gradient boosted trees application

klugman

After I got a model, I want to apply it outside rapidminer - for example on BI model. Where can I find the formula that calculated the prediction?

varunm1

Hello @klugman

Do you want to extract gradient boosted tree predictions and create a visualization? 

If so you can extract predictions from the apply model operators. You can also store the trained model using the "Store" operator in rapidminer and retrieve this to apply on a new dataset for predictions in rapidminer. I don't think you can apply the trained model outside of rapidminer environment.

Gradient boosted trees are additive models and there won't be a single equation that can be extracted from the model.

varunm1

Hello @klugman

Do you want to extract gradient boosted tree predictions and create a visualization? 

If so you can extract predictions from the apply model operators. You can also store the trained model using the "Store" operator in rapidminer and retrieve this to apply on a new dataset for predictions in rapidminer. I don't think you can apply the trained model outside of rapidminer environment.

Gradient boosted trees are additive models and there won't be a single equation that can be extracted from the model.

klugman

Hi Vaurun,

I need to apply this model in a device that is not connected to the world - I guess I will need to find a solution outside Rapid.

Many Thanks,
Klug

IngoRM

Hi,

You can use a command line version of the RapidMiner engine or install a deployment server on that device to use all models directly.  I would definitely consider this if possible.

Most models do not support the generation of "easy" formulas.  There is actually an operator for this extraction, but only few model types are supported.  GBT is not one of them.  Here is an example process nevertheless.

Cheers,
Ingo

<?xml version="1.0" encoding="UTF-8"?><process version="9.6.000-SNAPSHOT"><br>  <context><br>    <input/><br>    <output/><br>    <macros/><br>  </context><br>  <operator activated="true" class="process" compatibility="9.6.000-SNAPSHOT" expanded="true" name="Process"><br>    <parameter key="logverbosity" value="init"/><br>    <parameter key="random_seed" value="2001"/><br>    <parameter key="send_mail" value="never"/><br>    <parameter key="notification_email" value=""/><br>    <parameter key="process_duration_for_mail" value="30"/><br>    <parameter key="encoding" value="UTF-8"/><br>    <process expanded="true"><br>      <operator activated="true" class="retrieve" compatibility="9.6.000-SNAPSHOT" expanded="true" height="68" name="Retrieve Sonar" width="90" x="45" y="34"><br>        <parameter key="repository_entry" value="//Samples/data/Sonar"/><br>      </operator><br>      <operator activated="true" class="support_vector_machine" compatibility="9.6.000-SNAPSHOT" expanded="true" height="124" name="SVM" width="90" x="179" y="34"><br>        <parameter key="kernel_type" value="dot"/><br>        <parameter key="kernel_gamma" value="1.0"/><br>        <parameter key="kernel_sigma1" value="1.0"/><br>        <parameter key="kernel_sigma2" value="0.0"/><br>        <parameter key="kernel_sigma3" value="2.0"/><br>        <parameter key="kernel_shift" value="1.0"/><br>        <parameter key="kernel_degree" value="2.0"/><br>        <parameter key="kernel_a" value="1.0"/><br>        <parameter key="kernel_b" value="0.0"/><br>        <parameter key="kernel_cache" value="200"/><br>        <parameter key="C" value="0.0"/><br>        <parameter key="convergence_epsilon" value="0.001"/><br>        <parameter key="max_iterations" value="100000"/><br>        <parameter key="scale" value="true"/><br>        <parameter key="calculate_weights" value="true"/><br>        <parameter key="return_optimization_performance" value="true"/><br>        <parameter key="L_pos" value="1.0"/><br>        <parameter key="L_neg" value="1.0"/><br>        <parameter key="epsilon" value="0.0"/><br>        <parameter key="epsilon_plus" value="0.0"/><br>        <parameter key="epsilon_minus" value="0.0"/><br>        <parameter key="balance_cost" value="false"/><br>        <parameter key="quadratic_loss_pos" value="false"/><br>        <parameter key="quadratic_loss_neg" value="false"/><br>        <parameter key="estimate_performance" value="false"/><br>      </operator><br>      <operator activated="true" class="create_formula" compatibility="9.6.000-SNAPSHOT" expanded="true" height="82" name="Create Formula" width="90" x="313" y="34"/><br>      <connect from_op="Retrieve Sonar" from_port="output" to_op="SVM" to_port="training set"/><br>      <connect from_op="SVM" from_port="model" to_op="Create Formula" to_port="model"/><br>      <connect from_op="Create Formula" from_port="formula" to_port="result 1"/><br>      <portSpacing port="source_input 1" spacing="0"/><br>      <portSpacing port="sink_result 1" spacing="0"/><br>      <portSpacing port="sink_result 2" spacing="0"/><br>    </process><br>  </operator><br></process><br><br>

klugman

Hi Ingo,

I am truly very impressed by RapidMiner software and service. I was hoping to find an easy solution :-)
The options mentioned above "command line version of the RapidMiner engine" how can one get it?

Many Thanks,
Klug

IngoRM

Hi,
There have been a couple of examples here on the community for that, see for example:

https://community.rapidminer.com/discussion/52524/is-there-a-way-to-run-a-rapidminer-process-from-the-command-line-on-a-mac

https://community.rapidminer.com/discussion/41784/how-do-i-start-rapidminer-studio-without-gui-in-command-line-mode

Hope this helps,
Ingo