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
"Relevance Vector Machine problem"
stealthgate
Member Posts: 1 
Learner III
Hello, I'm a new user and currently I'm trying the relevance vector machine in rapidMiner.
I loaded my data from a csv file successfully, and I'm trying to do a binary classification task using RVM. I try it with just about ~50 data with 11 features and 2 labels. I have undergo the training process, but in the final result view I can't view the final weights (there's a message that the data can't be visualized). And also based on the reference papers, afaik there's also a correlation between my data index, the weights, the alpha values and the relevance vectors. The output did not give the index of data which becomes the relevance vectors! My question is, are these a problem with the RVM implementation in rapidMiner or maybe I do it in the wrong way? Thank you in advance,
Regards,
I loaded my data from a csv file successfully, and I'm trying to do a binary classification task using RVM. I try it with just about ~50 data with 11 features and 2 labels. I have undergo the training process, but in the final result view I can't view the final weights (there's a message that the data can't be visualized). And also based on the reference papers, afaik there's also a correlation between my data index, the weights, the alpha values and the relevance vectors. The output did not give the index of data which becomes the relevance vectors! My question is, are these a problem with the RVM implementation in rapidMiner or maybe I do it in the wrong way? Thank you in advance,
Regards,
Tagged:
1
Answers
to be honest: I don't know. The relevance vector machine is a brilliant idea, but it's computational so complex. that it renders the algorithm unusable in the practice.
Please send the process, I will take a look at that.
Greetings,
Sebastian
<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<process version="5.1.001">
<context>
<input/>
<output/>
<macros/>
</context>
<operator activated="true" class="process" compatibility="5.1.001" expanded="true" name="Process">
<process expanded="true" height="296" width="872">
<operator activated="true" class="generate_data" compatibility="5.1.001" expanded="true" height="60" name="Generate Data" width="90" x="514" y="30">
<parameter key="target_function" value="complicated function"/>
<parameter key="number_examples" value="1000"/>
<parameter key="number_of_attributes" value="12"/>
</operator>
<operator activated="true" class="relevance_vector_machine" compatibility="5.1.001" expanded="true" height="76" name="Relevance Vector Machine" width="90" x="648" y="30">
<parameter key="rvm_type" value="Constructive-Regression-RVM"/>
<parameter key="kernel_type" value="sigmoid"/>
</operator>
<connect from_op="Generate Data" from_port="output" to_op="Relevance Vector Machine" to_port="training set"/>
<connect from_op="Relevance Vector Machine" from_port="model" to_port="result 1"/>
<portSpacing port="source_input 1" spacing="0"/>
<portSpacing port="sink_result 1" spacing="0"/>
<portSpacing port="sink_result 2" spacing="0"/>
</process>
</operator>
</process>
yes, theres a bug in the operator. Don't know how long it was hidden there. Please file a bug at our bugtracker, if not already added.
Greetings,
Sebastian