SVM10分类问题（含数据集和代码）_SVM多分类算法_

% clc; clear; close all; %read data: reference: https://www.mathworks.com/help/matlab/ref/importdata.html X_train = importdata('X_train.mat'); y_train = importdata('y_train.mat'); X_test = importdata('X_test.mat'); y_test = importdata('y_test.mat'); %transposing the class label vectors y_train_transpose = transpose(y_train); y_test_transpose = transpose(y_test); %initialization %number of class-10 %number of test samples-3251 SVMModel = cell(10,1); label = zeros(10,3251); %1 in the place of index, other class 0 trainingClassLabelsMatrix = full(ind2vec(y_train_transpose,10)); %train the model one-vs-all for index=1:10 SVMModel{index} = fitcsvm(X_train,trainingClassLabelsMatrix(index,:),'OptimizeHyperparameters','all'); end %predict values for index=1:10 label(index,:) = predict(SVMModel{index},X_test); end %transform into index predictedLabel=vec2ind(label); %calculate accuracy accuracy = sum(y_test_transpose == predictedLabel)/length(y_test_transpose); accuracyPercentage = 100*accuracy; fprintf('Accuracy = %f%%\n',accuracyPercentage)