文件列表:

Spider\spider_jul2406.zip (5483499, 2008-07-08)
Spider\all_extra28-Jul-2006.zip (349307, 2008-07-08)
Spider\spider_jul2406\spider\basic\@algorithm\algorithm.m (2941, 2004-03-31)
Spider\spider_jul2406\spider\basic\@algorithm\am_i_data.m (154, 2003-05-12)
Spider\spider_jul2406\spider\basic\@algorithm\display.m (1434, 2003-05-12)
Spider\spider_jul2406\spider\basic\@algorithm\display_simple.m (1526, 2003-05-12)
Spider\spider_jul2406\spider\basic\@algorithm\extract.m (2111, 2004-03-11)
Spider\spider_jul2406\spider\basic\@algorithm\gen.m (254, 2003-05-12)
Spider\spider_jul2406\spider\basic\@algorithm\get_name.m (155, 2004-03-26)
Spider\spider_jul2406\spider\basic\@algorithm\isdeferred.m (56, 2004-03-23)
Spider\spider_jul2406\spider\basic\@algorithm\subsasgn.m (3354, 2004-11-18)
Spider\spider_jul2406\spider\basic\@algorithm\subsref.m (2943, 2004-08-23)
Spider\spider_jul2406\spider\basic\@algorithm\test.m (2211, 2004-12-02)
Spider\spider_jul2406\spider\basic\@algorithm\train.m (3690, 2004-11-18)
Spider\spider_jul2406\spider\basic\@algorithm\traintest.m (3929, 2004-03-26)
Spider\spider_jul2406\spider\basic\@algorithm\traintests.m (437, 2003-05-12)
Spider\spider_jul2406\spider\basic\@chain\chain.m (1112, 2006-07-24)
Spider\spider_jul2406\spider\basic\@chain\display.m (1461, 2003-05-12)
Spider\spider_jul2406\spider\basic\@chain\subsasgn.m (3417, 2004-08-23)
Spider\spider_jul2406\spider\basic\@chain\subsref.m (2998, 2004-08-23)
Spider\spider_jul2406\spider\basic\@chain\testing.m (105, 2003-05-12)
Spider\spider_jul2406\spider\basic\@chain\training.m (249, 2003-05-12)
Spider\spider_jul2406\spider\basic\@cv\cv.m (2334, 2006-07-24)
Spider\spider_jul2406\spider\basic\@cv\display.m (1829, 2003-05-12)
Spider\spider_jul2406\spider\basic\@cv\get_name.m (372, 2003-05-12)
Spider\spider_jul2406\spider\basic\@cv\subsasgn.m (3418, 2004-08-23)
Spider\spider_jul2406\spider\basic\@cv\subsref.m (3007, 2004-08-23)
Spider\spider_jul2406\spider\basic\@cv\testing.m (640, 2003-05-12)
Spider\spider_jul2406\spider\basic\@cv\training.m (4370, 2006-02-08)
Spider\spider_jul2406\spider\basic\@data\concatenate.m (720, 2003-05-12)
Spider\spider_jul2406\spider\basic\@data\data.asv (2514, 2006-02-18)
Spider\spider_jul2406\spider\basic\@data\data.m (2569, 2006-07-24)
Spider\spider_jul2406\spider\basic\@data\display.m (706, 2005-05-18)
Spider\spider_jul2406\spider\basic\@data\get.m (684, 2006-05-31)
Spider\spider_jul2406\spider\basic\@data\get_dim.m (653, 2006-07-19)
Spider\spider_jul2406\spider\basic\@data\get_index.m (71, 2003-05-12)
Spider\spider_jul2406\spider\basic\@data\get_name.m (213, 2004-03-26)
Spider\spider_jul2406\spider\basic\@data\get_x.m (431, 2003-05-12)
Spider\spider_jul2406\spider\basic\@data\get_xy.m (420, 2003-05-12)
Spider\spider_jul2406\spider\basic\@data\get_y.m (388, 2003-05-12)
... ...

STATfeatureselect software package ------------------------------------ This software ranks all the features using t-statistics, and computes Leave-One-Out (LOO) accuracy and test accuracy with respect to 2, 4, 8, ..., until all the features, using the naive Bayes classifier. (1) This package runs only on Matlab and consists of the following four modules: STATfeatureselect.m : main module for this software package. NaiveBayes.m : naive Bayes classification. tTestRanking.m : rank features based on t-statistics. normalizeData.m : make each instance unit length. (2) Input data files : training/testing input : must be a matrix. : each row is a feature, and each column is an instance. label : must be a column vector. if the data is binary, then the label must be either 1 or 2. if the data has three categories, then the label must be one of 1, 2, or 3. Output files : There are two output files from this program. One is about LOO accuracy and test accuracy (if testing data was provided) with respect to 2, 4, 8, ..., until all the features. The default output file name for this is 'statFeatSelOutput'. The other output file is about the ranked feature list and its file name is 'rankedFeatureScoreList'. (3) How to run : The easiest way to run this program is to write a .m file and the .m file (say, demo.m) may look like : % load data files. trainX = load('train.dat'); testX = load('test.dat'); trainY = load('trainLabel'); testY = load('testLabel'); normData = 0; % run the program. acc = STATfeatureselect(trainX, trainY, testX, testY,normData); In the STATfeatureselect function, 5th argument is for the normalization option. If it is 0, then data is normalized. If it is 1, then data is not normalized. If no 5th argument is specified (default), then data is normalized. Once you edit the demo.m, open the Matlab and on the matlab prompt, type in demo.

近期下载者：

相关文件：

评论：[我要评论] [举报此文件]

收藏者：