org.openimaj.math.model

Class LeastSquaresLinearModel

java.lang.Object

org.openimaj.math.model.LeastSquaresLinearModel

All Implemented Interfaces:

Cloneable, EstimatableModel<Integer,Integer>, Model<Integer,Integer>

public class LeastSquaresLinearModel
extends Object
implements EstimatableModel<Integer,Integer>

Model of mapping between pairs of integers learned from a least-squares regression. Basically this class learns the parameters m and c in y = mx + c.

Author:

Jonathon Hare (jsh2@ecs.soton.ac.uk)

Constructor Summary

Constructors

Constructor and Description
LeastSquaresLinearModel() Construct model
LeastSquaresLinearModel(int nEstimates) Construct model

Method Summary

Modifier and Type	Method and Description
LeastSquaresLinearModel	clone() Clone the model
boolean	estimate(List<? extends IndependentPair<Integer,Integer>> data) Using standard vertical linear regression as outlined here: http://mathworld.wolfram.com/LeastSquaresFitting.html calculate the m and c of a line of best fit given the data.
double	getC() Get the offset (c in y=mx+c)
double	getM() Get the gradient (m in y=mx+c)
int	numItemsToEstimate()
Integer	predict(Integer data) Uses the model to predict dependent data from an independent value.
String	toString()

Methods inherited from class java.lang.Object

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

Constructor Detail

LeastSquaresLinearModel

public LeastSquaresLinearModel()

Construct model

LeastSquaresLinearModel

public LeastSquaresLinearModel(int nEstimates)

Construct model

Parameters:

nEstimates - minimum number of samples required for estimating model when fitting

Method Detail

estimate

public boolean estimate(List<? extends IndependentPair<Integer,Integer>> data)

Using standard vertical linear regression as outlined here: http://mathworld.wolfram.com/LeastSquaresFitting.html calculate the m and c of a line of best fit given the data.

Specified by:

estimate in interface EstimatableModel<Integer,Integer>

Parameters:

data - Observed data

Returns:

true if a successful estimate was made, false otherwise

See Also:

EstimatableModel.numItemsToEstimate()

predict

public Integer predict(Integer data)

Description copied from interface: Model

Uses the model to predict dependent data from an independent value.

Specified by:

predict in interface Model<Integer,Integer>

Parameters:

data - the data (independent variable)

Returns:

Dependent variable(s) predicted from the independent ones.

numItemsToEstimate

public int numItemsToEstimate()

Specified by:

numItemsToEstimate in interface EstimatableModel<Integer,Integer>

Returns:

The minimum number of observations required to estimate the model.

clone

public LeastSquaresLinearModel clone()

Description copied from interface: EstimatableModel

Clone the model

Specified by:

clone in interface EstimatableModel<Integer,Integer>

Overrides:

clone in class Object

Returns:

a cloned copy

toString

public String toString()

Overrides:

toString in class Object

getM

public double getM()

Get the gradient (m in y=mx+c)

Returns:

the gradient

getC

public double getC()

Get the offset (c in y=mx+c)

Returns:

the offset