Model without correlated attributes.

Legacy User

Hallo

I have different problems with my models
Usually are in practice always a lot of attributes are correlated each other.
Unfortunately, there is no such generator, which produces correlated attributes.
I can therefore show the present example, not directly. But the principle is the same.

But I would still find a way
a) I want to test all attributes combinations with each other. This may not be correlated !!
b) With the operator Feature Selection  I would like to decide how many variables are to be incorporated into the model. It is often not useful to have 20 variables in the model.
c) he will find the best model to me and write out the linear model as well.
d) Unfortunately, I did not get the forecast (prediction) for the model

But I have some problems with my model. Therefore I hope for your good ideas :-X

Many thanks for you opinion and help

Momo




My Modelkonstruction:


<operator name="Root" class="Process" expanded="yes">
    <operator name="Daten laden  und vorbereiten" class="OperatorChain" expanded="yes">
        <operator name="ExampleSetGenerator" class="ExampleSetGenerator">
            <parameter key="target_function" value="random"/>
        </operator>
        <operator name="RandomOptimizer" class="RandomOptimizer" expanded="yes">
            <parameter key="iterations" value="100"/>
            <operator name="RemoveCorrelatedFeatures (2)" class="RemoveCorrelatedFeatures">
                <parameter key="correlation" value="0.5"/>
                <parameter key="attribute_order" value="random"/>
            </operator>
            <operator name="CorrelationMatrix" class="CorrelationMatrix">
            </operator>
            <operator name="FeatureSelection" class="FeatureSelection" expanded="yes">
                <parameter key="plot_generations" value="100"/>
                <parameter key="keep_best" value="3"/>
                <parameter key="generations_without_improval" value="-1"/>
                <parameter key="maximum_number_of_generations" value="3"/>
                <operator name="XValidation" class="XValidation" expanded="yes">
                    <parameter key="create_complete_model" value="true"/>
                    <parameter key="local_random_seed" value="1"/>
                    <operator name="LinearRegression" class="LinearRegression">
                        <parameter key="feature_selection" value="none"/>
                    </operator>
                    <operator name="OperatorChain" class="OperatorChain" expanded="yes">
                        <operator name="ModelApplier" class="ModelApplier">
                            <list key="application_parameters">
                            </list>
                        </operator>
                        <operator name="RegressionPerformance" class="RegressionPerformance">
                            <parameter key="main_criterion" value="squared_correlation"/>
                            <parameter key="root_mean_squared_error" value="true"/>
                            <parameter key="squared_correlation" value="true"/>
                        </operator>
                    </operator>
                </operator>
            </operator>
        </operator>
        <operator name="ModelApplier (2)" class="ModelApplier">
            <parameter key="keep_model" value="true"/>
            <list key="application_parameters">
            </list>
        </operator>
    </operator>
</operator>

land

Hi,
sorry but I didn't understand, what was your problem at all. So if you could give a more detailed overall description of your target, then I might be able to help you. But please keep in mind, that this is only a forum and I cannot design arbitrary complex processes. If you are interested in full personal consulting for example via phone, you might contact us at contact@rapid-i.com and ask of an individual solution, tailored to your needs.

Greetings,
  Sebastian

Legacy User

Dear Sebastian
I'm glad that there is this forum, because I can learn a lot from this for my own models. Thank you for your efforts for that.
My current difficulty is that my model. I need an output for the logistic model. I do not know how to do this.
The second ModelApplier (2) it always gives me an error message.

Many thanks
Momo

land

Hi,
you probably will need both an example set and a model for applying it. And the example set needs the same attribute signature as the one learned the model on.


Greetings,
  Sebastian