de.lmu.ifi.dbs.elki.math.scales

Class LinearScale

java.lang.Object

de.lmu.ifi.dbs.elki.math.scales.LinearScale

public class LinearScale
extends Object

Class to handle a linear scale for an axis. The computed scales are rounded to be on decimal borders, choosing an appropriate resolution to have between 4 and 31 major tics (3 to 30 intervals). Future versions might use major/minor tics to get even nicer values.

Author:

Erich Schubert

Field Summary

Fields

Modifier and Type	Field and Description
private double	delta Scale delta := max - min
private int	log10res Scale resolution in log10.
private double	max max value of the scale
private double	min min value of the scale
private double	res Scale resolution
private double	ZOOMFACTOR

Constructor Summary

Constructors

Constructor and Description
LinearScale(double min, double max) Constructor.

Method Summary

Methods

Modifier and Type	Method and Description
String	formatValue(double val) Format value according to the scales resolution (i.e. appropriate number of digits)
double	getLog10Res() Get resolution (scale interval size)
double	getMax() Get maximum value (scale, not data).
double	getMin() Get minimum value (scale, not data).
double	getRelativeScaled(double val) Covert a relative value to it's scale position
double	getRelativeScaled(double val, double smax, double smin) Covert a relative value to it's scale position
double	getRelativeUnscaled(double val) Covert a relative scale position to the actual value
double	getRelativeUnscaled(double val, double smin, double smax) Covert a relative scale position to the actual value
double	getRes() Get resolution (scale interval size)
double	getScaled(double val) Covert a value to it's scale position
double	getScaled(double val, double smin, double smax) Covert a value to it's scale position
double	getUnscaled(double val) Covert a scale position to the actual value
double	getUnscaled(double val, double smin, double smax) Covert a scale position to the actual value

Methods inherited from class java.lang.Object

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

Field Detail

ZOOMFACTOR

private final double ZOOMFACTOR

min

private double min

min value of the scale

max

private double max

max value of the scale

res

private double res

Scale resolution

log10res

private int log10res

Scale resolution in log10.

delta

private double delta

Scale delta := max - min

Constructor Detail

LinearScale

public LinearScale(double min,
           double max)

Constructor. Computes a scale covering the range of min-max with between 3 and 30 intervals, rounded to the appropriate number of digits.

Parameters:

min - actual minimum in the data

max - actual maximum in the data

Method Detail

getMin

public double getMin()

Get minimum value (scale, not data).

Returns:

min

getMax

public double getMax()

Get maximum value (scale, not data).

Returns:

max

getRes

public double getRes()

Get resolution (scale interval size)

Returns:

scale interval size

getLog10Res

public double getLog10Res()

Get resolution (scale interval size)

Returns:

scale interval size in logarithmic form

getScaled

public double getScaled(double val)

Covert a value to it's scale position

Parameters:

val - data value

Returns:

scale position in the interval [0:1]

getUnscaled

public double getUnscaled(double val)

Covert a scale position to the actual value

Parameters:

val - scale position in the interval [0:1]

Returns:

value on the original scale

getRelativeScaled

public double getRelativeScaled(double val)

Covert a relative value to it's scale position

Parameters:

val - relative data value

Returns:

relative scale position in the interval [0:1]

getRelativeUnscaled

public double getRelativeUnscaled(double val)

Covert a relative scale position to the actual value

Parameters:

val - relative scale position in the interval [0:1]

Returns:

relative value on the original scale

getScaled

public double getScaled(double val,
               double smin,
               double smax)

Covert a value to it's scale position

Parameters:

val - data value

smin - target scale minimum

smax - target scale maximum

Returns:

scale position in the interval [smin:smax]

getUnscaled

public double getUnscaled(double val,
                 double smin,
                 double smax)

Covert a scale position to the actual value

Parameters:

val - scale position in the interval [smin:smax]

smin - target scale minimum

smax - target scale maximum

Returns:

value on the original scale

getRelativeScaled

public double getRelativeScaled(double val,
                       double smax,
                       double smin)

Covert a relative value to it's scale position

Parameters:

val - relative data value

smin - target scale minimum

smax - target scale maximum

Returns:

relative scale position in the interval [smin:smax]

getRelativeUnscaled

public double getRelativeUnscaled(double val,
                         double smin,
                         double smax)

Covert a relative scale position to the actual value

Parameters:

val - relative scale position in the interval [smin:smax]

smin - target scale minimum

smax - target scale maximum

Returns:

relative value on the original scale

formatValue

public String formatValue(double val)

Format value according to the scales resolution (i.e. appropriate number of digits)

Parameters:

val - Value

Returns:

formatted number