ex2练习

 

• 特征映射（两个特征映射成六次多项式）

  X = mapFeature(X(:,1), X(:,2));

  degree = 6;
  out = ones(size(X1(:,1)));
  for i = 1:degree
      for j = 0:i
          out(:, end+1) = (X1.^(i-j)).*(X2.^j);% 增加特征新列
      end
  end

   

• 假设函数

  g = 1./(1+exp(-z)); % 计算时z用 X*theta 替换
                      % g是m*1向量，用./对每一项进行计算

         

•  代价函数和偏导 

% 无正则项
J =(1/m)*((-y)'*log(sigmoid(X*theta))-(1-y)'*log(1-sigmoid(X*theta)));
grad = ((1/m)*(sigmoid(X*theta)-y)'*X)'; % 和theta维度一样，m*1

% 引入正则项
theta_cp =theta;
theta_cp(1) = 0;
% 计算正则项不包括theta(1)
J = (1/m)*((-y)'*log(sigmoid(X*theta))-(1-y)'*log(1-sigmoid(X*theta)))+(lambda/(2*m))*(theta_cp'*theta_cp);
grad = ((1/m)*(sigmoid(X*theta)-y)'*X)' + (lambda/m)*theta_cp;

 无正则项：

引入正则项： 

       

       

        

 

 

 

• 使用fminunc得到theta 

  data = load('ex2data1.txt');
  X = data(:, [1, 2]);
  y = data(:, 3);
  [m, n] = size(X);
  X = [ones(m, 1) X];
  initial_theta = zeros(n + 1, 1);
  options = optimset('GradObj', 'on', 'MaxIter', 400);
  
  % 无正则项
  [theta, cost] = ...
  	fminunc(@(t)(costFunction(t, X, y)), initial_theta, options);
  
  % 有正则项
  lambda = 1;
  [theta, J, exit_flag] = ...
  	fminunc(@(t)(costFunctionReg(t, X, y, lambda)), initial_theta, options);
  

   

• 预测（阈值0.5）

  pos = find(sigmoid(X*theta) >= 0.5);
  neg = find(sigmoid(X*theta) < 0.5);
  % p是m*1维向量
  p(pos,1) = 1;
  p(neg,1) = 0;
  % 判断准确度
  fprintf('Train Accuracy: %f\n', mean(double(p == y)) * 100);

   

• 绘图
  • 绘样本图

    % 寻找正负样本
    pos = find(y==1); neg = find(y == 0);
    % 正负样本分别绘制
    plot(X(pos, 1), X(pos, 2), 'k+','LineWidth', 2, 'MarkerSize', 7);
    plot(X(neg, 1), X(neg, 2), 'ko', 'MarkerFaceColor', 'y','MarkerSize', 7);
    
    hold on;
    xlabel('Exam 1 score')
    ylabel('Exam 2 score')
    legend('Admitted', 'Not admitted')
    hold off;

     

  • 绘边界图

    plotData(X(:,2:3), y); % 绘制散点
    hold on
    if size(X, 2) <= 3     % 如果特征数小于等于3，即边界面为直线，只需要两点确定一直线
        plot_x = [min(X(:,2))-2,  max(X(:,2))+2]; % 按特征x(2),选择最大、最小两点
        plot_y = (-1./theta(3)).*(theta(2).*plot_x + theta(1));
                                           % 根据theta(1)+theta(2)x(1)+theta(3)x(2)=0,计算x(3)
        plot(plot_x, plot_y) % 绘制曲线
        legend('Admitted', 'Not admitted', 'Decision Boundary')
        axis([30, 100, 30, 100])
    else                   % 如果特征数大于3，即边界面为曲线
        u = linspace(-1, 1.5, 50);
        v = linspace(-1, 1.5, 50);
        z = zeros(length(u), length(v));
        for i = 1:length(u)
            for j = 1:length(v)
                z(i,j) = mapFeature(u(i), v(j))*theta;
            end
        end
        z = z';
        contour(u, v, z, [0, 0], 'LineWidth', 2) % z=0 刚好是分界面
    end
    hold off