weka.classifiers.timeseries.core

Class Utils

java.lang.Object

weka.classifiers.timeseries.core.Utils

public class Utils
extends Object

Static utility routines.

Version:

$Revision: 52593 $

Author:

Mark Hall (mhall{[at]}pentaho{[dot]}com)

Constructor Summary

Constructors

Constructor and Description
Utils()

Method Summary

Methods

Modifier and Type	Method and Description
static double	advanceSuppliedTimeValue(double valueToAdvance, TSLagMaker.PeriodicityHandler dateBasedPeriodicity) Utility method to advance a supplied time value by one unit.
static double	decrementSuppliedTimeValue(double valueToDecrement, TSLagMaker.PeriodicityHandler dateBasedPeriodicity) Utility method to decrement a supplied time value by one unit.
static weka.core.Instances	insertMissing(weka.core.Instances toInsert, weka.core.Attribute timeStampAtt, TSLagMaker.PeriodicityHandler periodicityHandler, String m_skipEntries, List<String> missingReport) Check to see if there are any instances (time steps) that are missing entirely from the data (and are not in the skip list).
static weka.core.Instances	replaceMissing(weka.core.Instances toReplace, List<String> targets, String timeStampName, boolean dateOnly, TSLagMaker.Periodicity userHint, String skipEntries, Object... missingReport) Replace missing target values by interpolation.

Methods inherited from class java.lang.Object

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

Constructor Detail

Utils

public Utils()

Method Detail

insertMissing

public static weka.core.Instances insertMissing(weka.core.Instances toInsert,
                                weka.core.Attribute timeStampAtt,
                                TSLagMaker.PeriodicityHandler periodicityHandler,
                                String m_skipEntries,
                                List<String> missingReport)

Check to see if there are any instances (time steps) that are missing entirely from the data (and are not in the skip list). We make the assumption that the data won't contain both missing date values *and* entirely missing rows as this is a kind of chicken and egg situation with regards to detecting missing rows. E.g. if we have missing date values bracketing one or more missing rows then we can't do the date comparisons necessary for detecting whether the intermediate rows are missing. Similarly the missing date handling routine assumes that there aren't missing rows so that it can advance the previous date value by one unit in order to compute the value for the current missing date value.

Parameters:

toInsert - the instances to check

timeStampAtt - the name of time stamp att

periodicityHandler - the periodicity handler

m_skipEntries - the list of data "holes" that are expected (i.e. don't actually count as time stamp increments)

missingReport - will hold time stamps of any instances we insert

Returns:

the (potentially) modified instances

replaceMissing

public static weka.core.Instances replaceMissing(weka.core.Instances toReplace,
                                 List<String> targets,
                                 String timeStampName,
                                 boolean dateOnly,
                                 TSLagMaker.Periodicity userHint,
                                 String skipEntries,
                                 Object... missingReport)

Replace missing target values by interpolation. Also replaces missing date values (if a date time stamp has been specified and if possible).

Parameters:

toReplace - the instances to replace missing target values and time stamp values

targets - a list of target attributes to check for missing values

timeStampName - the name of the time stamp attribute (or null if there is no time stamp)

dateOnly - if true, only replace missing date values and not missing target values (useful for hold-out test sets)

userHint - user-specified hint as to the data periodicity

skipEntries - any skip entries (may be null)

missingReport - a varargs parameter that, if provided, is expected to be up to two lists of Integers and one list of Strings. The first list will be populated with the instance numbers (duplicates are possible) of instances that have missing targets replaced. The second list will be populated with the instance numbers of instances that have missing time stamp values replaced. The third list will be populated with the time stamps for any new instances that are inserted into the data (i.e. if we detect that there are "holes" in the data that aren't covered by the skip list entries.

Returns:

the instances with missing targets and (possibly) missing time stamp values replaced.

advanceSuppliedTimeValue

public static double advanceSuppliedTimeValue(double valueToAdvance,
                              TSLagMaker.PeriodicityHandler dateBasedPeriodicity)

Utility method to advance a supplied time value by one unit.

Parameters:

valueToAdvance - the time value to advance

dateBasedPeriodicity - the periodicity to use for data arithmetic

Returns:

the advanced value or the original value if this lag maker is not adjusting for trends.

decrementSuppliedTimeValue

public static double decrementSuppliedTimeValue(double valueToDecrement,
                                TSLagMaker.PeriodicityHandler dateBasedPeriodicity)

Utility method to decrement a supplied time value by one unit.

Parameters:

valueToDecrement - the time value to decrement

dateBasedPeriodicity - the periodicity to use for data arithmetic

Returns:

the advanced value or the original value if this lag maker is not adjusting for trends.