资源简介
光谱信息的特征选择,通过云永欢等提出的VCPA来进行光谱信息的特征选择(文件中包含了VCPA,IRIV,VCPA-GA以及VCPA-IRIV等光谱的变量选择算法)。In this study, we propose a hybrid variable selection strategy based on the continuous shrinkage of variable space which is the core idea of variable combination population analysis (VCPA). The VCPA-based hybrid strategy continuously shrinks the variable space from big to small and optimizes it based on modified VCPA in the first step. It then employs iteratively retaining informative variables (IRIV) and a genetic algorithm (GA) to carry out further optimization in the second step. It takes full advantage of VCPA, GA, and IRIV, and makes up for their drawbacks in the face of high numbers of variables. Three NIR datasets and three variable selection methods including two widely-used methods (competitive adaptive reweighted sampling, CARS and genetic algorithm-interval partial least squares, GA–iPLS) and one hybrid method (variable importance in projection coupled with genetic algorithm, VIP–GA) were used to investigate the improvement of VCPA-based hybrid strategy.
代码片段和文件信息
function left_variables = backward_elimination(XyAfoldmethodvarnumber)
% Inputs:
%+++ X: The data matrix of size m x p
%+++ y: The reponse vector of size m x 1
%+++ A: the maximal principle to extract.
%+++ fold: the group number for cross validation.
%+++ method : pretreatment method.
%
% Outputs:
% left_variables : the variables that finally left.
CV=plscvfold(XyAfoldmethod);
RMSECV0 = CV.RMSECV;
screenProcess=nan(size(X2)-1size(X2)+2);
jj=1;
di=1:size(X2);
index=1:size(X2);
while size(X2)>1
parc=size(X2);
% fprintf(‘Cross-validation with whole variables of this turn: %g\n‘RMSECV0);
screenProcess(jj1)=RMSECV0;
RMSECV_temp=nan(1parc);
i=1;
for k=1:parc
cX=X;
cX(:k)=[];
CV1=plscvfold(cXyAfoldmethod0);
RMSECV_temp(k) = CV1.RMSECV;
% fprintf(‘Cross-validation without variables #%g: %g\n‘index(k)RMSECV_temp(k));
if di(index(k))==0
i=i+1;
else
% screenProcess: the screening process of backward_elimination
screenProcess(jjdi(index(k))+i)=RMSECV_temp(k);
end
end
[minRMSECVminRMSECVIndex]=min(RMSECV_temp);
jj=jj+1;
if minRMSECV<=RMSECV0
fprintf(‘Variable #%g has been washed out %d variables have been deleted.\n‘varnumber(minRMSECVIndex)jj-1);
screenProcess(jj-1end)=index(minRMSECVIndex);
di(index(minRMSECVIndex))=0;
X(:minRMSECVIndex)=[];
index(minRMSECVIndex)=[];
varnumber(minRMSECVIndex)=[];
RMSECV0=minRMSECV;
else
disp(‘No any variable-deleting is necessary screenning is finished.‘);
% disp([‘Left variables: ‘sprintf(‘%g‘varnumber)]);
break
end
end
left_variables=varnumber;
属性 大小 日期 时间 名称
----------- --------- ---------- ----- ----
文件 1781 2019-02-14 20:26 VCPA ba
文件 449024 2019-02-14 20:26 VCPA ba
文件 1018 2019-02-14 20:26 VCPA ba
文件 729 2019-02-14 20:26 VCPA ba
文件 992 2019-02-14 20:26 VCPA ba
文件 1364 2019-02-14 20:26 VCPA ba
文件 1266 2019-02-14 20:26 VCPA ba
文件 2764 2019-02-14 20:26 VCPA ba
文件 1211 2019-02-14 20:26 VCPA ba
文件 208 2019-02-14 20:26 VCPA ba
文件 342 2019-02-14 20:26 VCPA ba
文件 3531 2019-02-14 20:26 VCPA ba
文件 824 2019-02-14 20:26 VCPA ba
文件 1331 2019-02-14 20:26 VCPA ba
文件 7013 2019-02-14 20:26 VCPA ba
文件 5976 2019-02-14 20:26 VCPA ba
文件 6754 2019-02-14 20:26 VCPA ba
文件 7429 2019-02-14 20:26 VCPA ba
文件 22119 2019-02-14 20:26 VCPA ba
文件 19255 2019-02-14 20:26 VCPA ba
文件 726 2019-02-14 20:26 VCPA ba
文件 3645 2019-02-14 20:26 VCPA ba
文件 1784 2019-02-14 20:26 VCPA ba
文件 3479 2019-02-14 20:26 VCPA ba
文件 429 2019-02-14 20:26 VCPA ba
文件 48 2019-02-14 20:26 VCPA ba
文件 1365 2019-02-14 20:26 VCPA ba
文件 371 2019-02-14 20:26 VCPA ba
文件 2646 2019-02-14 20:26 VCPA ba
文件 962 2019-02-14 20:26 VCPA ba
文件 5012 2019-02-14 20:26 VCPA ba
............此处省略56个文件信息
- 上一篇:RCNP考试题目-交换路由
- 下一篇:无信息变量消除法
相关资源
- 搜索在s = 13 TeV pp碰撞中在pp碰撞中产
- 兴奋的重介子衰变为轻向量介子:光
- NMR和FTIR光谱在确定CeO的结构作用方面
- 简单二阶互联系统的非线性动力学分
- FEATURE SELECTION FOR KNOWLEDGE DISCOVERY AND
- 近红外光谱在油脂可皂化物含量测定
- 猕猴桃近红外光谱MLR建模中的波长选
- sNN = 5.02 TeV时p -Pb碰撞中强子p T光谱的
- 温度对短链多烯生物分子β胡萝卜
- IDL+ENVI开发多光谱遥感图像DN值求辐亮
- 手写数字识别-模板匹配法
- Stock_Watson_动态因子分析模型
- 果蝇优化算法优化支持向量回归程序
- 自己做的一个简单GUI扑克纸牌识别-
- 近红外光谱分析技术的应用及其局限
- multi output SVR
- AR过程的线性建模过程与各种功率谱估
- PCNN TOOLBOX
- plstoolbox.zip
- 中国国家基础地理信息系统GIS数据
- 粒子群微电网优化调度
- 矩阵分析-经典教材-中文版-Roger.A.Ho
- 压缩感知TwIST
- 基于最小错误率的贝叶斯手写数字分
- 最全系统辨识源代码,包括多种最小
- 导弹制导实验
- 模式识别之特征选择
- 画跟踪精确度图的程序.zip
- 重力场大地水准面及重力异常阶次误
- prtools5.2.3工具包
评论
共有 条评论