Options

SlidingWindowValidation

walden21walden21 Member Posts: 2 Contributor I
edited November 2018 in Help
Hello there!
I'm newbie here. I installed RM a few days ago.
I tried some data mining method with RM...and felt very interesting these days (and nights!)

Today, I met my first obstacle in RM...that's SlidingWindowValidation!

Error messages
G May 19, 2009 2:48:43 PM: [Fatal] RuntimeException occured in 1st application of SlidingWindowValidation (SlidingWindowValidation)
G May 19, 2009 2:48:43 PM: [Fatal] Process failed: operator cannot be executed (Illegal operator index in getOperator() (SlidingWindowValidation): 0). Check the log messages...
          Root[1] (Process)
          +- ExcelExampleSource[1] (ExcelExampleSource)
          +- FeatureNameFilter[1] (FeatureNameFilter)
          +- NearestNeighbors[1] (NearestNeighbors)
          +- ModelApplier[1] (ModelApplier)
here ==> +- SlidingWindowValidation[1] (SlidingWindowValidation)

Here's data and XML scripts...
God bless who are glad to help me with this problem!
<operator name="Root" class="Process" expanded="yes">
    <operator name="ExcelExampleSource" class="ExcelExampleSource">
        <parameter key="excel_file" value="C:\!Hard1\RMdata\KNN2_rm.xls"/>
        <parameter key="first_row_as_names" value="true"/>
        <parameter key="create_label" value="true"/>
        <parameter key="label_column" value="6"/>
        <parameter key="create_id" value="true"/>
    </operator>
    <operator name="FeatureNameFilter" class="FeatureNameFilter">
        <parameter key="filter_special_features" value="true"/>
        <parameter key="skip_features_with_name" value="Op"/>
    </operator>
    <operator name="NearestNeighbors" class="NearestNeighbors">
        <parameter key="keep_example_set" value="true"/>
        <parameter key="measure_types" value="NumericalMeasures"/>
    </operator>
    <operator name="ModelApplier" class="ModelApplier">
        <parameter key="keep_model" value="true"/>
        <list key="application_parameters">
        </list>
    </operator>
    <operator name="SlidingWindowValidation" class="SlidingWindowValidation" expanded="yes">
        <parameter key="training_window_width" value="50"/>
        <parameter key="test_window_width" value="50"/>
        <parameter key="cumulative_training" value="true"/>
        <parameter key="average_performances_only" value="false"/>
    </operator>
</operator>
I can't find how to attatch my data file to this board...
So I can't help posting some records as follow...
RowNo. Close date prediction(Close) Hi Lo CI Barnum
1 -0.35 10/29/07 -0.35 0.978 -0.309 0.959 66.0
2 1.15 10/30/07 1.15 0.23 -0.77 -0.48 75.0
3 1.2 10/31/07 1.2 0.804 -0.192 0.211 64.0
4 1.7 11/1/07 1.7 0.0 -1.368 -1.234 67.0
5 -5.6 11/2/07 -5.6 1.393 -0.039 1.258 62.0
6 -0.35 11/5/07 -0.35 0.274 -0.965 -0.176 55.0
7 2.4 11/6/07 2.4 1.207 0.0 1.11 75.0
8 -2.95 11/7/07 -2.95 1.308 -0.019 0.4 77.0
9 -3.7 11/8/07 -3.7 0.763 -0.433 0.137 55.0

Answers

  • Options
    haddockhaddock Member Posts: 849 Maven
    Welcome to Rapido land!

    You have my sympathies, Rapido offers much, and is wonderful to play with when you get over the basics, but sometimes getting there can be confusing. So a few tips to help you on your way.

    Learning from examples works - when you are not sure about something whizz through the samples to see whether something looks like what you want. In your case 03_XValidation_numerical looks cool.

    Take the guided tour - If you go Help>Tutorial those samples are presented in order, and give a pretty good overview of the games you can play.

    Don't re-invent wheels - File>Wizard can generate some starting points, or take an example and then tailor it to your needs.

    Press the buttons - Zap every single one until you know what they do, for instance the double tick button spots errors in your setup, and prints errors which sometimes are helpful. In your case it says that the validation needs two sub operators, and when you get to the first validation example on the guided tour you'll see why.

    But press on, it really is worth it!

Sign In or Register to comment.