Machine Learning by Andrew Ng --- neural network learning

The step of this exercise is show in the pdf which i have updoaded.

Neural network of this exercise is not easy to finish,okay,let me show U.

nnCostFunction:

function [J grad] = nnCostFunction(nn_params, ...
                                   input_layer_size, ...
                                   hidden_layer_size, ...
                                   num_labels, ...
                                   X, y, lambda)
%NNCOSTFUNCTION Implements the neural network cost function for a two layer
%neural network which performs classification
%   [J grad] = NNCOSTFUNCTON(nn_params, hidden_layer_size, num_labels, ...
%   X, y, lambda) computes the cost and gradient of the neural network. The
%   parameters for the neural network are "unrolled" into the vector
%   nn_params and need to be converted back into the weight matrices. 
% 
%   The returned parameter grad should be a "unrolled" vector of the
%   partial derivatives of the neural network.
%

% Reshape nn_params back into the parameters Theta1 and Theta2, the weight matrices
% for our 2 layer neural network
Theta1 = reshape(nn_params(1:hidden_layer_size * (input_layer_size + 1)), ...
                 hidden_layer_size, (input_layer_size + 1));

Theta2 = reshape(nn_params((1 + (hidden_layer_size * (input_layer_size + 1))):end), ...
                 num_labels, (hidden_layer_size + 1));

% Setup some useful variables
m = size(X, 1);
         
% You need to return the following variables correctly 
J = 0;
Theta1_grad = zeros(size(Theta1));
Theta2_grad = zeros(size(Theta2));

% ====================== YOUR CODE HERE ======================
% Instructions: You should complete the code by working through the
%               following parts.
%
% Part 1: Feedforward the neural network and return the cost in the
%         variable J. After implementing Part 1, you can verify that your
%         cost function computation is correct by verifying the cost
%         computed in ex4.m
%
% Part 2: Implement the backpropagation algorithm to compute the gradients
%         Theta1_grad and Theta2_grad. You should return the partial derivatives of
%         the cost function with respect to Theta1 and Theta2 in Theta1_grad and
%         Theta2_grad, respectively. After implementing Part 2, you can check
%         that your implementation is correct by running checkNNGradients
%
%         Note: The vector y passed into the function is a vector of labels
%               containing values from 1..K. You need to map this vector into a 
%               binary vector of 1's and 0's to be used with the neural network
%               cost function.
%
%         Hint: We recommend implementing backpropagation using a for-loop
%               over the training examples if you are implementing it for the 
%               first time.
%
% Part 3: Implement regularization with the cost function and gradients.
%
%         Hint: You can implement this around the code for
%               backpropagation. That is, you can compute the gradients for
%               the regularization separately and then add them to Theta1_grad
%               and Theta2_grad from Part 2.
%

%=================================Costfunction========================================
X=[ones(size(X,1),1) X];

a1=X;
z2=Theta1*a1';
a2=sigmoid(z2);

a2=[ones(1,m) ; a2];
z3=Theta2*a2;
a3=sigmoid(z3);

y_vec=zeros(num_labels,m);
for index=1:m
	y_vec(y(index),index)=1;
end

size(y_vec);

for index_num=1:m
	J+= sum(-1*y_vec(:,index_num).*log(a3(:,index_num))-(1-y_vec(:,index_num)).*log(1-a3(:,index_num)));
end 
J=J/m;
size(J);
J=J+(lambda/( 2*m )) * (sum  (sum  ( Theta1(:,2:end).^2 ) )+ sum(sum(Theta2(:,2:end).^2)) )  

%===================================backprogation=======================================
%Delta2=(zeros(size(Theta2))) (:,2:end);
%Delta1=(zeros(size(Theta1))) (:,2:end);


%for ind=1:m
%	delta3=a3(:,ind)-y_vec(:,ind);
%	delta2=sum( (Theta2'*delta3)(2:end).*sigmoidGradient(z2(:,ind)),2 );
%	Delta2=Delta2 + delta3*a2(2:end,ind)';
%	test=delta2*a1(ind,2:end);
%	Delta1=Delta1 + delta2*a1(ind,2:end);
%end
Delta1 = zeros( hidden_layer_size, (input_layer_size + 1));
Delta2 = zeros( num_labels, (hidden_layer_size + 1));

%Delta需要bias unit，delta不需要bias unit

for i=1:m
    delta3 = a3(:,i) - y_vec(:,i);  
      
    #注意这里的δ是不包含bias unit的delta的，毕竟bias unit永远是1，  
    #不需要计算delta, 下面的2:end,: 过滤掉了bias unit相关值  
    delta2 = (Theta2'*delta3)(2:end,:).*sigmoidGradient(z2(:,i));  
    #移除bias unit上的delta2，但是由于上面sigmoidGradient式子中  
    #的z，本身不包含bias unit，所以下面的过滤不必要，注释掉。  
    #delta2 = delta2(2:end);  
    Delta2 += delta3 * a2(:,i)';  
      
    #第一层的input是一行一行的，和后面的结构不一样，后面是一列作为一个example  
    Delta1 += delta2 * a1(i,:); 
end
Theta1_grad=Delta1/m;
Theta2_grad=Delta2/m;

Theta2_grad(:,2:end) = Theta2_grad(:,2:end) .+ lambda * Theta2(:,2:end) / m;  
Theta1_grad(:,2:end) = Theta1_grad(:,2:end) .+ lambda * Theta1(:,2:end) / m; 


% =========================================================================

% Unroll gradients
grad = [Theta1_grad(:) ; Theta2_grad(:)];

%训练集一个一个轮流训练，一次只训练一个set，前进后反向。
end

sigmoidGradient:

function g = sigmoidGradient(z)
%SIGMOIDGRADIENT returns the gradient of the sigmoid function
%evaluated at z
%   g = SIGMOIDGRADIENT(z) computes the gradient of the sigmoid function
%   evaluated at z. This should work regardless if z is a matrix or a
%   vector. In particular, if z is a vector or matrix, you should return
%   the gradient for each element.

g = zeros(size(z));

% ====================== YOUR CODE HERE ======================
% Instructions: Compute the gradient of the sigmoid function evaluated at
%               each value of z (z can be a matrix, vector or scalar).
%g=( 1.0 ./ (1.0 .+ exp(-z)) ) *( (1.0 .- (1.0 ./ (1.0 .+ exp(-z)) ) )' );
g=sigmoid(z).*(1-sigmoid(z)); 

randomInitializeWeights：

function W = randInitializeWeights(L_in, L_out)
%RANDINITIALIZEWEIGHTS Randomly initialize the weights of a layer with L_in
%incoming connections and L_out outgoing connections
%   W = RANDINITIALIZEWEIGHTS(L_in, L_out) randomly initializes the weights 
%   of a layer with L_in incoming connections and L_out outgoing 
%   connections. 
%
%   Note that W should be set to a matrix of size(L_out, 1 + L_in) as
%   the column row of W handles the "bias" terms
%

% You need to return the following variables correctly 
W = zeros(L_out, 1 + L_in);

% ====================== YOUR CODE HERE ======================
% Instructions: Initialize W randomly so that we break the symmetry while
%               training the neural network.
%
% Note: The first row of W corresponds to the parameters for the bias units
%
epsilon_init=0.12;
W=rand(L_out,L_in+1) * 2 * epsilon_init - epsilon_init;
% =========================================================================

end

Tips:

1. some build-in functions in MATLAB must pass a vector of theta,rather than matrix.

2. for y,if U use BP to solve multip-classification problems,then U must change every y to vector.

3. in BP,every single traning set is pass to BP one by one .for the first one,FP using x then BP using y,the second ,third...