Exercise6 - Week7 Finished

Author: ProgrammerYuan

machine-learning-ex6/ex6/dataset3Params.m

@@ -11,6 +11,14 @@
 C = 1;
 sigma = 0.3;
 
+% Predefine Enum
+C_vec = [0.01, 0.03, 0.1, 0.3, 1, 3, 10, 30];
+sigma_vec = [0.01, 0.03, 0.1, 0.3, 1, 3, 10, 30];
+
+% Prepare variables
+n = length(C_vec);
+error_matrix = zeros(n, n);
+
 % ====================== YOUR CODE HERE ======================
 % Instructions: Fill in this function to return the optimal C and sigma
 %               learning parameters found using the cross validation set.
@@ -23,11 +31,21 @@
 %        mean(double(predictions ~= yval))
 %
 
+for i = 1:n
+    for j = 1:n
+        C_value = C_vec(i);
+        sigma_value = sigma_vec(j);
+        model = svmTrain(X, y, C_value, @(x1, x2) gaussianKernel(x1, x2, sigma_value));
+        pred = svmPredict(model, Xval);
+        error_matrix(i, j) = mean(double(pred ~= yval));
+    end
+end
 
+[v, I] = min(error_matrix(:));
+[I1,I2] = ind2sub(size(error_matrix),I);
 
-
-
-
+C = C_vec(I1);
+sigma = sigma_vec(I2);
 
 % =========================================================================
 

machine-learning-ex6/ex6/emailFeatures.m

@@ -6,7 +6,8 @@
 
 % Total number of words in the dictionary
 n = 1899;
-
+% Total number of word_indices
+m = length(word_indices);
 % You need to return the following variables correctly.
 x = zeros(n, 1);
 
@@ -48,7 +49,9 @@
 %
 %
 
-
+for i = 1:m
+    x(word_indices(i)) = 1;
+end
 
 
 

machine-learning-ex6/ex6/gaussianKernel.m

@@ -16,7 +16,7 @@
 %
 %
 
-
+sim = exp(-sum((x1 - x2).^2) / (2 * (sigma.^2)));
 
 
 

machine-learning-ex6/ex6/processEmail.m

@@ -1,4 +1,4 @@
-function word_indices = processEmail(email_contents)
+  function word_indices = processEmail(email_contents)
 %PROCESSEMAIL preprocesses a the body of an email and
 %returns a list of word_indices 
 %   word_indices = PROCESSEMAIL(email_contents) preprocesses 
@@ -97,12 +97,13 @@
     %       str2). It will return 1 only if the two strings are equivalent.
     %
 
-
-
-
-
-
-
+    n = length(vocabList);
+    for i = 1:n
+        if strcmp(str, vocabList{i}) == 1
+            word_indices = [word_indices; i];
+            break;
+        end
+    end