Added a comment about playing around with the epsilon to improve

training time.

--HG--
extra : convert_revision : svn%3Afdd8eb12-d10e-0410-9acb-85c331704f74/trunk%403778

examples/rvm_ex.cpp

@@ -99,6 +99,11 @@ int main()
     // here we make an instance of the rvm_trainer object that uses our kernel type.
     rvm_trainer<kernel_type> trainer;
 
+    // One thing you can do to reduce the RVM training time is to make its
+    // stopping epsilon bigger.  However, this might make the outputs less
+    // reliable.  But sometimes it works out well.  0.001 is the default.
+    trainer.set_epsilon(0.001);
+
     // Now we loop over some different gamma values to see how good they are.  Note
     // that this is a very simple way to try out a few possible parameter choices.  You 
     // should look at the model_selection_ex.cpp program for examples of more sophisticated 

examples/rvm_regression_ex.cpp

@@ -61,6 +61,11 @@ int main()
     cout << "using gamma of " << gamma << endl;
     trainer.set_kernel(kernel_type(gamma));
 
+    // One thing you can do to reduce the RVM training time is to make its
+    // stopping epsilon bigger.  However, this might make the outputs less
+    // reliable.  But sometimes it works out well.  0.001 is the default.
+    trainer.set_epsilon(0.001);
+
     // now train a function based on our sample points
     decision_function<kernel_type> test = trainer.train(samples, labels);