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
"How do you apply Platt Scaling in X-Validation?"
Contributor II
Hello all,
I am having trouble using the Rescale Confidences operator. I looked over the sample file, and it seems straight forward,, but I can't figure out how to apply it in an X-Validation.
This is the best I can come up with. You can see that the labeled data which is output from the Apply Model operator does not contain confidence values -- therefore the Performance operator fails. I am new to data mining, so perhaps my thinking is way off the mark. (I used Rapid Miner 5)
I have read over Steffen's posts on this subject, but I am afraid I still can't figure it out. Any help would be much appreciated.
Regards, Charles
I am having trouble using the Rescale Confidences operator. I looked over the sample file, and it seems straight forward,, but I can't figure out how to apply it in an X-Validation.
This is the best I can come up with. You can see that the labeled data which is output from the Apply Model operator does not contain confidence values -- therefore the Performance operator fails. I am new to data mining, so perhaps my thinking is way off the mark. (I used Rapid Miner 5)
<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<process version="5.0">
<context>
<input/>
<output/>
<macros/>
</context>
<operator activated="true" class="process" expanded="true" name="Root">
<process expanded="true" height="395" width="748">
<operator activated="true" class="generate_data" expanded="true" height="60" name="ExampleSetGenerator" width="90" x="45" y="30">
<parameter key="target_function" value="checkerboard classification"/>
<parameter key="number_examples" value="500"/>
<parameter key="number_of_attributes" value="2"/>
</operator>
<operator activated="true" class="x_validation" expanded="true" height="112" name="Validation" width="90" x="246" y="30">
<process expanded="true" height="414" width="433">
<operator activated="true" class="support_vector_machine" expanded="true" height="112" name="JMySVMLearner" width="90" x="45" y="30">
<parameter key="kernel_type" value="radial"/>
</operator>
<operator activated="true" class="rescale_confidences" expanded="true" height="76" name="PlattScaling" width="90" x="179" y="30"/>
<connect from_port="training" to_op="JMySVMLearner" to_port="training set"/>
<connect from_op="JMySVMLearner" from_port="model" to_op="PlattScaling" to_port="prediction model"/>
<connect from_op="JMySVMLearner" from_port="exampleSet" to_op="PlattScaling" to_port="example set"/>
<connect from_op="PlattScaling" 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="414" width="435">
<operator activated="true" class="apply_model" expanded="true" height="76" name="Apply Model" width="90" x="45" y="30">
<list key="application_parameters"/>
</operator>
<operator activated="true" class="rescale_confidences" expanded="true" height="76" name="Rescale Confidences" width="90" x="179" y="30"/>
<operator activated="true" class="performance" expanded="true" height="76" name="Performance" width="90" x="313" y="30"/>
<connect from_port="model" to_op="Apply Model" to_port="model"/>
<connect from_port="test set" to_op="Apply Model" to_port="unlabelled data"/>
<connect from_op="Apply Model" from_port="labelled data" to_op="Rescale Confidences" to_port="example set"/>
<connect from_op="Apply Model" from_port="model" to_op="Rescale Confidences" to_port="prediction model"/>
<connect from_op="Rescale Confidences" from_port="example set" 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"/>
<portSpacing port="sink_averagable 3" spacing="0"/>
</process>
</operator>
<connect from_op="ExampleSetGenerator" from_port="output" to_op="Validation" to_port="training"/>
<connect from_op="Validation" from_port="model" to_port="result 1"/>
<connect from_op="Validation" from_port="averagable 1" 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>
I have read over Steffen's posts on this subject, but I am afraid I still can't figure it out. Any help would be much appreciated.
Regards, Charles
Tagged:
0
Answers
just delete the PlatScaling. The second one will do the trick. PlatScaling internally applies the model, but does not alter it. Here's how it works: Greetings,
Sebastian
Thanks so much for the clear - and very quick - reply. I had not realized that the Platt scaling operator replaced the need for the model applier. As usual, I understand the complex eventually... the obvious takes me a little longer.
Unfortunately, the configuration your offered gives me the same wonky output as my original process. However, at least I know that the problem lies somewhere other than with the scaling. I will try running it again with a larger sample. Thanks again for sharing your expertise. It probably saved me numerous hours of futile experimentation. Have a great day.
Regards, Charles