Info
此问题已关闭。 请重新打开它进行编辑或回答。
weird shape of ROC for my data
1 次查看(过去 30 天)
显示 更早的评论
this is the code that results in generating ROC for my data. i have attached my roundMeas matrix. the problem is that my ROCs look really strange. they are not actually like curves. and i don't know what is wrong. i also don't know anything about setting thresholds for plotting ROC. can any body help?
TPRvect=zeros(27,300); FPRvect=zeros(27,300); u=13; for e=1:6 for i=1:27 for j=1:300 xx=roundMeas(j,3,e); TPRvect(i,j)=xx{1,1}(i); yy=roundMeas(j,6,e); FPRvect(i,j)=yy{1,1}(i); end end for i=1:27 qq=TPRvect(i,:); ww=FPRvect(i,:); figure (u) plot(ww,qq) xlabel('FPR') ylabel('TPR') str=sprintf('ROC of class %d for clustering%d', i, e); title(str) h=figure(u); saveas(h,sprintf('FIG%d.jpeg',u)) u=u+1; end e=e+1; end
0 个评论
回答(0 个)
此问题已关闭。
另请参阅
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!