Options

"SVM generates same value for output"

billsbills Member Posts: 3 Contributor I
edited June 2019 in Help
Hello

I'm trying to cluster data using the SVMCluster model from Samples/Processes/07 but only get 1 cluster from my data.  I scaled the data to percentages and want to use a windowing process to examine some customer history then predict future buying.

I changed the model to LibSVMLearner and used the genetic parameter optimizer to try to find acceptable parameters.  The output (prediction) is the same value for all examples.  When I use your SVMclustering process with LibSVM it does generate different prediction values.

Can you explain what I am doing wrong?  Do I need to use the original data instead of percentages?

Here are a modified RM sample that generates different prediction values (for simplicity I used the same data for training and testing just to get some results), my process that generates the same prediction value and some sample data.  My process also takes a very long time to read the model after training but the model file is only 35K.  Is there something wrong with model storage that could cause the problem?

(This post may apply      http://rapid-i.com/rapidforum/index.php?PHPSESSID=4f6c22a9df5bbb0be3e6e41d1eb2c119&topic=490.0)

Thank you.

Data:  
https://www.dropbox.com/s/fs7up70iyr0ugbz/rapidsvm.csv

description and two process examples
https://www.dropbox.com/s/nrntxs397uun95m/rapidproblem.txt

(My process exceeds the 20k limit for posts)

simple example adapted from 07_EvolutionaryParameterOptimization 

<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<process version="5.3.005">
  <context>
    <input/>
    <output/>
    <macros/>
  </context>
  <operator activated="true" class="process" compatibility="5.3.005" expanded="true" name="Root">
    <description>&lt;p&gt;This process is also a parameter optimization process like the first one discussed in the meta group. In this case, 

an evolutionary approach is used for the search of the best parameter combination. This approach is often more appropriate and leads to better 

results without defining the parameter combinations which should be tested (as for the Grid Search and  the quadratic parameter optimization 

approaches).&lt;/p&gt; &lt;p&gt;The parameters for the evolutionary parameter optimization approach are defined in the same way as for the 

other parameter optimization operators. Instead of a comma separated list of parameters which should be checked the user has to define a colon 

separated pair which is used as lower and upper bound for the specific parameters.&lt;/p&gt; </description>
    <process expanded="true">
      <operator activated="true" class="retrieve" compatibility="5.3.005" expanded="true" height="60" name="Retrieve" width="90" x="45" y="30">
        <parameter key="repository_entry" value="//Samples/data/Polynomial"/>
      </operator>
      <operator activated="true" class="multiply" compatibility="5.3.005" expanded="true" height="94" name="Multiply" width="90" x="179" 

y="30"/>
      <operator activated="true" class="optimize_parameters_evolutionary" compatibility="5.3.005" expanded="true" height="94" 

name="ParameterOptimization" width="90" x="380" y="30">
        <list key="parameters">
          <parameter key="LibSVMLearner.C" value="[0.0;10000]"/>
        </list>
        <parameter key="max_generations" value="10"/>
        <parameter key="tournament_fraction" value="0.75"/>
        <parameter key="crossover_prob" value="1.0"/>
        <process expanded="true">
          <operator activated="true" class="loop_and_average" compatibility="5.3.005" expanded="true" height="76" 

name="IteratingPerformanceAverage" width="90" x="45" y="30">
            <parameter key="iterations" value="3"/>
            <process expanded="true">
              <operator activated="true" class="x_validation" compatibility="5.3.005" expanded="true" height="112" name="Validation" width="90" 

x="112" y="30">
                <parameter key="number_of_validations" value="2"/>
                <parameter key="sampling_type" value="shuffled sampling"/>
                <process expanded="true">
                  <operator activated="true" class="support_vector_machine_libsvm" compatibility="5.3.005" expanded="true" height="76" 

name="LibSVMLearner" width="90" x="112" y="30">
                    <parameter key="svm_type" value="epsilon-SVR"/>
                    <parameter key="degree" value="40"/>
                    <parameter key="gamma" value="10.0"/>
                    <parameter key="C" value="2205.039608219743"/>
                    <parameter key="epsilon" value="1.0"/>
                    <list key="class_weights"/>
                  </operator>
                  <connect from_port="training" to_op="LibSVMLearner" to_port="training set"/>
                  <connect from_op="LibSVMLearner" from_port="model" to_port="model"/>
                  <portSpacing port="source_training" spacing="0"/>
                  <portSpacing port="sink_model" spacing="0"/>
                  <portSpacing port="sink_through 1" spacing="0"/>
                </process>
                <process expanded="true">
                  <operator activated="true" class="apply_model" compatibility="5.3.005" expanded="true" height="76" name="Test" width="90" 

x="45" y="30">
                    <list key="application_parameters"/>
                  </operator>
                  <operator activated="true" class="performance" compatibility="5.3.005" expanded="true" height="76" name="Performance" 

width="90" x="227" y="30"/>
                  <connect from_port="model" to_op="Test" to_port="model"/>
                  <connect from_port="test set" to_op="Test" to_port="unlabelled data"/>
                  <connect from_op="Test" from_port="labelled data" to_op="Performance" to_port="labelled data"/>
                  <connect from_op="Performance" from_port="performance" to_port="averagable 1"/>
                  <portSpacing port="source_model" spacing="0"/>
                  <portSpacing port="source_test set" spacing="0"/>
                  <portSpacing port="source_through 1" spacing="0"/>
                  <portSpacing port="sink_averagable 1" spacing="0"/>
                  <portSpacing port="sink_averagable 2" spacing="0"/>
                </process>
              </operator>
              <operator activated="true" class="write_model" compatibility="5.3.005" expanded="true" height="60" name="Write Model" width="90" 

x="380" y="120">
                <parameter key="model_file" value="c:\rapid\test.mdl"/>
                <parameter key="output_type" value="Binary"/>
              </operator>
              <connect from_port="in 1" to_op="Validation" to_port="training"/>
              <connect from_op="Validation" from_port="model" to_op="Write Model" to_port="input"/>
              <connect from_op="Validation" from_port="averagable 1" to_port="averagable 1"/>
              <portSpacing port="source_in 1" spacing="0"/>
              <portSpacing port="source_in 2" spacing="0"/>
              <portSpacing port="sink_averagable 1" spacing="0"/>
              <portSpacing port="sink_averagable 2" spacing="0"/>
            </process>
          </operator>
          <operator activated="true" class="log" compatibility="5.3.005" expanded="true" height="76" name="Log" width="90" x="425" y="30">
            <parameter key="filename" value="paraopt.log"/>
            <list key="log">
              <parameter key="C" value="operator.LibSVMLearner.parameter.C"/>
              <parameter key="degree" value="operator.LibSVMLearner.parameter.degree"/>
              <parameter key="error" value="operator.IteratingPerformanceAverage.value.performance"/>
            </list>
          </operator>
          <connect from_port="input 1" to_op="IteratingPerformanceAverage" to_port="in 1"/>
          <connect from_op="IteratingPerformanceAverage" from_port="averagable 1" to_op="Log" to_port="through 1"/>
          <connect from_op="Log" from_port="through 1" to_port="performance"/>
          <portSpacing port="source_input 1" spacing="0"/>
          <portSpacing port="source_input 2" spacing="0"/>
          <portSpacing port="sink_performance" spacing="0"/>
          <portSpacing port="sink_result 1" spacing="0"/>
        </process>
      </operator>
      <operator activated="true" class="read_model" compatibility="5.3.005" expanded="true" height="60" name="Read Model" width="90" x="179" 

y="255">
        <parameter key="model_file" value="c:\rapid\test.mdl"/>
      </operator>
      <operator activated="true" class="apply_model" compatibility="5.3.005" expanded="true" height="76" name="Apply Model" width="90" x="447" 

y="210">
        <list key="application_parameters"/>
      </operator>
      <connect from_op="Retrieve" from_port="output" to_op="Multiply" to_port="input"/>
      <connect from_op="Multiply" from_port="output 1" to_op="ParameterOptimization" to_port="input 1"/>
      <connect from_op="Multiply" from_port="output 2" to_op="Apply Model" to_port="unlabelled data"/>
      <connect from_op="ParameterOptimization" from_port="performance" to_port="result 1"/>
      <connect from_op="ParameterOptimization" from_port="parameter" to_port="result 2"/>
      <connect from_op="Read Model" from_port="output" to_op="Apply Model" to_port="model"/>
      <connect from_op="Apply Model" from_port="labelled data" to_port="result 3"/>
      <connect from_op="Apply Model" from_port="model" to_port="result 4"/>
      <portSpacing port="source_input 1" spacing="0"/>
      <portSpacing port="sink_result 1" spacing="0"/>
      <portSpacing port="sink_result 2" spacing="0"/>
      <portSpacing port="sink_result 3" spacing="0"/>
      <portSpacing port="sink_result 4" spacing="0"/>
      <portSpacing port="sink_result 5" spacing="0"/>
    </process>
  </operator>
</process>

[ /code]
Tagged:

Answers

  • Options
    MariusHelfMariusHelf RapidMiner Certified Expert, Member Posts: 1,869 Unicorn
    Hi,

    you have two errors in your process setup, plus some less-then-optimal operators choices:

    1. you are using an SVM with rbf kernel, that means in addition to C you have to optimize the gamma operators
    2. you are writing the model for each iteration of the parameter optimization. That means that in the end the file contains the model of the *last* parameter combination, not of the *best* one

    Those were the errors. Below some more hints:
    3. To optimize only one or two parameters in a well-defined range you should not use the evolutionary optimization, but grid optimization. Try values for both C and gamma between 1e-6 to 100 on a logarithmic scale.
    4. Read/Write Model is deprecated. You should not write to model to disk with those operators, but use Store and Retrieve to write/read the model to the repository.

    To work around point 2, leave away the Loop and Average operator, and use more iterations for the X-Validation instead. Connect its model output to the result (res) output of the optimization operator. That makes sure that in the end the best model will be returned.
    For reference have a look at the attached process.

    Best regards,
    Marius 
    <?xml version="1.0" encoding="UTF-8" standalone="no"?>
    <process version="5.3.007">
      <context>
        <input/>
        <output/>
        <macros/>
      </context>
      <operator activated="true" class="process" compatibility="5.3.007" expanded="true" name="Root">
        <description>&lt;p&gt;This process is also a parameter optimization process like the first one discussed in the meta group. In this case, 

    an evolutionary approach is used for the search of the best parameter combination. This approach is often more appropriate and leads to better 

    results without defining the parameter combinations which should be tested (as for the Grid Search and  the quadratic parameter optimization 

    approaches).&lt;/p&gt; &lt;p&gt;The parameters for the evolutionary parameter optimization approach are defined in the same way as for the 

    other parameter optimization operators. Instead of a comma separated list of parameters which should be checked the user has to define a colon 

    separated pair which is used as lower and upper bound for the specific parameters.&lt;/p&gt; </description>
        <process expanded="true">
          <operator activated="true" class="retrieve" compatibility="5.3.007" expanded="true" height="60" name="Retrieve" width="90" x="45" y="30">
            <parameter key="repository_entry" value="//Samples/data/Polynomial"/>
          </operator>
          <operator activated="true" class="multiply" compatibility="5.3.007" expanded="true" height="94" name="Multiply" width="90" x="179" y="30"/>
          <operator activated="true" class="optimize_parameters_grid" compatibility="5.3.007" expanded="true" height="112" name="Optimize Parameters (Grid)" width="90" x="447" y="30">
            <list key="parameters">
              <parameter key="LibSVMLearner.C" value="[0.0001;1;5;logarithmic]"/>
              <parameter key="LibSVMLearner.gamma" value="[0.0001;10;5;logarithmic]"/>
            </list>
            <process expanded="true">
              <operator activated="true" class="x_validation" compatibility="5.3.007" expanded="true" height="112" name="Validation" width="90" x="45" y="30">
                <parameter key="number_of_validations" value="5"/>
                <parameter key="sampling_type" value="shuffled sampling"/>
                <process expanded="true">
                  <operator activated="true" class="support_vector_machine_libsvm" compatibility="5.3.007" expanded="true" height="76" name="LibSVMLearner" width="90" x="174" y="30">
                    <parameter key="svm_type" value="epsilon-SVR"/>
                    <parameter key="degree" value="40"/>
                    <parameter key="gamma" value="1.0E-4"/>
                    <parameter key="C" value="1.0E-4"/>
                    <parameter key="epsilon" value="1.0"/>
                    <list key="class_weights"/>
                  </operator>
                  <connect from_port="training" to_op="LibSVMLearner" to_port="training set"/>
                  <connect from_op="LibSVMLearner" from_port="model" to_port="model"/>
                  <portSpacing port="source_training" spacing="0"/>
                  <portSpacing port="sink_model" spacing="0"/>
                  <portSpacing port="sink_through 1" spacing="0"/>
                </process>
                <process expanded="true">
                  <operator activated="true" class="apply_model" compatibility="5.3.007" expanded="true" height="76" name="Test" width="90" x="45" y="30">
                    <list key="application_parameters"/>
                  </operator>
                  <operator activated="true" class="performance" compatibility="5.3.007" expanded="true" height="76" name="Performance" width="90" x="278" y="30"/>
                  <connect from_port="model" to_op="Test" to_port="model"/>
                  <connect from_port="test set" to_op="Test" to_port="unlabelled data"/>
                  <connect from_op="Test" from_port="labelled data" to_op="Performance" to_port="labelled data"/>
                  <connect from_op="Performance" from_port="performance" to_port="averagable 1"/>
                  <portSpacing port="source_model" spacing="0"/>
                  <portSpacing port="source_test set" spacing="0"/>
                  <portSpacing port="source_through 1" spacing="0"/>
                  <portSpacing port="sink_averagable 1" spacing="0"/>
                  <portSpacing port="sink_averagable 2" spacing="0"/>
                </process>
              </operator>
              <operator activated="true" class="log" compatibility="5.3.007" expanded="true" height="76" name="Log" width="90" x="447" y="30">
                <parameter key="filename" value="paraopt.log"/>
                <list key="log">
                  <parameter key="C" value="operator.LibSVMLearner.parameter.C"/>
                  <parameter key="degree" value="operator.LibSVMLearner.parameter.degree"/>
                  <parameter key="error" value="operator.Validation.value.performance"/>
                  <parameter key="gamma" value="operator.LibSVMLearner.parameter.gamma"/>
                </list>
              </operator>
              <operator activated="true" class="write_model" compatibility="5.3.007" expanded="true" height="60" name="Write Model" width="90" x="447" y="120">
                <parameter key="model_file" value="c:\rapid\test.mdl"/>
                <parameter key="output_type" value="Binary"/>
              </operator>
              <connect from_port="input 1" to_op="Validation" to_port="training"/>
              <connect from_op="Validation" from_port="model" to_op="Write Model" to_port="input"/>
              <connect from_op="Validation" from_port="averagable 1" to_op="Log" to_port="through 1"/>
              <connect from_op="Log" from_port="through 1" to_port="performance"/>
              <connect from_op="Write Model" from_port="through" to_port="result 1"/>
              <portSpacing port="source_input 1" spacing="0"/>
              <portSpacing port="source_input 2" spacing="0"/>
              <portSpacing port="sink_performance" spacing="0"/>
              <portSpacing port="sink_result 1" spacing="0"/>
              <portSpacing port="sink_result 2" spacing="0"/>
            </process>
          </operator>
          <operator activated="false" class="read_model" compatibility="5.3.007" expanded="true" height="60" name="Read Model" width="90" x="179" y="255">
            <parameter key="model_file" value="c:\rapid\test.mdl"/>
          </operator>
          <operator activated="true" class="apply_model" compatibility="5.3.007" expanded="true" height="76" name="Apply Model" width="90" x="581" y="210">
            <list key="application_parameters"/>
          </operator>
          <connect from_op="Retrieve" from_port="output" to_op="Multiply" to_port="input"/>
          <connect from_op="Multiply" from_port="output 1" to_op="Apply Model" to_port="unlabelled data"/>
          <connect from_op="Multiply" from_port="output 2" to_op="Optimize Parameters (Grid)" to_port="input 1"/>
          <connect from_op="Optimize Parameters (Grid)" from_port="performance" to_port="result 4"/>
          <connect from_op="Optimize Parameters (Grid)" from_port="parameter" to_port="result 3"/>
          <connect from_op="Optimize Parameters (Grid)" from_port="result 1" to_op="Apply Model" to_port="model"/>
          <connect from_op="Apply Model" from_port="labelled data" to_port="result 1"/>
          <connect from_op="Apply Model" from_port="model" to_port="result 2"/>
          <portSpacing port="source_input 1" spacing="0"/>
          <portSpacing port="sink_result 1" spacing="0"/>
          <portSpacing port="sink_result 2" spacing="0"/>
          <portSpacing port="sink_result 3" spacing="0"/>
          <portSpacing port="sink_result 4" spacing="0"/>
          <portSpacing port="sink_result 5" spacing="0"/>
        </process>
      </operator>
    </process>
  • Options
    billsbills Member Posts: 3 Contributor I
    Marius,
    Thanks for your suggestions.  I've updated my process based on your sample.  I still have one value as the result for all examples.

    label rownumber prediction(label) attrib1-attrib7
    0.02 100.0 -0.010000000000000002 -0.00445 0.3857 -0.00159 0.017248 0.017248 -1.531244 -0.073051
    -0.02 101.0 -0.010000000000000002 0.023435 0.8366 0.009283 0.021341 0.026825 -1.970992 -0.096819
    -0.01 102.0 -0.010000000000000002 0.001783 0.9629 -2.64E-4 0.010065 0.010065 -1.028535 0.046995
    -0.03 103.0 -0.010000000000000002 0.001799 0.9545 -0.006372 0.011877 0.011877 -0.482728 -0.041898
    0.0 104.0 -0.010000000000000002 0.017011 0.802 -0.026941 0.057796 0.057796 2.960633 0.168071

    There's also an error code from running this process:
    SEVERE: There is more than one renderable candidate for the result of com.rapidminer.operator.learner.functions.kernel.LibSVMModel

    This error also appears when I run your code.

    What else can I try to fix this?  



    <?xml version="1.0" encoding="UTF-8" standalone="no"?>
    <process version="5.3.005">
      <context>
        <input/>
        <output/>
        <macros/>
      </context>
      <operator activated="true" class="process" compatibility="5.3.005" expanded="true" name="Root">
        <description>&lt;p&gt;This process is also a parameter optimization process like the first one discussed in the meta group. In this case, an evolutionary approach is used for the search of the best parameter combination. This approach is often more appropriate and leads to better results without defining the parameter combinations which should be tested (as for the Grid Search and  the quadratic parameter optimization approaches).&lt;/p&gt; &lt;p&gt;The parameters for the evolutionary parameter optimization approach are defined in the same way as for the other parameter optimization operators. Instead of a comma separated list of parameters which should be checked the user has to define a colon separated pair which is used as lower and upper bound for the specific parameters.&lt;/p&gt; </description>
        <parameter key="parallelize_main_process" value="true"/>
        <process expanded="true">
          <operator activated="true" class="read_csv" compatibility="5.3.005" expanded="true" height="60" name="Read CSV" width="90" x="45" y="30">
            <parameter key="csv_file" value="C:\rapid\rapidsvm.csv"/>
            <parameter key="column_separators" value=","/>
            <parameter key="first_row_as_names" value="false"/>
            <list key="annotations">
              <parameter key="0" value="Name"/>
            </list>
            <parameter key="encoding" value="windows-1252"/>
            <list key="data_set_meta_data_information">
              <parameter key="0" value="custid.true.integer.attribute"/>
              <parameter key="1" value="attrib1.true.real.attribute"/>
              <parameter key="2" value="attrib2.true.real.attribute"/>
              <parameter key="3" value="attrib3.true.real.attribute"/>
              <parameter key="4" value="attrib4.true.real.attribute"/>
              <parameter key="5" value="attrib5.true.real.attribute"/>
              <parameter key="6" value="attrib6.true.real.attribute"/>
              <parameter key="7" value="attrib7.true.real.attribute"/>
              <parameter key="8" value="label.true.real.label"/>
              <parameter key="9" value="rownumber.true.integer.id"/>
            </list>
          </operator>
          <operator activated="true" class="select_attributes" compatibility="5.3.005" expanded="true" height="76" name="Select Attributes" width="90" x="45" y="120">
            <parameter key="attribute_filter_type" value="subset"/>
            <parameter key="attribute" value="custid"/>
            <parameter key="attributes" value="|attrib1|attrib2|attrib7|attrib6|attrib5|attrib4|attrib3|label|rownumber"/>
            <parameter key="include_special_attributes" value="true"/>
          </operator>
          <operator activated="true" class="filter_examples" compatibility="5.3.005" expanded="true" height="76" name="Filter Examples" width="90" x="179" y="75">
            <parameter key="condition_class" value="attribute_value_filter"/>
            <parameter key="parameter_string" value="rownumber&lt;100"/>
          </operator>
          <operator activated="true" class="multiply" compatibility="5.3.005" expanded="true" height="94" name="Multiply" width="90" x="380" y="75"/>
          <operator activated="true" class="support_vector_machine_libsvm" compatibility="5.3.005" expanded="true" height="76" name="SVMLearner" width="90" x="581" y="165">
            <parameter key="svm_type" value="epsilon-SVR"/>
            <parameter key="degree" value="1"/>
            <parameter key="gamma" value="100.0"/>
            <parameter key="C" value="0.0063095734448019355"/>
            <parameter key="epsilon" value="1.0"/>
            <list key="class_weights"/>
          </operator>
          <operator activated="true" class="optimize_parameters_grid" compatibility="5.3.005" expanded="true" height="94" name="Optimize Parameters (Grid)" width="90" x="581" y="30">
            <list key="parameters">
              <parameter key="SVMOptimize.C" value="[0.00001;100;5;logarithmic]"/>
              <parameter key="SVMOptimize.gamma" value="[0.0001;100;5;logarithmic]"/>
            </list>
            <parameter key="parallelize_optimization_process" value="true"/>
            <process expanded="true">
              <operator activated="true" class="x_validation" compatibility="5.3.005" expanded="true" height="112" name="Validation (2)" width="90" x="112" y="75">
                <parameter key="number_of_validations" value="2"/>
                <parameter key="sampling_type" value="shuffled sampling"/>
                <process expanded="true">
                  <operator activated="true" class="support_vector_machine_libsvm" compatibility="5.3.005" expanded="true" height="76" name="SVMOptimize" width="90" x="112" y="30">
                    <parameter key="svm_type" value="epsilon-SVR"/>
                    <parameter key="degree" value="100"/>
                    <parameter key="gamma" value="100.0"/>
                    <parameter key="C" value="100.00000000000001"/>
                    <parameter key="epsilon" value="1.0"/>
                    <list key="class_weights"/>
                  </operator>
                  <connect from_port="training" to_op="SVMOptimize" to_port="training set"/>
                  <connect from_op="SVMOptimize" from_port="model" to_port="model"/>
                  <portSpacing port="source_training" spacing="0"/>
                  <portSpacing port="sink_model" spacing="0"/>
                  <portSpacing port="sink_through 1" spacing="0"/>
                </process>
                <process expanded="true">
                  <operator activated="true" class="apply_model" compatibility="5.3.005" expanded="true" height="76" name="ApplyOptimize" width="90" x="45" y="30">
                    <list key="application_parameters"/>
                  </operator>
                  <operator activated="true" class="performance" compatibility="5.3.005" expanded="true" height="76" name="Performance (2)" width="90" x="209" y="30"/>
                  <connect from_port="model" to_op="ApplyOptimize" to_port="model"/>
                  <connect from_port="test set" to_op="ApplyOptimize" to_port="unlabelled data"/>
                  <connect from_op="ApplyOptimize" from_port="labelled data" to_op="Performance (2)" to_port="labelled data"/>
                  <connect from_op="Performance (2)" from_port="performance" to_port="averagable 1"/>
                  <portSpacing port="source_model" spacing="0"/>
                  <portSpacing port="source_test set" spacing="0"/>
                  <portSpacing port="source_through 1" spacing="0"/>
                  <portSpacing port="sink_averagable 1" spacing="0"/>
                  <portSpacing port="sink_averagable 2" spacing="0"/>
                </process>
              </operator>
              <connect from_port="input 1" to_op="Validation (2)" to_port="training"/>
              <connect from_op="Validation (2)" from_port="averagable 1" to_port="performance"/>
              <portSpacing port="source_input 1" spacing="0"/>
              <portSpacing port="source_input 2" spacing="0"/>
              <portSpacing port="sink_performance" spacing="0"/>
              <portSpacing port="sink_result 1" spacing="0"/>
            </process>
          </operator>
          <operator activated="true" class="set_parameters" compatibility="5.3.005" expanded="true" height="76" name="ParameterSetter" width="90" x="715" y="75">
            <list key="name_map">
              <parameter key="SVMOptimize" value="SVMLearner"/>
            </list>
          </operator>
          <operator activated="true" class="read_csv" compatibility="5.3.005" expanded="true" height="60" name="Read CSV (2)" width="90" x="112" y="300">
            <parameter key="csv_file" value="C:\rapid\rapidsvm.csv"/>
            <parameter key="column_separators" value=","/>
            <parameter key="first_row_as_names" value="false"/>
            <list key="annotations">
              <parameter key="0" value="Name"/>
            </list>
            <parameter key="encoding" value="windows-1252"/>
            <list key="data_set_meta_data_information">
              <parameter key="0" value="custid.true.integer.attribute"/>
              <parameter key="1" value="attrib1.true.real.attribute"/>
              <parameter key="2" value="attrib2.true.real.attribute"/>
              <parameter key="3" value="attrib3.true.real.attribute"/>
              <parameter key="4" value="attrib4.true.real.attribute"/>
              <parameter key="5" value="attrib5.true.real.attribute"/>
              <parameter key="6" value="attrib6.true.real.attribute"/>
              <parameter key="7" value="attrib7.true.real.attribute"/>
              <parameter key="8" value="label.true.real.label"/>
              <parameter key="9" value="rownumber.true.integer.id"/>
            </list>
          </operator>
          <operator activated="true" class="select_attributes" compatibility="5.3.005" expanded="true" height="76" name="Select Attributes (2)" width="90" x="246" y="300">
            <parameter key="attribute_filter_type" value="subset"/>
            <parameter key="attribute" value="custid"/>
            <parameter key="attributes" value="|rownumber|label|attrib7|attrib6|attrib5|attrib4|attrib3|attrib2|attrib1"/>
          </operator>
          <operator activated="true" class="filter_examples" compatibility="5.3.005" expanded="true" height="76" name="Filter Examples (2)" width="90" x="380" y="300">
            <parameter key="condition_class" value="attribute_value_filter"/>
            <parameter key="parameter_string" value="rownumber&gt;=100"/>
          </operator>
          <operator activated="true" class="apply_model" compatibility="5.3.005" expanded="true" height="76" name="Apply Model" width="90" x="648" y="300">
            <list key="application_parameters"/>
          </operator>
          <connect from_op="Read CSV" from_port="output" to_op="Select Attributes" to_port="example set input"/>
          <connect from_op="Select Attributes" from_port="example set output" to_op="Filter Examples" to_port="example set input"/>
          <connect from_op="Filter Examples" from_port="example set output" to_op="Multiply" to_port="input"/>
          <connect from_op="Multiply" from_port="output 1" to_op="Optimize Parameters (Grid)" to_port="input 1"/>
          <connect from_op="Multiply" from_port="output 2" to_op="SVMLearner" to_port="training set"/>
          <connect from_op="SVMLearner" from_port="model" to_op="Apply Model" to_port="model"/>
          <connect from_op="Optimize Parameters (Grid)" from_port="performance" to_port="result 3"/>
          <connect from_op="Optimize Parameters (Grid)" from_port="parameter" to_op="ParameterSetter" to_port="parameter set"/>
          <connect from_op="Read CSV (2)" from_port="output" to_op="Select Attributes (2)" to_port="example set input"/>
          <connect from_op="Select Attributes (2)" from_port="example set output" to_op="Filter Examples (2)" to_port="example set input"/>
          <connect from_op="Filter Examples (2)" from_port="example set output" to_op="Apply Model" to_port="unlabelled data"/>
          <connect from_op="Apply Model" from_port="labelled data" to_port="result 1"/>
          <connect from_op="Apply Model" from_port="model" to_port="result 2"/>
          <portSpacing port="source_input 1" spacing="0"/>
          <portSpacing port="sink_result 1" spacing="0"/>
          <portSpacing port="sink_result 2" spacing="0"/>
          <portSpacing port="sink_result 3" spacing="0"/>
          <portSpacing port="sink_result 4" spacing="0"/>
        </process>
      </operator>
    </process>


  • Options
    MariusHelfMariusHelf RapidMiner Certified Expert, Member Posts: 1,869 Unicorn
    Hi,

    the SVMLearner is executed before the Parameter Optimization. In this process setup, the connections alone do not define a unique process order, so you have to adjust it manually. To do so, enter process ordering mode by clicking the icon with the blue arrow and the question mark, and adjust the process order such that the optimization and the Parameter Setter are executed before the learner.

    Best regards,
    Marius
  • Options
    billsbills Member Posts: 3 Contributor I
    Marius,

    I updated the model with your suggestions but I still have the same output value for each entry.  I've created a simple process that trains a svm then uses the same data for forecasting.  It generates the same output value.

    I also created the same basic process in R and it generates output predictions close to the original values.

    The solution is something simple I'm sure but from searching the forum site for possible solutions and using the R parameters I still get the same value for all predictions in Rapidminer. 

    Any other ideas I can test?  Thanks 

    <?xml version="1.0" encoding="UTF-8" standalone="no"?>
    <process version="5.3.007">
      <context>
        <input/>
        <output/>
        <macros/>
      </context>
      <operator activated="true" class="process" compatibility="5.3.007" expanded="true" name="Root">
        <description>&lt;p&gt;This process is also a parameter optimization process like the first one discussed in the meta group. In this case, an evolutionary approach is used for the search of the best parameter combination. This approach is often more appropriate and leads to better results without defining the parameter combinations which should be tested (as for the Grid Search and  the quadratic parameter optimization approaches).&lt;/p&gt; &lt;p&gt;The parameters for the evolutionary parameter optimization approach are defined in the same way as for the other parameter optimization operators. Instead of a comma separated list of parameters which should be checked the user has to define a colon separated pair which is used as lower and upper bound for the specific parameters.&lt;/p&gt; </description>
        <parameter key="parallelize_main_process" value="true"/>
        <process expanded="true">
          <operator activated="true" class="read_csv" compatibility="5.3.007" expanded="true" height="60" name="Read CSV" width="90" x="45" y="30">
            <parameter key="csv_file" value="C:\rapid\rapidsvm.csv"/>
            <parameter key="column_separators" value=","/>
            <parameter key="first_row_as_names" value="false"/>
            <list key="annotations">
              <parameter key="0" value="Name"/>
            </list>
            <parameter key="encoding" value="windows-1252"/>
            <list key="data_set_meta_data_information">
              <parameter key="0" value="custid.false.integer.attribute"/>
              <parameter key="1" value="attrib1.true.real.attribute"/>
              <parameter key="2" value="attrib2.true.real.attribute"/>
              <parameter key="3" value="attrib3.true.real.attribute"/>
              <parameter key="4" value="attrib4.true.real.attribute"/>
              <parameter key="5" value="attrib5.true.real.attribute"/>
              <parameter key="6" value="attrib6.true.real.attribute"/>
              <parameter key="7" value="attrib7.true.real.attribute"/>
              <parameter key="8" value="label.true.real.label"/>
              <parameter key="9" value="rownumber.true.integer.id"/>
            </list>
          </operator>
          <operator activated="true" class="select_attributes" compatibility="5.3.007" expanded="true" height="76" name="Select Attributes" width="90" x="179" y="30">
            <parameter key="attribute_filter_type" value="subset"/>
            <parameter key="attribute" value="custid"/>
            <parameter key="attributes" value="label|rownumber|attrib1|attrib2|attrib3|attrib4|attrib5|attrib6|attrib7|"/>
            <parameter key="include_special_attributes" value="true"/>
          </operator>
          <operator activated="true" class="multiply" compatibility="5.3.007" expanded="true" height="94" name="Multiply" width="90" x="380" y="75"/>
          <operator activated="true" class="support_vector_machine_libsvm" compatibility="5.3.007" expanded="true" height="76" name="SVMLearner" width="90" x="581" y="165">
            <parameter key="svm_type" value="epsilon-SVR"/>
            <parameter key="degree" value="1"/>
            <parameter key="gamma" value="63.0"/>
            <parameter key="C" value="2.0"/>
            <parameter key="epsilon" value="1.0"/>
            <list key="class_weights"/>
          </operator>
          <operator activated="true" class="apply_model" compatibility="5.3.007" expanded="true" height="76" name="Apply Model" width="90" x="648" y="300">
            <list key="application_parameters"/>
          </operator>
          <connect from_op="Read CSV" from_port="output" to_op="Select Attributes" to_port="example set input"/>
          <connect from_op="Select Attributes" from_port="example set output" to_op="Multiply" to_port="input"/>
          <connect from_op="Multiply" from_port="output 1" to_op="SVMLearner" to_port="training set"/>
          <connect from_op="Multiply" from_port="output 2" to_op="Apply Model" to_port="unlabelled data"/>
          <connect from_op="SVMLearner" from_port="model" to_op="Apply Model" to_port="model"/>
          <connect from_op="Apply Model" from_port="labelled data" to_port="result 1"/>
          <connect from_op="Apply Model" from_port="model" to_port="result 2"/>
          <portSpacing port="source_input 1" spacing="0"/>
          <portSpacing port="sink_result 1" spacing="0"/>
          <portSpacing port="sink_result 2" spacing="0"/>
          <portSpacing port="sink_result 3" spacing="0"/>
        </process>
      </operator>
    </process>




    ## copy data file rapidsvm.csv to c:\rapid
    ##
    library(e1071)
    ## read and prepare data
    setwd('c:\\rapid')
    custdata <- read.table( 'rapidsvm.csv', header=T, dec='.', sep = ',', na.strings=c('XXXXXXX'))
    rownames(custdata) <- custdata$rownbr
    head(custdata)
    custdata$custid <- NULL
    custdata$rownumber <- NULL
    ## find best svm parameters
    ## e1071.pdf, pages 49-55
    tunedsvm <- tune.svm(label~., data = custdata, type='eps-regression', gamma = 2^(4:6), cost = 2^(1:2))
    summary(tunedsvm)
    tund.gamma <- tunedsvm$best.parameters[[1]]
    tund.cost <- tunedsvm$best.parameters[[2]]
    ## use best parameters in model
    svmmodel <- svm(label ~ ., data = custdata, type='eps-regression', gamma = tund.gamma, cost = tund.cost)
    ## compare original to svm predicted value
    pred <- fitted(svmmodel)
    compare <- as.data.frame(custdata[1:20,'label'])
    colnames(compare) <- 'orig_value'
    compare$svm_pred  <-  round(pred[1:20], digits = 3)
    compare


  • Options
    MariusHelfMariusHelf RapidMiner Certified Expert, Member Posts: 1,869 Unicorn
    Hm, how did you optimize the SVM? From which process do the values for C=2 and gamma=63? Definitely not from the process you posted above! Those values are not tested in the Optimize Parameters operator configured there.
    Furthermore, in the above process you use only a 2-Fold cross validation to optimize the parameters. Try to increase the number of folds for better results!

    Best regards,
    Marius
Sign In or Register to comment.