"Fossies" - the Fresh Open Source Software Archive

Member "rapidminer-studio-9.5.0/src/main/resources/com/rapidminer/resources/samples/processes/03_Validation/01_PerformanceEvaluator_Nominal.rmp" (7 Nov 2019, 6213 Bytes) of package /linux/misc/rapidminer-studio-9.5.0-src.tar.gz:


As a special service "Fossies" has tried to format the requested text file into HTML format (style: standard) with prefixed line numbers. Alternatively you can here view or download the uninterpreted source code file.

    1 <?xml version="1.0" encoding="UTF-8"?><process version="7.3.000-SNAPSHOT">
    2   <context>
    3     <input/>
    4     <output/>
    5     <macros/>
    6   </context>
    7   <operator activated="true" class="process" compatibility="7.3.000-SNAPSHOT" expanded="true" name="Root">
    8     <parameter key="logverbosity" value="init"/>
    9     <parameter key="random_seed" value="2001"/>
   10     <parameter key="send_mail" value="never"/>
   11     <parameter key="notification_email" value=""/>
   12     <parameter key="process_duration_for_mail" value="30"/>
   13     <parameter key="encoding" value="SYSTEM"/>
   14     <process expanded="true">
   15       <operator activated="true" class="retrieve" compatibility="7.3.000-SNAPSHOT" expanded="true" height="68" name="Retrieve" width="90" x="45" y="34">
   16         <parameter key="repository_entry" value="../../data/Labor-Negotiations"/>
   17       </operator>
   18       <operator activated="true" class="replace_missing_values" compatibility="7.1.001" expanded="true" height="103" name="MissingValueReplenishment" width="90" x="179" y="34">
   19         <parameter key="return_preprocessing_model" value="false"/>
   20         <parameter key="create_view" value="false"/>
   21         <parameter key="attribute_filter_type" value="all"/>
   22         <parameter key="attribute" value=""/>
   23         <parameter key="attributes" value=""/>
   24         <parameter key="use_except_expression" value="false"/>
   25         <parameter key="value_type" value="attribute_value"/>
   26         <parameter key="use_value_type_exception" value="false"/>
   27         <parameter key="except_value_type" value="time"/>
   28         <parameter key="block_type" value="attribute_block"/>
   29         <parameter key="use_block_type_exception" value="false"/>
   30         <parameter key="except_block_type" value="value_matrix_row_start"/>
   31         <parameter key="invert_selection" value="false"/>
   32         <parameter key="include_special_attributes" value="false"/>
   33         <parameter key="default" value="average"/>
   34         <list key="columns"/>
   35       </operator>
   36       <operator activated="true" class="decision_stump" compatibility="7.3.000-SNAPSHOT" expanded="true" height="82" name="DecisionStump" width="90" x="313" y="34">
   37         <parameter key="criterion" value="information_gain"/>
   38         <parameter key="minimal_leaf_size" value="1"/>
   39       </operator>
   40       <operator activated="true" class="apply_model" compatibility="7.1.001" expanded="true" height="82" name="ModelApplier" width="90" x="447" y="34">
   41         <list key="application_parameters"/>
   42         <parameter key="create_view" value="false"/>
   43       </operator>
   44       <operator activated="true" class="performance_classification" compatibility="7.3.000-SNAPSHOT" expanded="true" height="82" name="ClassificationPerformance" width="90" x="581" y="34">
   45         <parameter key="main_criterion" value="first"/>
   46         <parameter key="accuracy" value="true"/>
   47         <parameter key="classification_error" value="true"/>
   48         <parameter key="kappa" value="false"/>
   49         <parameter key="weighted_mean_recall" value="false"/>
   50         <parameter key="weighted_mean_precision" value="false"/>
   51         <parameter key="spearman_rho" value="false"/>
   52         <parameter key="kendall_tau" value="false"/>
   53         <parameter key="absolute_error" value="false"/>
   54         <parameter key="relative_error" value="false"/>
   55         <parameter key="relative_error_lenient" value="false"/>
   56         <parameter key="relative_error_strict" value="false"/>
   57         <parameter key="normalized_absolute_error" value="false"/>
   58         <parameter key="root_mean_squared_error" value="true"/>
   59         <parameter key="root_relative_squared_error" value="false"/>
   60         <parameter key="squared_error" value="false"/>
   61         <parameter key="correlation" value="false"/>
   62         <parameter key="squared_correlation" value="false"/>
   63         <parameter key="cross-entropy" value="false"/>
   64         <parameter key="margin" value="false"/>
   65         <parameter key="soft_margin_loss" value="false"/>
   66         <parameter key="logistic_loss" value="false"/>
   67         <parameter key="skip_undefined_labels" value="true"/>
   68         <parameter key="use_example_weights" value="true"/>
   69         <list key="class_weights"/>
   70       </operator>
   71       <connect from_op="Retrieve" from_port="output" to_op="MissingValueReplenishment" to_port="example set input"/>
   72       <connect from_op="MissingValueReplenishment" from_port="example set output" to_op="DecisionStump" to_port="training set"/>
   73       <connect from_op="DecisionStump" from_port="model" to_op="ModelApplier" to_port="model"/>
   74       <connect from_op="DecisionStump" from_port="exampleSet" to_op="ModelApplier" to_port="unlabelled data"/>
   75       <connect from_op="ModelApplier" from_port="labelled data" to_op="ClassificationPerformance" to_port="labelled data"/>
   76       <connect from_op="ClassificationPerformance" from_port="performance" to_port="result 1"/>
   77       <portSpacing port="source_input 1" spacing="0"/>
   78       <portSpacing port="sink_result 1" spacing="0"/>
   79       <portSpacing port="sink_result 2" spacing="0"/>
   80       <description align="left" color="yellow" colored="false" height="345" resized="true" width="469" x="43" y="150">This process introduces another very important operator for almost all RapidMiner processes: the PerformanceEvaluator operator.&lt;br/&gt;&lt;br/&gt;This operator expects an input example set with both a true and a predicted label attribute. The ClassificationPerformance calculates one or more performance measures (performance criteria) and delivers them as PerformanceVector. &lt;br/&gt;&lt;br/&gt;Such a performance vector can be used as estimation of a learner prediction accuracy (in this process it simply calculates the training error which usually isn't a good idea) or as fitness for an outer optimization scheme. &lt;br/&gt;&lt;br/&gt;Please note that RapidMiner provides a huge set of performance criteria which are suitable for classification tasks. Beside the well-known classification criteria like accuracy or precision all regression criteria like RMS can also be calculated for classification tasks. In this case the difference between 1 and the confidence for the desired true label value is used as example deviation.</description>
   81     </process>
   82   </operator>
   83 </process>