In my project, I generate random samples from N1, N2 ,N3 and i plot them and x is a 2 dimensional guassian distribution N(meani,sigmai). but I want also to plot the bayes decision regions for equal and different initial prob. can anyboby help me plz?
1 次查看(过去 30 天)
显示 更早的评论
clear all
close all
%N1
mean1=[0 0];
sigma1=[1 0; 0 1];
%N2
mean2=[1 1];
sigma2=[1 0;0 4];
%N3
mean3=[2 2];
sigma3=[1 -1;-1 4];
x1 = -4:.2:4;
x2 = -4:.2:4;
[X1,X2] = meshgrid(x1,x2);
y1=mvnpdf([X1(:) X2(:)],mean1,sigma1);
y1 = reshape(y1,length(x2),length(x1));
surf(x1,x2,y1);
caxis([min(y1(:))-.5*range(y1(:)),max(y1(:))]);
axis([-4 4 -4 4 0 .2])
xlabel('x1'); ylabel('x2'); zlabel('Probability Density');
figure(1)
hold on
y2=mvnpdf([X1(:) X2(:)],mean2,sigma2);
y2 = reshape(y2,length(x2),length(x1));
surf(x1,x2,y2);
caxis([min(y2(:))-.5*range(y2(:)),max(y2(:))]);
axis([-4 4 -4 4 0 .2])
xlabel('x1'); ylabel('x2'); zlabel('Probability Density');
hold on
y3=mvnpdf([X1(:) X2(:)],mean3,sigma3);
y3 = reshape(y3,length(x2),length(x1));
surf(x1,x2,y3);
caxis([min(y3(:))-.5*range(y3(:)),max(y3(:))]);
axis([-4 4 -4 4 0 .2])
xlabel('x1'); ylabel('x2'); zlabel('Probability Density');
s1=mvnrnd(mean1,sigma1,30);
i=zeros(1,30);
figure(2)
plot(s1,i,'v')
hold on
s2=mvnrnd(mean2,sigma2,30);
i=zeros(1,30);
plot(s2,i,'r*')
hold on
s3=mvnrnd(mean3,sigma3,30);
i=zeros(1,30);
plot(s3,i,'+')
figure(3)
c1=contour(x1,x2,y1);
hold on
c2=contour(x1,x2,y2);
hold on
c3=contour(x1,x2,y3);
0 个评论
回答(0 个)
另请参阅
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!