Evolutionary Parameter Optimization, bagging and W-RepTree

s-a-s-h

Hallo together,

I´ve tried to optimize of a bagging scheme combined with W-RepTree by a parameter optimization. Optimization regarding the Bagging (iterations, sample_ratio) works fine. Optimization of the W-RepTree unfortunately works not with the code below.

Does anybody have advice ?

Thank you,
Sascha

<operator name="Root" class="Process" expanded="yes">
    <operator name="ExcelExampleSource" class="ExcelExampleSource">
        <parameter key="create_label"	value="true"/>
        <parameter key="excel_file"	value="D:\Dissdaten\Literatur\Papers\ab08-08\DataMiningSoftware\RapidminerYALE\Trainingsdaten5.xls"/>
        <parameter key="first_row_as_names"	value="true"/>
    </operator>
    <operator name="EvolutionaryParameterOptimization" class="EvolutionaryParameterOptimization" expanded="yes">
        <list key="parameters">
          <parameter key="Bagging2.iterations"	value="[5.0;100.0]"/>
          <parameter key="Bagging2.sample_ratio"	value="[0.0;1.0]"/>
          <parameter key="Training.M"	value="[-1.0;5.0]"/>
        </list>
        <operator name="Validation" class="XValidation" expanded="yes">
            <parameter key="sampling_type"	value="shuffled sampling"/>
            <operator name="Bagging2" class="Bagging" expanded="yes">
                <parameter key="iterations"	value="64"/>
                <parameter key="sample_ratio"	value="0.8633563476940593"/>
                <operator name="Training" class="W-REPTree">
                    <parameter key="L"	value="-10.0"/>
                    <parameter key="M"	value="2.387723832548856"/>
                    <parameter key="N"	value="5.0"/>
                    <parameter key="P"	value="true"/>
                    <parameter key="S"	value="56.0"/>
                </operator>
            </operator>
            <operator name="ApplierChain" class="OperatorChain" expanded="yes">
                <operator name="Test" class="ModelApplier">
                    <list key="application_parameters">
                    </list>
                </operator>
                <operator name="Evaluation" class="RegressionPerformance">
                    <parameter key="absolute_error"	value="true"/>
                    <parameter key="main_criterion"	value="absolute_error"/>
                    <parameter key="normalized_absolute_error"	value="true"/>
                    <parameter key="root_mean_squared_error"	value="true"/>
                    <parameter key="squared_error"	value="true"/>
                </operator>
            </operator>
        </operator>
        <operator name="Log (2)" class="ProcessLog">
            <parameter key="filename"	value="paraopt.log"/>
            <list key="log">
              <parameter key="sample_ratio"	value="operator.Bagging2.parameter.keep_example_set"/>
              <parameter key="M"	value="operator.Training.parameter.M"/>
              <parameter key="V"	value="operator.Training.parameter.V"/>
              <parameter key="N"	value="operator.Training.parameter.N"/>
              <parameter key="S"	value="operator.Training.parameter.S"/>
              <parameter key="P"	value="operator.Training.parameter.P"/>
              <parameter key="L"	value="operator.Training.parameter.L"/>
              <parameter key="absolute"	value="operator.Evaluation.value.absolute_error"/>
              <parameter key="iterations"	value="operator.Bagging2.parameter.iterations"/>
            </list>
        </operator>
    </operator>

s-a-s-h

an extension: the same problems appear with W-FT of the WEKA package...

land

Hi Sascha,
the problem is within the weka learner W-REPTree. It defines the parameter M as real, but needs an integer. Unfortunatly we have no influence on the weka parameters.
An solution would be to remove the M value from evoultionary parameter optimization and sourround the evolutionary optimization with a gridParameterOptimization. Then you could only insert integer values.

Greetings,
  Sebastian