Source

WiktionaryIdioms / WiktionaryIdioms / config / classifierconfig.xml

Full commit
<?xml version="1.0" encoding="UTF-8"?>
<!-- config.xml -->
<config>
	<MySQL>
		<classify>
			<database>database_to_draw_from</database>
			<table>table_to_draw_from</table>
			<column>data_set_column</column>
			<testData>what_the_data_set_column_is_for_the_test_set</testData>
		</classify>
	</MySQL>

	<default>
		<trainPath>./data/train.txt</trainPath>
		<testPath>./data/dev_annotated.txt</testPath>
	    <percents></percents>
		
		<features>1, 2, 4, 6, 8, 9, 10, 14</features>
		<errorBound>.000001</errorBound>
		<specificsPath>./specifics/specifics.txt</specificsPath>
		<classifierType>perceptron</classifierType>
		<precisionRecallGranularity>.01</precisionRecallGranularity>
		
		<cleanTest>false</cleanTest>
		<testCorrectedLabel>true</testCorrectedLabel>
		
		<verbose>true</verbose>
		<shouldOutputGeneral>true</shouldOutputGeneral>
		<shouldOutputModel>true</shouldOutputModel>
		
		<output>
			<file>file_name</file>
			<generalInfo>true</generalInfo>
			<confidenceList>true</confidenceList>
			<precisionRecallPoints>true</precisionRecallPoints>
			<randomErrorAnalysis>100</randomErrorAnalysis>
		</output>
		<outputModel>
			<file>test</file>
		</outputModel>
	</default>
	
    
    <BasicApply>
    	<learningRate>1</learningRate>
		<iterations>93</iterations>
		<modelFile></modelFile>
    </BasicApply>
    
    <GridSearch>
    	<group>groups</group>
    	<iterationMax>100</iterationMax>
    	<learningDelta>1</learningDelta>
    </GridSearch>
    
    <CompareFeatures>
    	<features>6,8,9,10,14</features>

		<buildUp>true</buildUp>
		<buildByBest>false</buildByBest>
    	<iterationMax>100</iterationMax>
    	<learningDelta>1</learningDelta>
    </CompareFeatures>
    
     <CompareGroups>
    	<iterationMax>100</iterationMax>
    	<learningDelta>1</learningDelta>
    </CompareGroups>
    
    <LabelData>
    	<modelFile></modelFile>
    	<precisionBoundary></precisionBoundary>
    </LabelData>
</config>