如何使用grid.py进行参数选择? [英] how can i use grid.py for parameter selection?

问题描述

我想使用C-SVM分类选择参数c和gamma 带有libsvm \ tools \ grid.py的RBF(径向基函数)内核，但是我不知道这怎么可能?我安装了libsvm，gnuplot和python，并在python中运行了grid.py，但是它有错误并且没有显示结果.

解决方案

%grid of parameters
folds = 5; 
[C,gamma] = meshgrid(-5:2:15, -15:2:3); 
%# grid search, and cross-validation 
cv_acc = zeros(numel(C),1); 
d= 2;
for i=1:numel(C)   
    cv_acc(i) = svmtrain(TrainLabel,TrainVec, ...          
        sprintf('-c %f -g %f -v %d -t %d', 2^C(i), 2^gamma(i), folds,d));
end
%# pair (C,gamma) with best accuracy
[~,idx] = max(cv_acc); 
%# contour plot of paramter selection 
contour(C, gamma, reshape(cv_acc,size(C))), colorbar
hold on;
text(C(idx), gamma(idx), sprintf('Acc = %.2f %%',cv_acc(idx)), ...  
    'HorizontalAlign','left', 'VerticalAlign','top') 
hold off 
xlabel('log_2(C)'), ylabel('log_2(\gamma)'), title('Cross-Validation Accuracy') 
%# now you can train you model using best_C and best_gamma
best_C = 2^C(idx); best_gamma = 2^gamma(idx); %# ...

这也执行网格搜索...但是使用matlab ...不使用grid.py ...也许有帮助...

i want to select parameters c and gamma for C-SVM classification using the RBF (radial basis function) kernel with libsvm\tools\grid.py, but i dont know how it's possible? i installed libsvm and gnuplot and python and runned the grid.py in python, but it had error and didn't show the results.

解决方案

%grid of parameters
folds = 5; 
[C,gamma] = meshgrid(-5:2:15, -15:2:3); 
%# grid search, and cross-validation 
cv_acc = zeros(numel(C),1); 
d= 2;
for i=1:numel(C)   
    cv_acc(i) = svmtrain(TrainLabel,TrainVec, ...          
        sprintf('-c %f -g %f -v %d -t %d', 2^C(i), 2^gamma(i), folds,d));
end
%# pair (C,gamma) with best accuracy
[~,idx] = max(cv_acc); 
%# contour plot of paramter selection 
contour(C, gamma, reshape(cv_acc,size(C))), colorbar
hold on;
text(C(idx), gamma(idx), sprintf('Acc = %.2f %%',cv_acc(idx)), ...  
    'HorizontalAlign','left', 'VerticalAlign','top') 
hold off 
xlabel('log_2(C)'), ylabel('log_2(\gamma)'), title('Cross-Validation Accuracy') 
%# now you can train you model using best_C and best_gamma
best_C = 2^C(idx); best_gamma = 2^gamma(idx); %# ...

This performs grid search as well... but using matlab... not using grid.py... maybe this helps...

这篇关于如何使用grid.py进行参数选择?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！