"ModelWriter: XML file"

nicugeorgian

Hi,

Within the training chain of a cross validation process, I have inserted a ModelWriter operator after a learner operator (in my case, W-M5P). The corresponding part of the code looks as follows:

<operator name="XValidation" class="XValidation" expanded="yes">
        <operator name="TrainingChain" class="OperatorChain" expanded="yes">
            <operator name="W-M5P" class="W-M5P">
                <parameter key="M"	value="10.0"/>
                <parameter key="R"	value="true"/>
                <parameter key="keep_example_set"	value="true"/>
            </operator>
            <operator name="ModelWriter" class="ModelWriter">
                <parameter key="model_file"	value="model_%{a}.mod"/>
            </operator>
        </operator>

As the cross validation has 10 number_of_validations, the final output consists of 10 file models: model_i.mod, where i = 1,10.

My question is: why I don't get also 10 XML zipped files correspoding to the 10 models? I think that the default value of the parameter output_type of the operator ModelWriter is XML Zipped, right? How can I get the models in the XML (or XML zipped) format?

Thanks,
Geo

IngoRM

Hi,

ahem, I am not sure if I get your point. The created models "model_i.mod" actually are zipped XML files. Just rename them to model_i.zip and you will be able to unzip them.

Cheers,
Ingo

nicugeorgian

Hi Ingo,

many thanks for the answer.

I have renamed the mod files to zip files, but I was not able to unzip them. Anyway, I ran again the process, this time asking the ModelWriter to yield xml files as opposed to zipped xml files. And yes, I was I able to open them as xml files.

They, however, do not looas as I expected  In my case, the learner, W-M5P, yields a regression tree. I was expecting to see only the regression tree in the file returned by ModelWriter. There is, however, a plenty of other information in this model file 

Is there a way to get an xml file with the detail of the regression tree only?

Thanks,
Geo

IngoRM

Hi again,

the reason why it did not work with the zip files probably is that the files are actually zipped using gzip, so renaming them to "gz" and using a unzipper which is able to unzip them should work.

Is there a way to get an xml file with the detail of the regression tree only?

Try a ResultWriter:

<operator name="Root" class="Process" expanded="yes">
    <operator name="ExampleSetGenerator" class="ExampleSetGenerator">
        <parameter key="target_function"	value="sum"/>
    </operator>
    <operator name="BinDiscretization" class="BinDiscretization">
        <parameter key="number_of_bins"	value="3"/>
    </operator>
    <operator name="XValidation" class="XValidation" expanded="yes">
        <operator name="OperatorChain (2)" class="OperatorChain" expanded="yes">
            <operator name="W-M5P" class="W-M5P">
                <parameter key="N"	value="true"/>
            </operator>
            <operator name="ResultWriter" class="ResultWriter">
                <parameter key="result_file"	value="model_%{a}.txt"/>
            </operator>
        </operator>
        <operator name="OperatorChain" class="OperatorChain" expanded="yes">
            <operator name="ModelApplier" class="ModelApplier">
                <list key="application_parameters">
                </list>
            </operator>
            <operator name="Performance" class="Performance">
            </operator>
        </operator>
    </operator>
</operator>

This will produce files with exactly the same output as the text view.

Cheers,
Ingo

nicugeorgian

mierswa wrote:
This will produce files with exactly the same output as the text view.

Yes, it worked Thank you!

I now have to find a way to parse such a file / or to convert it to an XML file.

Cheers,
Geo