matlab机器学习库

knn = fitcknn(meas,species,'NumNeighbors',5);

CVMdl = crossval(knn);

kloss = kfoldLoss(CVMdl);

predict(knn,ones(1,size(meas,2)))

//latent:特征值（从大到小),score特征向量

[coeff, score, latent, tsquared, explained] = pca(data);

//score即为从大到小 排序 后的特征矩阵，取前k列即为取样本最具代表性的k个属性

//explained即为每一列对应的影响力，所有列加起来为100

svm = fitcsvm(meas,species);

CVMdl = crossval(svm);

kloss = kfoldLoss(CVMdl);

naivebayes = fitcnb(meas, species);

nb = crossval(naivebayes);

kloss = kfoldLoss(nb);

cart = fitctree(meas,species);

CVMdl = crossval(cart);

kloss = kfoldLoss(CVMdl);

b = TreeBagger(nTree,meas,species,'OOBPrediction','on');

rf = oobError(b);

kloss = rf(nTree,1);

ada = fitensemble(meas,species,'AdaBoostM1',100,'Tree',

'Holdout',0.5);

kloss = kfoldLoss(ada,'mode','cumulative');

kloss = kloss(100,1);