be achieved. Program, select the operator uses the best individual preservation and roulette wheel selection strategy of combining, the final analysis of the optimal ratio of individual saving optimization results impact. more reduction)

说明：  用matlab实现BP网络，各种激活函数任你选择。(Use matlab to realize BP network. You can choose all kinds of activation functions.)

在计算机上产生一组实验数据，首先产生一段零均值白噪声数据u（n）,令功率为 ，让u(n)通过一个三阶FIR： 得到y(n). .y(n)上加三个实正弦信号f1’=0.1,f2’=0.25,f3’=0.26调整 和正弦信号幅度信噪比大致为10dB,50dB,50dB. (1) 令N=256,描绘xn波形； (2)得出真实功率谱密度 . (3) 利用此实验数据Pisarenko谐波分解法估计该实验数据的正弦频率及幅度。 (On the computer to generate a set of experimental data, first generate a zero mean white noise data u (n), so that power, let u (n) through a third-order FIR: get y (n).. Y (n) added three real sinusoidal signal f1 ' = 0.1, f2' = 0.25, f3 ' = 0.26 and sine signal amplitude adjustment signal to noise ratio is approximately 10dB, 50dB, 50dB. (1) that N = 256, depicted xn waveform (2) the true power spectral density obtained. (3) using this experimental data, Pisarenko harmonic decomposition method to estimate the experimental data of the sinusoidal frequency and amplitude.)

la commande vectorille assossié avec l onduleur

说明：  用k均值聚类实现随机n个数分类到k类中。 k和n是可变的。用图形化显示聚类结果。 需要把两个文件放在同一目录下，运行wkmeans2D即可。(k-means )

无线传感网络定位代码，其中有很多算法的实现代码，包括误差分析等，介绍的非常详细，值得一看(Wireless sensor network localization code)

《提问的智慧》，作为一个程序员，你曾经遇到过在论坛上发帖无人问津，看看提问的智慧把(" Question the wisdom" as a programmer, you have encountered no one is interested to post in the forum to see the wisdom of the questions)

本例是一个用PIC16F73编写的门禁读卡器解码并显示的源程序(This case is a PIC16F73 prepared to use access control card reader and display the source decoder)

This work describes a framework for image hiding that exploits spatial domain color properties of natural images combined with spectral properties of the Fourier magnitude and phase of these images