PLS_pls_PLS近红外_plsmatlab_Spectrum_光谱

% x=xlsread('D:\行星科学\dataSPEC345.csv') % z=diff(x,2);%二阶导数 % % z = diff(x,1);%一阶导数 % % z=sgolayfilt(x,3,5);%sg 3次 5点平滑 % % z = filter([0.2,0.2,0.2,0.2,0.2],1,x);% 5点移动平滑 % %归一化偏最小二乘法程序 close all; clear; clc; format short pzz=xlsread('D:\行星科学\dataSPEC345.csv');%每一列为一个指标，每一行为一个样本 A = []; ncv = size(pzz,1);%读出样本数量 for i=1:1:ncv %留一交叉验证，循环次数由样本数确定 pz=pzz'; y_out = pzz(i,1); data_train = pzz(i,2:end); pzz(i,:) = []; [data,data_mean,d_std]=zscore(pzz); %数据标准化 n=1; % n 是因变量的个数 y=data(:,1:n);x=data(:,n+1:end);%定义因变量SIO2含量为第一列，自变量光谱数据为2到最后一列 [xl,yl,xs,ys,beta,pctvar,mse,stats] = plsregress(x,y,8); plot(1:8,cumsum(100*pctvar(2,:)),'-bo'); xlabel('Number of PLS components'); ylabel('Percent Variance Explained in y'); data_trainn = (data_train - data_mean(1,2:end))./d_std(1,2:end); yfit = [ones(size(data_train,1),1) data_trainn]*beta; yfi = yfit.*d_std(1,1) + data_mean(1,1); A(i,1) = yfi; A(i,2) = y_out; A(i,3)= abs( yfi - y_out); end % xlswrite('42',A) %未归一化 偏最小二乘结合留一交叉验证 % % close all; % clear; % clc; % format short % pzz=xlsread('4123');%每一列为一个指标，每一行为一个样本 % A = []; % ncv = size(pzz,2); % for i=1:1:ncv % pzz=pzz'; % y_out = pzz(i,1); % data_train = pzz(i,2:end); % pzz(i,:) = []; % %[data,data_mean,d_std]=zscore(pz); %数据标准化 % n=1; % n 是因变量的个数 % y=pzz(:,1:n);x=pzz(:,n+1:end);%定义因变量为第一列 % [xl,yl,xs,ys,beta,pctvar,mse,stats] = plsregress(x,y,4); % plot(1:4,cumsum(100*pctvar(2,:)),'-bo'); % % xlabel('Number of PLS components'); % % ylabel('Percent Variance Explained in y'); % %data_trainn = (data_train - data_mean(1,2:end))./d_std(1,2:end); % yfit = [ones(size(data_train,1),1) data_train]*beta; % % yfi = yfit.*d_std(1,1) + data_mean(1,1); % A(i,1) = yfit; % A(i,2) = y_out; % A(i,3)= abs( yfit - y_out); % % end % % xlswrite('31',A)