How to use W vector in adaboost SVM?

2 次查看（过去 30 天）

Fateme Jalali 2016-8-20

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/300235-how-to-use-w-vector-in-adaboost-svm

评论： Image Analyst 2016-8-20

Hello,I wrote this code for multiclass classification using svm and adaboost. I do not know how to use W vector for libsvm option (cmd). because W in cmd must be a number, not a vector.can any one help me plz? Thanks.

    N = length(X); % X training labels
    W = 1/N * ones(N,1); %Weights initialization
    M = 10; % Number of boosting iterations 
    for m=1:M
        C = 10; %The cost parameters of the linear SVM, you can...
                 perform a grid search for the optimal value as well           
        %Calculate the error and alpha in adaBoost with cross validation
        cmd = ['-c ', num2str(C), ' -w ', num2str(W)];
        model = svmtrain(X, Y, cmd);
        [Xout, acc, ~] = svmpredict(X,Y,cmd);
        err = sum(.5 * W .* acc * N)/sum(W);
        alpha = log( (1-err)/err );
        % update the weight
        W = W.*exp( - alpha.*Xout.*X );
        W = W/norm(W);
    end