Add files via upload

Author: ttungl

week7/machine-learning-ex6/ex6/dataset3Params.m

@@ -23,11 +23,28 @@
 %        mean(double(predictions ~= yval))
 %
 
+test_range = [0.01, 0.03, 0.1, 0.3, 1, 3, 10, 30];
+C_opt = C;
+sigma_opt = sigma;
+test_range_len = size(test_range,2);
+error_min = 10^6;
+for i = 1:test_range_len
+	for j = 1:test_range_len
+		C = test_range(i);
+		sigma = test_range(j);
+		model = svmTrain(X, y, C, @(x1, x2) gaussianKernel(x1, x2, sigma));
+		predictions = svmPredict(model, Xval);
+		error_fraction = mean(double(predictions ~= yval));
+		if error_fraction < error_min 
+			C_opt = C;
+			sigma_opt = sigma;
+			error_min = error_fraction;
+		end		
+	end
+end
 
-
-
-
-
+C = C_opt;
+sigma = sigma_opt;
 
 % =========================================================================
 

week7/machine-learning-ex6/ex6/emailFeatures.m

@@ -48,6 +48,11 @@
 %
 %
 
+for i = word_indices
+	x(i) = 1;
+end
+
+
 
 
 

week7/machine-learning-ex6/ex6/ex6.m

@@ -51,6 +51,7 @@
 % You should try to change the C value below and see how the decision
 % boundary varies (e.g., try C = 1000)
 C = 1;
+% C = 100;
 model = svmTrain(X, y, C, @linearKernel, 1e-3, 20);
 visualizeBoundaryLinear(X, y, model);
 

week7/machine-learning-ex6/ex6/gaussianKernel.m

@@ -15,11 +15,30 @@
 %               sigma
 %
 %
+summation = norm(x1-x2) .^2;
 
+% K_gaussian 
+sim = exp(-(summation/(2*sigma .^2)));
 
 
+% ----
+% Tung Le's result
+% >> ex6
+% Loading and Visualizing Data ...
+% Program paused. Press enter to continue.
 
+% Training Linear SVM ...
 
+% Training ......................................................................
+% ........................................ Done!
+
+% Program paused. Press enter to continue.
+
+% Evaluating the Gaussian Kernel ...
+% Gaussian Kernel between x1 = [1; 2; 1], x2 = [0; 4; -1], sigma = 2.000000 :
+%         0.324652
+% (for sigma = 2, this value should be about 0.324652)
+% Program paused. Press enter to continue.
 
 % =============================================================
 

week7/machine-learning-ex6/ex6/octave-workspace

@@ -33,24 +33,23 @@
 % Look for one or more characters between 0-9
 email_contents = regexprep(email_contents, '[0-9]+', 'number');
 
-% Handle URLS
+% Handle URLS % replace the URL to httpaddr.
 % Look for strings starting with http:// or https://
 email_contents = regexprep(email_contents, ...
                            '(http|https)://[^\s]*', 'httpaddr');
 
-% Handle Email Addresses
+% Handle Email Addresses, replace by emailaddr
 % Look for strings with @ in the middle
 email_contents = regexprep(email_contents, '[^\s]+@[^\s]+', 'emailaddr');
 
-% Handle $ sign
+% Handle $ sign, replace the dollar sign by dollar char
 email_contents = regexprep(email_contents, '[$]+', 'dollar');
 
-
 % ========================== Tokenize Email ===========================
 
 % Output the email to screen as well
 fprintf('\n==== Processed Email ====\n\n');
-
+% fprintf('%s ', email_contents);
 % Process file
 l = 0;
 
@@ -97,14 +96,13 @@
     %       str2). It will return 1 only if the two strings are equivalent.
     %
 
-
-
-
-
-
-
-
-
+    % function: 
+    % If a match exists, add the index of 'str' to word_indices.
+    for i = 1:length(vocabList)
+        if(strcmp(str,vocabList{i}))
+            word_indices = [word_indices; i]
+        end
+    end
 
     % =============================================================
 

week7/machine-learning-ex6/ex6/processEmail.m

@@ -0,0 +1,15 @@
+# Created by Octave 3.8.0, Tue Mar 28 00:00:33 2017 CDT <tungthanhle@Tungs-MacBook-Pro.local>
+# name: email
+# type: sq_string
+# elements: 1
+# length: 16
+ttungl@gmail.com
+
+
+# name: token
+# type: sq_string
+# elements: 1
+# length: 16
+NsyW3miuT2G4WGYI
+
+

week7/machine-learning-ex6/ex6/token.mat

@@ -0,0 +1,112 @@
+
+word_indices =
+471
+1893
+1809
+1699
+997
+479
+1182
+1064
+1231
+1827
+810
+1893
+103
+1070
+74
+1346
+837
+1852
+1242
+1699
+1893
+1631
+1665
+1852
+997
+1893
+479
+1120
+1231
+1827
+1182
+1064
+1676
+877
+1113
+234
+1191
+1120
+792
+1287
+1377
+1120
+1699
+708
+1666
+440
+1120
+1120
+1120
+1093
+1230
+1844
+1809
+1699
+997
+1490
+1060
+997
+1666
+234
+74
+708
+1666
+608
+869
+1120
+1048
+825
+1896
+1117
+1120
+1120
+1120
+976
+676
+1699
+1895
+234
+74
+1852
+867
+1893
+1699
+1230
+956
+1896
+1844
+103
+387
+997
+479
+1120
+1265
+1231
+1827
+1120
+1120
+1120
+
+number numberljgvnumb numberleannumberlrmsnumb
+numberwxhonumberqiytnumb numberrjuvnumberhqcfnumb numbereidbnumberdmtvlnumb
+
+=========================
+
+Processed spamSample1.txt
+
+Spam Classification: 1
+(1 indicates spam, 0 indicates not spam)
+
+>>