mirror of
https://github.com/jlengrand/MachineLearningCoursera.git
synced 2026-03-10 08:31:20 +00:00
38 lines
1.2 KiB
Matlab
38 lines
1.2 KiB
Matlab
function [theta, J_history] = gradientDescent(X, y, theta, alpha, num_iters)
|
|
%GRADIENTDESCENT Performs gradient descent to learn theta
|
|
% theta = GRADIENTDESENT(X, y, theta, alpha, num_iters) updates theta by
|
|
% taking num_iters gradient steps with learning rate alpha
|
|
|
|
% Initialize some useful values
|
|
m = length(y); % number of training examples
|
|
J_history = zeros(num_iters, 1);
|
|
|
|
for iter = 1:num_iters
|
|
|
|
% ====================== YOUR CODE HERE ======================
|
|
% Instructions: Perform a single gradient step on the parameter vector
|
|
% theta.
|
|
%
|
|
% Hint: While debugging, it can be useful to print out the values
|
|
% of the cost function (computeCost) and gradient here.
|
|
%
|
|
|
|
tmat = repmat( theta, 1, size(X, 1) );
|
|
h_theta = tmat' .* X;
|
|
to_sum = (sum(h_theta,2) - y);
|
|
|
|
to_sum_1 = to_sum .* X(:, 1);
|
|
to_sum_2 = to_sum .* X(:, 2);
|
|
|
|
theta(1) = theta(1) - (alpha * (1/m) * sum(to_sum_1));
|
|
theta(2) = theta(2) - (alpha * (1/m) * sum(to_sum_2));
|
|
|
|
% ============================================================
|
|
|
|
% Save the cost J in every iteration
|
|
J_history(iter) = computeCost(X, y, theta);
|
|
|
|
end
|
|
|
|
end
|