org.apache.ctakes.ytex.kernel

Interface SparseDataFormatter

All Known Implementing Classes:

BaseSparseDataFormatter, LibSVMFormatterFactory.LibSVMFormatter, SemiLFormatterFactory.SemiLDataFormatter, SGTFormatterFactory.SGTFormatter, SparseMatrixFormatterFactory.SparseMatrixDataFormatter, SVMLightFormatterFactory.SVMLightFormatter, SVMLinFormatterFactory.SVMLinDataFormatter, WekaAttributeEvaluatorImpl.WekaAttributeEvaluatorFormatter, WekaFormatterFactory.WekaFormatter

public interface SparseDataFormatter

stateful class called by the sparseDataExporter to export sparse data in a specific format. This is created, called for each fold, training, and test set, then thrown away.

Author:

vijay

Field Summary

Fields

Modifier and Type	Field and Description
static String	ATTR_INSTANCE_ID value instance_id.
static String	SCOPE scope property key
static String	SCOPE_FOLD fold value for scope
static String	SCOPE_LABEL label value for scope

Method Summary

Modifier and Type	Method and Description
void	clearFold() clear all data structures set up during initializeFold
void	clearLabel()
void	exportFold(SparseData sparseData, SortedMap<Long,String> sortedMap, boolean train, String label, Integer run, Integer fold) export the fold train/test set.
void	initializeExport(InstanceData instanceLabel, Properties properties, SparseData sparseData) initialize export - called once
void	initializeFold(SparseData sparseData, String label, Integer run, Integer fold, SortedMap<Boolean,SortedMap<Long,String>> foldInstanceLabelMap) initialize data structures for the fold that will be exported.
void	initializeLabel(String label, SortedMap<Integer,SortedMap<Integer,SortedMap<Boolean,SortedMap<Long,String>>>> labelInstances, Properties properties, SparseData sparseData)

Field Detail

SCOPE

static final String SCOPE

scope property key

See Also:

Constant Field Values

SCOPE_FOLD

static final String SCOPE_FOLD

fold value for scope

See Also:

Constant Field Values

SCOPE_LABEL

static final String SCOPE_LABEL

label value for scope

See Also:

Constant Field Values

ATTR_INSTANCE_ID

static final String ATTR_INSTANCE_ID

value instance_id. SparseMatrix adds the instance_id attribute to the matrix. This is a reserved attribute name.

See Also:

Constant Field Values

Method Detail

initializeFold

void initializeFold(SparseData sparseData,
                    String label,
                    Integer run,
                    Integer fold,
                    SortedMap<Boolean,SortedMap<Long,String>> foldInstanceLabelMap)
             throws IOException

initialize data structures for the fold that will be exported. called before export.

Parameters:

sparseData -

label -

run -

fold -

foldInstanceLabelMap -

Throws:

IOException

exportFold

void exportFold(SparseData sparseData,
                SortedMap<Long,String> sortedMap,
                boolean train,
                String label,
                Integer run,
                Integer fold)
         throws IOException

export the fold train/test set. called once per train/test set, 2x per fold.

Parameters:

sparseData -

sortedMap -

train -

label -

run -

fold -

Throws:

IOException

initializeExport

void initializeExport(InstanceData instanceLabel,
                      Properties properties,
                      SparseData sparseData)
               throws IOException

initialize export - called once

Parameters:

instanceLabel -

properties -

Throws:

IOException

clearFold

void clearFold()

clear all data structures set up during initializeFold

initializeLabel

void initializeLabel(String label,
                     SortedMap<Integer,SortedMap<Integer,SortedMap<Boolean,SortedMap<Long,String>>>> labelInstances,
                     Properties properties,
                     SparseData sparseData)
              throws IOException

Throws:

IOException

clearLabel

void clearLabel()