perceptron and ANN

%% Perceptron Regression
close all
clear

%%load data
x = load('ex4x.dat');
y = load('ex4y.dat');

x=ones(80,2);
for i=1:80
    x(i,1)=mod(i,10);
    x(i,2)=floor(i/10);
end

for i=1:80
    if (x(i,1)+x(i,2))<10
        y(i)=0;
    else
        y(i)=1;
    end
end

[m, n] = size(x);

% Add intercept term to x
x = [ones(m, 1), x];

%%draw picture
% find returns the indices of the
% rows meeting the specified condition
pos = find(y == 1);
neg = find(y == 0);
% Assume the features are in the 2nd and 3rd
% columns of x
figure('NumberTitle', 'off', 'Name', '感知机');
plot(x(pos, 2), x(pos,3), '+');
hold on;
plot(x(neg, 2), x(neg, 3), 'o');

%进行初始化
s = 1;                     % 标识符，当s=0时，表示迭代终止
n = 0;                     % 表示迭代的次数
N = 10000;                 %定义N为最大分类判别次数，判别次数超过此值则认定样本无法分类。
w= [0,0,0]';                % 取权向量初始值

% 开始迭代
while s
    J = 0;                % 假设初始的分类全部正确
    for i = 1:size(pos)
        if (x(pos(i),:)*w)<=0   % 查看x1分类是否错误，在x属于w1却被错误分类的情况下，w'x<0
            w = w+x(pos(i),:)';% 分类错误，对权向量进行修正
            J = 1;         % 置错误标志位为1
        end
    end
    for i = 1:size(neg)
        if (x(neg(i),:)*w)>=0    % 查看x2分类是否错误，在x属于w2却被错误分类的情况下，w'x>0
            w = w-x(neg(i),:)';% 分类错误，对权向量进行修正
            J = 1;         % 置错误标志位为1
        end
    end
    if J==0                 % 代价为0，即分类均正确
        s = 0;              % 终止迭代
    end
    n = n+1;            % 迭代次数加1
    if n == N
        s=0;
    end
end

w=[1;w(2)/w(1);w(3)/w(1)];

%%Calculate the decision boundary line
plot_x = [min(x(:,2)),  max(x(:,2))];
plot_y = (-1./w(3)).*(w(2).*plot_x +w(1));
plot(plot_x, plot_y)
legend('Admitted', 'Not admitted', 'Decision Boundary')
hold off