UFLDL Tutorial 逻辑回归

function [f,g] = logistic_regression(theta, X,y)
  %
  % Arguments:
  %   theta - A column vector containing the parameter values to optimize.
  %   X - The examples stored in a matrix.  
  %       X(i,j) is the i'th coordinate of the j'th example.
  %   y - The label for each example.  y(j) is the j'th example's label.
  %


  m=size(X,2);
  
  % initialize objective value and gradient.
  f = 0;
  g = zeros(size(theta));




  %
  % TODO:  Compute the objective function by looping over the dataset and summing
  %        up the objective values for each example.  Store the result in 'f'.
  %
  % TODO:  Compute the gradient of the objective by looping over the dataset and summing
  %        up the gradients (df/dtheta) for each example. Store the result in 'g'.
  %
%%% YOUR CODE HERE %%%
h = sigmoid(X'*theta);
f = -y*log(h)+(1-y)*log(1-h);

g = X*(h-y');

Step Size below progTol
Optimization took 0.208678 seconds.
Training accuracy: 99.7%
Test accuracy: 99.9%