Hi all,
I'm trying to interatively optimize my classifier with WeightOptimization from RapidMiner 4. (Many of my processes are still in RM4, so I'm still somehow tied to RM4).
Can anybody share a running example? Haven't found anything about this operator.
I always get the error "Parameter optimization not supported for non-number parameter type 'BinominalClassificationPerformance.false_positive'"
Any clue?
Alex
<operator name="Classifier" class="Process" expanded="yes">
<operator name="NominalExampleSetGenerator" class="NominalExampleSetGenerator">
</operator>
<operator name="InfoGainRatioWeighting" class="InfoGainRatioWeighting">
</operator>
<operator name="WeightOptimization" class="WeightOptimization" expanded="yes">
<parameter key="parameter" value="BinominalClassificationPerformance.false_positive"/>
<operator name="XValidation" class="XValidation" expanded="yes">
<operator name="Weka-SMO" class="W-SMO">
<parameter key="D" value="true"/>
<parameter key="C" value="0.01"/>
<parameter key="N" value="1.0"/>
<parameter key="M" value="true"/>
<parameter key="K" value="weka.classifiers.functions.supportVector.PolyKernel -D -C 0 -E 1.0"/>
</operator>
<operator name="OperatorChain" class="OperatorChain" expanded="yes">
<operator name="Apply Model" class="ModelApplier">
<list key="application_parameters">
</list>
</operator>
<operator name="BinominalClassificationPerformance" class="BinominalClassificationPerformance">
<parameter key="keep_example_set" value="true"/>
<parameter key="precision" value="true"/>
<parameter key="recall" value="true"/>
<parameter key="f_measure" value="true"/>
<parameter key="false_positive" value="true"/>
<parameter key="false_negative" value="true"/>
<parameter key="true_positive" value="true"/>
<parameter key="true_negative" value="true"/>
</operator>
</operator>
</operator>
</operator>
</operator>
