The Altair Community is migrating to a new platform to provide a better experience for you. In preparation for the migration, the Altair Community is on read-only mode from October 28 - November 6, 2024. Technical support via cases will continue to work as is. For any urgent requests from Students/Faculty members, please submit the form linked here
Problem by optimization and select subprocess operator
Contributor II
Hi,
in my process I use the parameter optimization in a select subprocess operator in order to deceide whether using the optimization of the parameters or not. Surprisingly the results in optimization widely differ from not using the select subprocess operator in combination with the optimization operator (Evolutionary). Have everybody an idea what was wrong or is it a bug?
Till
in my process I use the parameter optimization in a select subprocess operator in order to deceide whether using the optimization of the parameters or not. Surprisingly the results in optimization widely differ from not using the select subprocess operator in combination with the optimization operator (Evolutionary). Have everybody an idea what was wrong or is it a bug?
Regards
<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<process version="5.0">
<context>
<input>
<location/>
</input>
<output>
<location/>
<location/>
<location/>
<location/>
</output>
<macros/>
</context>
<operator activated="true" class="process" expanded="true" name="classificationOfWeeds">
<process expanded="true" height="581" width="835">
<operator activated="true" class="read_aml" expanded="true" height="60" name="loadFirstPrediction" width="90" x="715" y="120">
<parameter key="attributes" value="C:\Users\Till\Desktop\PredictionAllData.aml"/>
<parameter key="local_random_seed" value="-1"/>
</operator>
<operator activated="true" class="generate_data" expanded="true" height="60" name="Generate Data" width="90" x="45" y="30"/>
<operator activated="true" class="work_on_subset" expanded="true" height="76" name="Work on Subset" width="90" x="179" y="30">
<parameter key="attribute_filter_type" value="subset"/>
<parameter key="attributes" value="label"/>
<parameter key="include_special_attributes" value="true"/>
<process expanded="true" height="691" width="909">
<operator activated="true" class="numerical_to_binominal" expanded="true" height="76" name="Numerical to Binominal" width="90" x="380" y="30"/>
<connect from_port="exampleSet" to_op="Numerical to Binominal" to_port="example set input"/>
<connect from_op="Numerical to Binominal" from_port="example set output" to_port="example set"/>
<portSpacing port="source_exampleSet" spacing="0"/>
<portSpacing port="sink_example set" spacing="0"/>
<portSpacing port="sink_through 1" spacing="0"/>
</process>
</operator>
<operator activated="true" class="select_subprocess" expanded="true" height="94" name="firstClassification" width="90" x="313" y="30">
<process expanded="true" height="518" width="435">
<operator activated="true" class="multiply" expanded="true" height="94" name="Multiply" width="90" x="45" y="30"/>
<operator activated="true" class="optimize_parameters_evolutionary" expanded="true" height="130" name="Optimize Parameters (Evolutionary)" width="90" x="179" y="30">
<list key="parameters">
<parameter key="learnerTrainingFirst.C" value="[0.0;300]"/>
<parameter key="learnerTrainingFirst.gamma" value="[0.0;1]"/>
</list>
<process expanded="true" height="691" width="909">
<operator activated="true" class="x_validation" expanded="true" height="112" name="XValidation (7)" width="90" x="380" y="30">
<parameter key="create_complete_model" value="true"/>
<parameter key="local_random_seed" value="-1"/>
<process expanded="true" height="518" width="440">
<operator activated="true" class="support_vector_machine_libsvm" expanded="true" height="76" name="learnerTrainingFirst" width="90" x="179" y="30">
<parameter key="gamma" value="0.5"/>
<parameter key="C" value="120.0"/>
<parameter key="cache_size" value="1000"/>
<list key="class_weights"/>
<parameter key="calculate_confidences" value="true"/>
</operator>
<connect from_port="training" to_op="learnerTrainingFirst" to_port="training set"/>
<connect from_op="learnerTrainingFirst" 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" height="518" width="440">
<operator activated="true" class="apply_model" expanded="true" height="76" name="ModelApplier (6)" width="90" x="112" y="30">
<list key="application_parameters"/>
</operator>
<operator activated="true" class="performance_classification" expanded="true" height="76" name="trainingPerformance (2)" width="90" x="242" y="30">
<parameter key="main_criterion" value="kappa"/>
<parameter key="accuracy" value="true"/>
<parameter key="kappa" value="true"/>
<list key="class_weights">
<parameter key="GALAP" value="2.0"/>
<parameter key="CIRAR" value="2.0"/>
</list>
</operator>
<connect from_port="model" to_op="ModelApplier (6)" to_port="model"/>
<connect from_port="test set" to_op="ModelApplier (6)" to_port="unlabelled data"/>
<connect from_op="ModelApplier (6)" from_port="labelled data" to_op="trainingPerformance (2)" to_port="labelled data"/>
<connect from_op="trainingPerformance (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="XValidation (7)" to_port="training"/>
<connect from_op="XValidation (7)" from_port="model" to_port="result 1"/>
<connect from_op="XValidation (7)" from_port="training" to_port="result 2"/>
<connect from_op="XValidation (7)" 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"/>
<portSpacing port="sink_result 2" spacing="0"/>
<portSpacing port="sink_result 3" spacing="0"/>
</process>
</operator>
<operator activated="true" class="subprocess" expanded="true" height="94" name="applyOptimalParameter (2)" width="90" x="313" y="30">
<process expanded="true" height="500" width="835">
<operator activated="true" class="set_parameters" expanded="true" height="60" name="Set Parameters (2)" width="90" x="179" y="30">
<list key="name_map">
<parameter key="learnerTrainingFirst" value="optimalLearnerFirst"/>
</list>
</operator>
<operator activated="true" class="x_validation" expanded="true" height="112" name="XValidation (9)" width="90" x="179" y="120">
<parameter key="create_complete_model" value="true"/>
<parameter key="local_random_seed" value="-1"/>
<process expanded="true" height="518" width="440">
<operator activated="true" class="support_vector_machine_libsvm" expanded="true" height="76" name="optimalLearnerFirst" width="90" x="179" y="30">
<parameter key="gamma" value="0.3"/>
<parameter key="C" value="40.0"/>
<parameter key="cache_size" value="2000"/>
<list key="class_weights"/>
<parameter key="calculate_confidences" value="true"/>
<parameter key="confidence_for_multiclass" value="false"/>
</operator>
<connect from_port="training" to_op="optimalLearnerFirst" to_port="training set"/>
<connect from_op="optimalLearnerFirst" 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" height="518" width="440">
<operator activated="true" class="apply_model" expanded="true" height="76" name="ModelApplier (8)" width="90" x="45" y="30">
<list key="application_parameters"/>
</operator>
<operator activated="true" class="performance_classification" expanded="true" height="76" name="performanceFirst" width="90" x="179" y="30">
<parameter key="accuracy" value="true"/>
<parameter key="kappa" value="true"/>
<list key="class_weights"/>
</operator>
<operator activated="true" class="write_aml" expanded="true" height="60" name="write AML (3)" width="90" x="313" y="120">
<parameter key="example_set_file" value="C:\Users\Till\Desktop\PredictionAllData.dat"/>
<parameter key="attribute_description_file" value="C:\Users\Till\Desktop\PredictionAllData.aml"/>
</operator>
<connect from_port="model" to_op="ModelApplier (8)" to_port="model"/>
<connect from_port="test set" to_op="ModelApplier (8)" to_port="unlabelled data"/>
<connect from_op="ModelApplier (8)" from_port="labelled data" to_op="performanceFirst" to_port="labelled data"/>
<connect from_op="performanceFirst" from_port="performance" to_port="averagable 1"/>
<connect from_op="performanceFirst" from_port="example set" to_op="write AML (3)" to_port="input"/>
<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="in 1" to_op="Set Parameters (2)" to_port="parameter set"/>
<connect from_port="in 2" to_op="XValidation (9)" to_port="training"/>
<connect from_op="XValidation (9)" from_port="model" to_port="out 1"/>
<connect from_op="XValidation (9)" from_port="averagable 1" to_port="out 2"/>
<portSpacing port="source_in 1" spacing="0"/>
<portSpacing port="source_in 2" spacing="0"/>
<portSpacing port="source_in 3" spacing="0"/>
<portSpacing port="sink_out 1" spacing="90"/>
<portSpacing port="sink_out 2" spacing="18"/>
<portSpacing port="sink_out 3" spacing="0"/>
</process>
</operator>
<connect from_port="input 1" to_op="Multiply" to_port="input"/>
<connect from_op="Multiply" from_port="output 1" to_op="applyOptimalParameter (2)" to_port="in 2"/>
<connect from_op="Multiply" from_port="output 2" to_op="Optimize Parameters (Evolutionary)" to_port="input 1"/>
<connect from_op="Optimize Parameters (Evolutionary)" from_port="parameter" to_op="applyOptimalParameter (2)" to_port="in 1"/>
<connect from_op="applyOptimalParameter (2)" from_port="out 1" to_port="output 1"/>
<connect from_op="applyOptimalParameter (2)" from_port="out 2" to_port="output 2"/>
<portSpacing port="source_input 1" spacing="0"/>
<portSpacing port="source_input 2" spacing="0"/>
<portSpacing port="sink_output 1" spacing="0"/>
<portSpacing port="sink_output 2" spacing="0"/>
<portSpacing port="sink_output 3" spacing="0"/>
</process>
<process expanded="true" height="518" width="299">
<operator activated="true" class="x_validation" expanded="true" height="112" name="Validation (5)" width="90" x="112" y="30">
<process expanded="true" height="551" width="299">
<operator activated="true" class="support_vector_machine_libsvm" expanded="true" height="76" name="SVM" width="90" x="112" y="30">
<parameter key="gamma" value="0.15"/>
<parameter key="C" value="108.0"/>
<list key="class_weights"/>
</operator>
<connect from_port="training" to_op="SVM" to_port="training set"/>
<connect from_op="SVM" 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" height="551" width="480">
<operator activated="true" class="apply_model" expanded="true" height="76" name="Apply Model (5)" width="90" x="112" y="30">
<list key="application_parameters"/>
</operator>
<operator activated="true" class="performance_classification" expanded="true" height="76" name="performanceFirst (2)" width="90" x="246" y="30">
<parameter key="main_criterion" value="kappa"/>
<parameter key="accuracy" value="true"/>
<parameter key="kappa" value="true"/>
<list key="class_weights"/>
</operator>
<operator activated="true" class="write_aml" expanded="true" height="60" name="Write AML (3)" width="90" x="380" y="120">
<parameter key="example_set_file" value="C:\Users\Till\Desktop\PredictionAllData.dat"/>
<parameter key="attribute_description_file" value="C:\Users\Till\Desktop\PredictionAllData.aml"/>
</operator>
<connect from_port="model" to_op="Apply Model (5)" to_port="model"/>
<connect from_port="test set" to_op="Apply Model (5)" to_port="unlabelled data"/>
<connect from_op="Apply Model (5)" from_port="labelled data" to_op="performanceFirst (2)" to_port="labelled data"/>
<connect from_op="performanceFirst (2)" from_port="performance" to_port="averagable 1"/>
<connect from_op="performanceFirst (2)" from_port="example set" to_op="Write AML (3)" to_port="input"/>
<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 (5)" to_port="training"/>
<connect from_op="Validation (5)" from_port="model" to_port="output 1"/>
<connect from_op="Validation (5)" from_port="averagable 1" to_port="output 2"/>
<portSpacing port="source_input 1" spacing="0"/>
<portSpacing port="source_input 2" spacing="0"/>
<portSpacing port="sink_output 1" spacing="0"/>
<portSpacing port="sink_output 2" spacing="0"/>
<portSpacing port="sink_output 3" spacing="0"/>
</process>
</operator>
<connect from_op="loadFirstPrediction" from_port="output" to_port="result 3"/>
<connect from_op="Generate Data" from_port="output" to_op="Work on Subset" to_port="example set"/>
<connect from_op="Work on Subset" from_port="example set" to_op="firstClassification" to_port="input 1"/>
<connect from_op="firstClassification" from_port="output 1" to_port="result 1"/>
<connect from_op="firstClassification" from_port="output 2" 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>
Till
0
Answers
It could be useful to note that the numerical to binominal operator switches all the numerical values in the label attribute to "true".
if I understood you correctly, the problem is, that the results differ when you execute the optimization within a select subprocess operator from when executing it without?
Greetings,
Sebastian
Regards
Till
have you tried to use a local random seed on the optimization to avoid problems resulting from an altered random number sequence?
Greetings,
Sebastian