Coursera-MachineLearning-Week2编程题目整理

warmUpExercise.m

A = eye(5);

plotData.m

plot(x, y, 'rx', 'MarkerSize', 10);
ylabel('Profit in $10,000s');
xlabel('Population of City in 10,000s');

gradientDescent.m

    theta_temp = theta; %记录当前梯度for j=1:length(theta)  %对其中的所有都进行更新theta_temp(j) = theta(j) -alpha/m*(X*theta-y)'*X(:,j);  %endtheta = theta_temp;  %更新参数

computeCost.m

predictions = X * theta;
sqrErrors = (predictions - y) .^ 2 ;% 求每一项的平方误差
J = 1 / (2 * m) * sum(sqrErrors); %求代价函数

gradientDescentMulti.m

    S = (1 / m) * (X' * (X * theta - y));  %求导theta = theta - alpha .* S;  %梯度更新

computeCostMulti.m

J = 1 / (2 * m) * sum( (X * theta - y) .^ 2);

featureNormalize.m

mu = mean(X);  %计算均值
sigma = std(X, 1, 1);  %计算标准差for i = 1:size(X, 2)  %遍历所有列X_norm(:, i) = (X(:, i) - mu(i)) ./ sigma(i);  %进行标准化
end

normalEqn.m

theta = pinv((X'*X))*X'*y;