net.librec.data.convertor

Class ArffDataConvertor

java.lang.Object

net.librec.job.progress.ProgressReporter

net.librec.data.convertor.AbstractDataConvertor

net.librec.data.convertor.ArffDataConvertor

All Implemented Interfaces:

java.lang.Runnable, DataConvertor, Progressable

public class ArffDataConvertor
extends AbstractDataConvertor

A ArffDataConvertor is a class to convert a data file from ARFF format to a target format.

Author:

Tang Jiaxi and Ma Chen

Field Summary

Fields

Modifier and Type	Field and Description
SparseMatrix	oneHotFeatureMatrix
DenseVector	oneHotRatingVector

Fields inherited from class net.librec.data.convertor.AbstractDataConvertor

datetimeMatrix, preferenceMatrix, sparseTensor

Fields inherited from class net.librec.job.progress.ProgressReporter

PROGRESS_INTERVAL

Constructor Summary

Constructors

Constructor and Description
ArffDataConvertor(java.lang.String path) Initializes a newly created ArffDataConvertor object with the path of the input data file.
ArffDataConvertor(java.lang.String path, java.util.ArrayList<com.google.common.collect.BiMap<java.lang.String,java.lang.Integer>> featureMapping)

Method Summary

Methods

Modifier and Type	Method and Description
java.util.ArrayList<com.google.common.collect.BiMap<java.lang.String,java.lang.Integer>>	getAllFeatureIds() Return user, item, appender {raw id, inner id} mapping
java.util.ArrayList<ArffAttribute>	getAttributes() Return the attributes the input data.
java.util.ArrayList<ArffInstance>	getInstances() Return the instances of the input data.
com.google.common.collect.BiMap<java.lang.String,java.lang.Integer>	getItemIds() Return item {rawid, inner id} mappings
java.lang.String	getRelationName() Return the relation name of input data.
com.google.common.collect.BiMap<java.lang.String,java.lang.Integer>	getUserIds() Return user {rawid, inner id} mappings
void	oneHotEncoding() Build the oneHotFeatureMatrix and oneHotRatingVector
void	processData() Process the input data.
void	progress() Report progress to the Librec framework.
void	readData() Read data from the data file.

Methods inherited from class net.librec.data.convertor.AbstractDataConvertor

getDatetimeMatrix, getPreferenceMatrix, getSparseTensor

Methods inherited from class net.librec.job.progress.ProgressReporter

getJobStatus, progressx, run

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

oneHotFeatureMatrix

public SparseMatrix oneHotFeatureMatrix

oneHotRatingVector

public DenseVector oneHotRatingVector

Constructor Detail

ArffDataConvertor

public ArffDataConvertor(java.lang.String path)

Initializes a newly created ArffDataConvertor object with the path of the input data file.

Parameters:

path - path of the input data file.

ArffDataConvertor

public ArffDataConvertor(java.lang.String path,
                 java.util.ArrayList<com.google.common.collect.BiMap<java.lang.String,java.lang.Integer>> featureMapping)

Method Detail

readData

public void readData()
              throws java.io.IOException

Read data from the data file.

Throws:

java.io.IOException - if the path is not valid

processData

public void processData()
                 throws java.io.IOException

Process the input data.

Throws:

java.io.IOException - if the path is not valid

progress

public void progress()

Description copied from interface: Progressable

Report progress to the Librec framework.

oneHotEncoding

public void oneHotEncoding()

Build the oneHotFeatureMatrix and oneHotRatingVector

getRelationName

public java.lang.String getRelationName()

Return the relation name of input data.

Returns:

relationName

getInstances

public java.util.ArrayList<ArffInstance> getInstances()

Return the instances of the input data.

Returns:

instances

getAttributes

public java.util.ArrayList<ArffAttribute> getAttributes()

Return the attributes the input data.

Returns:

attributes

getUserIds

public com.google.common.collect.BiMap<java.lang.String,java.lang.Integer> getUserIds()

Return user {rawid, inner id} mappings

Returns:

the mapping between row id and inner id of users

getItemIds

public com.google.common.collect.BiMap<java.lang.String,java.lang.Integer> getItemIds()

Return item {rawid, inner id} mappings

Returns:

the mapping between row id and inner id of items

getAllFeatureIds

public java.util.ArrayList<com.google.common.collect.BiMap<java.lang.String,java.lang.Integer>> getAllFeatureIds()

Return user, item, appender {raw id, inner id} mapping

Returns:

the mapping between row id and inner id of each columns in data set