搜索资源列表
Subsample_MaxNeighborDistance_R
- A subsample of an input population that has max pair-distance and min projection error
Kmeanschuantong
- 传统k-means程序,包括最大误差和、迭代次数、运行时间-The traditional k- means the program, including the maximum error and, number of iterations, running time
zishiyingguolvfa
- 自适应过滤法是根据一组给定的权数对时间数列的历史观察值进行加权平均计算一个预测值,然后根据预测误差调整权数以减少误差,这样反复进行直至找出一组“最佳”权数,使误差减少到最低限度,再利用最佳权数进行加权平均预测。-Adaptive filtering method is based on the number of a given set of rights to compute a weighted average of the predicted value of historical tim
isodata-cluster
- isodata迭代自组织聚类算法源代码,直接运行,效果不错-Iterative Selforganizing Data Analysis cluster algorithm,source code without error
SgdClassifier
- 随机梯度下降分类器。本实验的实验平台为eclipse,只需导入(import)即可运行。输出方式为控制台输出,能够提供的评价数据有test error, percision, recall以及F1-measure。-Stochastic gradient descent classifier. In this study, experimental platform for eclipse, just import (import) to run. Output of the console o
SVM_regression_1001
- 一种可以择优选择参数的SVM拟合回归程序,通过for循环,可以调节参数,调出误差最小的参数设置。有时候,默认的参数也会有很不错的精度。-One can choose the best parameters of the SVM fitting regression procedures, through the for cycle, you can adjust the parameters, minimize the error of the parameter settings. Somet
fengqunsuanfaPID
- 针对工业控制中常用的PID控制器参数整定困难的问题,提出一种基于人工蜂群算法的参数整定方法。将PID控制器待整定参数看作蜜源,利用蜂群特有的角色转变机制搜索优质的参数组合 选取绝对误差矩积分性能指标作为参数寻优的目标函数。仿真实验结果表明,所采用的算法能够提高控制系统的动态性能,增强系统的快速性和稳定性,适用于PID控制器的自整定。 -For industrial control commonly used in PID controller tuning difficult problem
plot_cv_predict
- 等渗的插图对生成的数据回归。等张回归发现引入近似函数的同时最小化均方误差的训练数据。-An illustration of the isotonic regression on generated data. The isotonic regression finds a non-decreasing approximation of a function while minimizing the mean squared error on the training data.
yalemain
- 人脸识别,验证参数错误率,选择错误率小的参数 (对yale人脸库进行人脸识别),代码完整-Face recognition, validate parameter error rates, choose the parameters of the error rate is small (library of Yale face for face recognition), code integrity
698154
- forward error correction program performance algorithms for use()
knn
- 模式识别中的k近邻算法,经过测试,运行结果很好。 最小距离分类器 : 它将各类训练样本划分成若干子类,并在 每个子类中确定代表点 。测试样本的类别则以其与这些代表点距离最近作决策。该方法的缺点是所选择的代表点并不一定能很好地代表各类,其后果将使错误率增加。(The k nearest neighbor algorithm in pattern recognition has been tested and the result is very good. Minimum distance c
