Options

Newb: Can't get a Training and Test Set working with EvoSVM

benniebennie Member Posts: 3 Contributor I
edited November 2018 in Help
Hi Guys,

i'm trying for hours to get a training set and test some unlabeled data to that set. i read the tutorial and searched the forum but i can't get it working :(

i created two processes. the first should create my training set and the second give my an output for what my testdata is classified into.

1)
 
<operator name="Root" class="Process" expanded="yes">
    <operator name="ExampleSource" class="ExampleSource">
        <parameter key="attributes" value="C:\Users\Admin\Documents\rm_workspace\SDNN5Min.aml"/>
    </operator>
    <operator name="XValidation" class="XValidation" expanded="yes">
        <parameter key="create_complete_model" value="true"/>
        <parameter key="number_of_validations" value="2"/>
        <parameter key="sampling_type" value="shuffled sampling"/>
        <operator name="EvoSVM" class="EvoSVM">
        </operator>
        <operator name="OperatorChain" class="OperatorChain" expanded="yes">
            <operator name="ModelApplier" class="ModelApplier">
                <list key="application_parameters">
                </list>
            </operator>
            <operator name="Performance" class="Performance">
            </operator>
        </operator>
    </operator>
    <operator name="ModelWriter" class="ModelWriter">
        <parameter key="model_file" value="C:\Users\Admin\Documents\rm_workspace\Train_SDNN5Min.mod"/>
    </operator>
</operator>
and

<operator name="Root" class="Process" expanded="yes">
    <operator name="ExampleSource" class="ExampleSource">
        <parameter key="attributes" value="C:\Users\Admin\Documents\rm_workspace\test.aml"/>
    </operator>
    <operator name="ModelLoader" class="ModelLoader">
        <parameter key="model_file" value="C:\Users\Admin\Documents\rm_workspace\Train_SDNN5Min.mod"/>
    </operator>
    <operator name="ModelApplier" class="ModelApplier">
        <list key="application_parameters">
        </list>
    </operator>
    <operator name="ExampleSetWriter" class="ExampleSetWriter">
        <parameter key="example_set_file" value="C:\Users\Admin\Documents\rm_workspace\asd.dat"/>
    </operator>
</operator>
My input data is for the trainingset (first column = Attrib, second = label )
0.022826,1
0.044986,1
0.037715,1
0.034799,1
0.039011,1
0.024744,1
0.045759,1
0.036254,1
0.020361,1
0.088304,1
0.035647,0
0.033225,0
0.18196,0
0.063602,0
0.032698,0
0.053571,0
0.055212,0
0.11204,0
0.043181,0
0.10037,0

TestSet: (Attribut)
0.033349
0.023054
0.042332
0.043541
0.037389
0.032592
0.021912
0.041374

what im doing wrong? :\
thanks for your help

best regards
bennie

Answers

  • Options
    benniebennie Member Posts: 3 Contributor I
    i kinda edited my process to ( stolen by http://rapid-i.com/rapidforum/index.php/topic,1659.0.html )

    <operator name="Root" class="Process" expanded="yes">
        <operator name="Trainingset" class="ExampleSource">
            <parameter key="attributes" value="C:\Users\Admin\Documents\rm_workspace\SDNN5Min.aml"/>
        </operator>
        <operator name="XValidation" class="XValidation" expanded="yes">
            <parameter key="create_complete_model" value="true"/>
            <parameter key="number_of_validations" value="2"/>
            <parameter key="sampling_type" value="shuffled sampling"/>
            <operator name="OperatorChain" class="OperatorChain" expanded="yes">
                <operator name="EvoSVM" class="EvoSVM">
                </operator>
                <operator name="ModelWriter" class="ModelWriter">
                    <parameter key="model_file" value="C:\Users\Admin\Documents\rm_workspace\TrainSDNN5.mod"/>
                </operator>
            </operator>
            <operator name="OperatorChain (2)" class="OperatorChain" expanded="yes">
                <operator name="ModelApplier" class="ModelApplier">
                    <list key="application_parameters">
                    </list>
                </operator>
                <operator name="Performance" class="Performance">
                </operator>
            </operator>
        </operator>
        <operator name="TestSet" class="ExampleSource">
            <parameter key="attributes" value="C:\Users\Admin\Documents\rm_workspace\SDNN5minTestset.aml"/>
        </operator>
        <operator name="ModelLoader" class="ModelLoader">
            <parameter key="model_file" value="C:\Users\Admin\Documents\rm_workspace\TrainSDNN5.mod"/>
        </operator>
        <operator name="ModelApplier (2)" class="ModelApplier">
            <list key="application_parameters">
            </list>
        </operator>
    </operator>
    the result for the prediction of my test set is from [9.9799 to 9.800]. well what does it mean?
    my label was 0 to 1 and now i got 9.800 so its pritty sure to be in class with the label 1? and if it would be 0.5 it would be exaclty between this two classes?
  • Options
    benniebennie Member Posts: 3 Contributor I
    now my output is correct
    tried it again and it worked dunno what i did wrong
    /closed
  • Options
    landland RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 2,531 Unicorn
    Hi,
    anyway, if you just start to work with RapidMiner, I would suggest to switch directly to RapidMiner 5.0. It will help you a lot on such problems, since the automatic meta data checking will indicate errors before executing the process and will offer Quickfixes if possible, which solve the problem.


    By the way: If you just started, how did you received RapidMiner? Did you download explicitly RapidMiner 4.6 on source forge?


    Greetings,
      Sebastian.
Sign In or Register to comment.