Simultaneous Model Output and Perofrmance Output

atifshaikh4514

Hallo,
First of all, thanx for this tool.

I have a problem I never faced before, I have the following operator chain:

<operator name="Root" class="Process" expanded="yes">
    <operator name="DirectMailingExampleSetGenerator" class="DirectMailingExampleSetGenerator">
    </operator>
    <operator name="XValidation" class="XValidation" expanded="yes">
        <parameter key="keep_example_set"	value="true"/>
        <parameter key="sampling_type"	value="linear sampling"/>
        <operator name="MetaCost" class="MetaCost" expanded="yes">
            <parameter key="cost_matrix"	value="[10.0 3.0 1.0;1.0 0.0 2.0;6.0 1.0 0.0]"/>
            <parameter key="keep_example_set"	value="true"/>
            <parameter key="sampling_with_replacement"	value="false"/>
            <operator name="W-JRip" class="W-JRip">
                <parameter key="keep_example_set"	value="true"/>
            </operator>
        </operator>
        <operator name="OperatorChain" class="OperatorChain" expanded="yes">
            <operator name="ModelApplier" class="ModelApplier">
                <list key="application_parameters">
                </list>
                <parameter key="keep_model"	value="true"/>
            </operator>
            <operator name="Performance" class="Performance">
                <parameter key="keep_example_set"	value="true"/>
            </operator>
        </operator>
    </operator>
</operator>

I want to view each individual rule set generated by all iterations of JRip as well as the performance ouptut. Usually, without the performance output, a model simply outputs the rule sets without any explicit operator, but in this case, I am kinda stuck.

any ideas?

IngoRM

Hi,

the first idea could be to just add a breakpoint right after the learner (you can (de-)activate breakpoints by double-clicking on operators in the operator tree or you can select a breakpoint from the context menu which pops up after a right click on an operator). But maybe you will have a look into all models simultaneously? This is also possible with a little trick (using the %{a} macro):

<operator name="Root" class="Process" expanded="yes">
    <operator name="DirectMailingExampleSetGenerator" class="DirectMailingExampleSetGenerator">
    </operator>
    <operator name="XValidation" class="XValidation" expanded="yes">
        <parameter key="sampling_type"	value="linear sampling"/>
        <operator name="TrainingChain" class="OperatorChain" expanded="yes">
            <operator name="MetaCost" class="MetaCost" expanded="yes">
                <parameter key="cost_matrix"	value="[10.0 3.0 1.0;1.0 0.0 2.0;6.0 1.0 0.0]"/>
                <parameter key="keep_example_set"	value="true"/>
                <parameter key="sampling_with_replacement"	value="false"/>
                <operator name="W-JRip" class="W-JRip">
                    <parameter key="keep_example_set"	value="true"/>
                </operator>
            </operator>
            <operator name="ModelWriter" class="ModelWriter">
                <parameter key="model_file"	value="model_%{a}.mod"/>
            </operator>
        </operator>
        <operator name="TestingChain" class="OperatorChain" expanded="yes">
            <operator name="ModelApplier" class="ModelApplier">
                <list key="application_parameters">
                </list>
                <parameter key="keep_model"	value="true"/>
            </operator>
            <operator name="Performance" class="Performance">
                <parameter key="keep_example_set"	value="true"/>
            </operator>
        </operator>
    </operator>
    <operator name="IteratingOperatorChain" class="IteratingOperatorChain" expanded="yes">
        <parameter key="iterations"	value="10"/>
        <operator name="ModelLoader" class="ModelLoader">
            <parameter key="model_file"	value="model_%{a}.mod"/>
        </operator>
    </operator>
</operator>

Hope that helps. Cheers,
Ingo

atifshaikh4514

Thanks Ingo,
the trick did the trick :-)
;D

IngoRM

Hi,



great to hear, thanks for the response.

Cheers,
Ingo