====== About ======
If you are new to RVKDE, we suggest to read the [[rvkde:readme|README]] first, and then this document.  The README introduces using two wrapper scripts (__kde-train.pl__ and __kde-predict.pl__) to execute RVKDE, which is very similar to the procedure of using the well-know [[http://www.csie.ntu.edu.tw/~cjlin/libsvm/|LIBSVM]].  An alternative way to use RVKDE is directly executing the __rvkde__ (or __rvkde.exe__ on Windows system) binary executable file.  There are two benefits by doing so:
  - The usage of __rvkde__ is much powerful and flexible.
  - You don't need to install the Perl environment, which might be a little annoying on Windows system.
The following sections are written mainly for Linux users.  The mapping of most operations are trivial on Windows system.

====== Preparation ======
  * Create a directory for this document.  Here we use __~/tmp__ as an example.<code>
cd ~
mkdir tmp
</code>
  * Download and install [[http://mbi.ee.ncku.edu.tw/rvkde/|RVKDE]].<code>
cd ~/tmp
wget http://mbi.ee.ncku.edu.tw/rvkde/res/rvkde-current-linux32.tgz
tar zxvf rvkde-current-linux32.tgz
</code>

====== Classify the satimage dataset ======
  * Change the current path __~/tmp__.  All following steps are supposed to execute in this directory.<code>
cd ~/tmp
</code>
  * Classify the satimage dataset.  Notice that your RVKDE version might be different.<code>
rvkde-0.2.3-final/rvkde --classify --train -v rvkde-0.2.3-final/satimage.scale -m rvkde-0.2.3-final/satimage.scale.model --ks 10 # training
rvkde-0.2.3-final/rvkde --classify --predict -m rvkde-0.2.3-final/satimage.scale.model -V rvkde-0.2.3-final/satimage.scale.t -a 1 -b 1 --ks 10 --kt 10 # testing
</code>
  * Classify the satimage dataset in one-step.<code>
rvkde-0.2.3-final/rvkde --classify --predict -v rvkde-0.2.3-final/satimage.scale -V rvkde-0.2.3-final/satimage.scale.t -a 1 -b 1 --ks 10 --kt 10
</code>

====== Parameter selection ======
Most machine learning tools provide some parameters for users.  For example, the __k__ in knn classification algorithm and the __k__ in k-means clustering algorithm.  From the optimistic view, these parameters provide flexibility and make machine learning tools more powerful.  However, from another point of view, these machine learning techniques cannot determine (or learn) some parameters automatically so that users must specify by themselves.

RVKDE provides two alternative ways to do its parameter selection as described in the two following sections.

====== Cross-validation ======
  * Change the current path to __~/tmp__.  All following steps are supposed to execute in this path.
  * Cross-validation on __satimage.scale__.<code>
rvkde-0.2.3-final/rvkde --cv --classify --acc -n 5 -v rvkde-0.2.3-final/satimage.scale -a 1 -b 1,2,0.5 --ks 1,30,1 --kt 1,30
</code>
Let's take a look at the command.
^  --cv | Switch rvkde into cross-validation mode. |
^  --classify | Tell rvkde we want to do classification rather than regression now. |
^  --acc | Use [[wp>accuracy|accuracy]] as the evaluation index. |
^  -n | Do __n__-fold cross-validation. |
^  -v | Followed by the dataset for cross-validation. |
^  -a | Set the range (begin, end and step) of __alpha__ values of RVKDE.  In this example, 1 is the only possible __alpha__ value. |
^  -b | Set the range (begin, end and step) of __beta__ values of RVKDE.  In this example, the possible __beta__ values are 1, 1.5 and 2. |
^  --ks | Set the range (begin, end and step) of __ks__ values of RVKDE.  In this example, the possible __ks__ values are 1, 2, ... 30. |
^  --kt | Set the range (begin, end and step) of __kt__ values of RVKDE.  In this example, the possible __kt__ values are also 1, 2, ... 30 since the default step is 1. |

  * The result looks like<code>
[0.918602] a=1 b=1 s=8 t=21...
</code>Which tell us the best parameter combination is __alpha__ = 1, __beta__ = 1, __ks__ = 8 and __kt__ = 21.  In addition, the accuracy under the best parameters is 0.918602.

===== Predict with the selected parameters =====
Now we have a parameter combination derived from cross-validation.
  * Use these parameters to predict __satimage.scale.t__.<code>
rvkde-0.2.3-final/rvkde --predict --classify --acc -v rvkde-0.2.3-final/satimage.scale -V rvkde-0.2.3-final/satimage.scale.t -a 1 -b 1 --ks 8 --kt 21
</code>
Let's take a look at the command.
^ --predict | Switch RVKDE into prediction mode (rather than cross-validation). |
^ -v | Followed by the training dataset. |
^ -V | Followed by the testing dataset. |

  * The result looks like<code>
[0.9175] a=1 b=1 s=8 t=21...
</code>It indicates that RVKDE can yield a accuracy of 0.9175 under this parameter combination when using __satimage.scale__ to predict __satimage.scale.t__.

====== Train, validate, and then test ======
Another common procedure for parameter selection is to create an independent validation set.  For example, you can use __satimage.scale.tr__ and __satimage.scale.val__ to do parameter selection and see how good the parameters are when applying on __satimage.scale.t__.
  * Use __satimage.scale.tr__ (as training set) and __satimage.scale.val__ (as validation set) to select the parameters.<code>
rvkde-0.2.3-final/rvkde --predict --classify --acc -v rvkde-0.2.3-final/satimage.scale.tr -V rvkde-0.2.3-final/satimage.scale.val -a 1 -b 1,2,0.5 --ks 1,30,1 --kt 1,30,1
</code>
  * The result looks like<code>
[0.913599] a=1 b=1 s=8 t=23...
</code>
  * Predict __satimage.scale.t__ with the selected parameters.<code>
rvkde-0.2.3-final/rvkde --predict --classify --acc -v rvkde-0.2.3-final/satimage.scale.tr -V rvkde-0.2.3-final/satimage.scale.t -a 1 -b 1 --ks 8 --kt 23
</code>
  * The result looks like<code>
[0.917] a=1 b=1 s=8 t=23...
</code>It reveals that RVKDE yields very close accuracies with these two parameter selection schemes.