astro.fits
Class Overscan.WifsipOverscan

java.lang.Object
  vec_math.AbstractDataModel
      vec_math.AbstractGradientModel
          astro.fits.Overscan.WifsipOverscan

All Implemented Interfaces:

PropertyAware, DataModel, GradientModel

Enclosing class:

Overscan

public static class Overscan.WifsipOverscan
extends AbstractGradientModel
implements PropertyAware

A non-linear model for fitting the exponentially glow in WiFSIP data. We use a double-exponential to model the overscan region from pixel row zero to approximately 350. The characteristic fall-off length, the two non-linear parameters in the model, have been fitted on an averaged twenty-second dark over five individual exposures.

Nested Class Summary

Nested classes/interfaces inherited from class vec_math.AbstractDataModel

AbstractDataModel.Construct

Field Summary

private int	decay Number of exponential decays with different tau that overlap.
private static int	DEFDECAYCOUNT Number of exponential decays overlapping.
static String	KEY_DECAYCOUNT The number of decaying exp-f we overlay in the non-linear fit.
static String	KEY_STARTROW If we are not stating at row zero.
private double	start Starting row number, zero in most cases.

Fields inherited from class vec_math.AbstractDataModel

sigma, x, y

Constructor Summary

Overscan.WifsipOverscan(VectorG[] row, double[] aduav, double[] aduerr)
The array of VectorGs actually only contains the row number of the fit, i.e.

Overscan.WifsipOverscan(VectorG[] row, double[] aduav, double[] aduerr, int exp)
The array of VectorGs actually only contains the row number of the fit, i.e.

Method Summary

double	evaluateModel(VectorG curfit, VectorG t) Returns the model adu for the queried row in vector t, which again is a Vector1D, t.a0 = row.
Matrix	getGradientMatrix(VectorG curfit) Returns the matrix that is the gradient to the two-exponential model.
int	getParameterCount() The number of parameters is fixed to five.
void	setComposedProperties(PropertySupplying prop) If we have the KEY_DECAYCOUNT property set, we record the number of exponential decay laws.

Methods inherited from class vec_math.AbstractGradientModel

check, check, getNegativeChi2HalfGradient, getNegativeChi2HalfGradient, getPseudoHessian, getPseudoHessian, LevenbergMarquardtSolver

Methods inherited from class vec_math.AbstractDataModel

getChiSquareModel, getChiSquareModel, getLorentzianModel, getMeasureCount, getMeasurementErrors, getMeasures, getModel, getModel, getResiduals, getResiduals, getRms, getRobustModel, getTimes, getTotalErrors, LorentzianSolver, RobustSolver, SimplexSolver

Methods inherited from class java.lang.Object

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

Methods inherited from interface vec_math.DataModel

getMeasureCount, getMeasurementErrors, getMeasures, getModel, getResiduals, getTimes, getTotalErrors

Field Detail

KEY_DECAYCOUNT

public static final String KEY_DECAYCOUNT

The number of decaying exp-f we overlay in the non-linear fit.

See Also:

Constant Field Values

KEY_STARTROW

public static final String KEY_STARTROW

If we are not stating at row zero.

See Also:

Constant Field Values

DEFDECAYCOUNT

private static final int DEFDECAYCOUNT

Number of exponential decays overlapping.

See Also:

Constant Field Values

decay

private int decay

Number of exponential decays with different tau that overlap.

start

private double start

Starting row number, zero in most cases.

Constructor Detail

Overscan.WifsipOverscan

public Overscan.WifsipOverscan(VectorG[] row,
                               double[] aduav,
                               double[] aduerr)

The array of VectorGs actually only contains the row number of the fit, i.e. VectorG[i] = Vector1D.a0=i.

Overscan.WifsipOverscan

public Overscan.WifsipOverscan(VectorG[] row,
                               double[] aduav,
                               double[] aduerr,
                               int exp)

The array of VectorGs actually only contains the row number of the fit, i.e. VectorG[i] = Vector1D.a0=i.

Method Detail

setComposedProperties

public void setComposedProperties(PropertySupplying prop)

If we have the KEY_DECAYCOUNT property set, we record the number of exponential decay laws.

Specified by:

setComposedProperties in interface PropertyAware

getParameterCount

public int getParameterCount()

The number of parameters is fixed to five.

Specified by:

getParameterCount in interface DataModel

Returns:

The number of parameters in the model. This equals the dimension of the input vector in DataModel.getModel(vec_math.VectorG).

evaluateModel

public double evaluateModel(VectorG curfit,
                            VectorG t)

Returns the model adu for the queried row in vector t, which again is a Vector1D, t.a0 = row. The model we fit has the following formular

 ADU(r=row) = a0 + a1*exp(-a3*r) + a2*exp(-a4*r)+...
 

where the number of exponentials that are overlayed is stored in decay

Specified by:

evaluateModel in interface DataModel

Parameters:

curfit - The model parameters

t - The (one)dimensional vector of independant variables.

getGradientMatrix

public Matrix getGradientMatrix(VectorG curfit)

Returns the matrix that is the gradient to the two-exponential model.

Specified by:

getGradientMatrix in interface GradientModel