def findPeaks(data,qData,initialGuess,numberOfChi,stddev,verbose=1):
if numberOfChi <=1:
raise Exception("This program must look for at least 2 chi slices.")
chiUpper = 360.0
chiLower = 0.11
# the -1 allows for us to include chiLower & chiUpper in our count
chiStep = (chiUpper-chiLower)*1.0/(numberOfChi-1)
peakList = []
if verbose: print ' - %d total Q values. Calculating peaks for Q =' % (len(qData.getAllQPairs())),
for Q,dQ in qData.getAllQPairs():
if verbose: print ' %f' % Q,
for chi in frange(chiLower,chiUpper,chiStep):
try:
x,y,qFit,width=FitWrap.getPeak(data,stddev,initialGuess.getCenterX()['val'],
initialGuess.getCenterY()['val'],initialGuess.getDistance()['val'],
initialGuess.getEnergy()['val'],initialGuess.getAlpha()['val'],
initialGuess.getBeta()['val'],initialGuess.getRotation()['val'],Q-dQ,Q+dQ,chi,
initialGuess.getPixelLength()['val'],initialGuess.getPixelHeight()['val'])
peakList.append([x,y,Q,qFit,chi,width])
except ValueError: # this means no peak found
pass
print
return peakList
def getPeakList(data, qData, initialGuess, numberOfChi, stddev, verbose=1):
if numberOfChi <= 1:
raise Exception("This program must look for at least 2 chi slices.")
chiUpper = 360.0
chiLower = 0.11
# the -1 allows for us to include chiLower & chiUpper in our count
chiStep = (chiUpper - chiLower) * 1.0 / (numberOfChi - 1)
peakList = PeakList.PeakList()
for Q, dQ in qData.getAllQPairs():
for chi in frange(chiLower, chiUpper, chiStep):
try:
x, y, qFit, width = FitWrap.getPeak(
data, stddev,
initialGuess.getCenterX()['val'],
initialGuess.getCenterY()['val'],
initialGuess.getDistance()['val'],
initialGuess.getEnergy()['val'],
initialGuess.getAlpha()['val'],
initialGuess.getBeta()['val'],
initialGuess.getRotation()['val'], Q - dQ, Q + dQ, chi,
initialGuess.getPixelLength()['val'],
initialGuess.getPixelHeight()['val'])
peakList.addPeak(x, y, Q, qFit, chi, width)
except ValueError: # this means no peak found
pass
return peakList
def fit(data,initialGuess,peakList,maskedPixelInfo,verbose=1):
# create nice Numeric data structures to store all of
# the peak stuff in. This lets us read our data easily
# once we get into C
xValues,yValues,qReal,chi,width,intensity = peakList.getMaskedPeakArrays(maskedPixelInfo,data)
# make C do the hard work :)
centerX,centerY,distance,energy,alpha,beta,rotation, \
covariance,initialResidual,finalResidual, \
reasonForQuitting = FitWrap.fitCalibrationParameters(
xValues,yValues,qReal,intensity,
initialGuess.getCenterX()['val'],
initialGuess.getCenterX()['fixed'],
initialGuess.getCenterY()['val'],
initialGuess.getCenterY()['fixed'],
initialGuess.getDistance()['val'],
initialGuess.getDistance()['fixed'],
initialGuess.getEnergy()['val'],
initialGuess.getEnergy()['fixed'],
initialGuess.getAlpha()['val'],
initialGuess.getAlpha()['fixed'],
initialGuess.getBeta()['val'],
initialGuess.getBeta()['fixed'],
initialGuess.getRotation()['val'],
initialGuess.getRotation()['fixed'],
initialGuess.getPixelLength()['val'],
initialGuess.getPixelHeight()['val'])
# add the fit data to a new calibration file
bestGuess = CalibrationData.CalibrationData()
bestGuess.setCenterX(centerX, fixed = initialGuess.getCenterX()['fixed'] )
bestGuess.setCenterY(centerY, fixed = initialGuess.getCenterY()['fixed'] )
bestGuess.setDistance(distance, fixed = initialGuess.getDistance()['fixed'] )
bestGuess.setEnergy(energy, fixed = initialGuess.getEnergy()['fixed'] )
bestGuess.setAlpha(alpha, fixed = initialGuess.getAlpha()['fixed'] )
bestGuess.setBeta(beta, fixed = initialGuess.getBeta()['fixed'] )
bestGuess.setRotation(rotation, fixed = initialGuess.getRotation()['fixed'] )
# pixel length & height don't change
bestGuess.setPixelLength(initialGuess.getPixelLength()['val'])
bestGuess.setPixelHeight(initialGuess.getPixelHeight()['val'])
return bestGuess,peakList,covariance,initialResidual,finalResidual,reasonForQuitting
def fit(data, initialGuess, peakList, maskedPixelInfo, verbose=1):
# create nice Numeric data structures to store all of
# the peak stuff in. This lets us read our data easily
# once we get into C
xValues, yValues, qReal, chi, width, intensity = peakList.getMaskedPeakArrays(
maskedPixelInfo, data)
# make C do the hard work :)
centerX,centerY,distance,energy,alpha,beta,rotation, \
covariance,initialResidual,finalResidual, \
reasonForQuitting = FitWrap.fitCalibrationParameters(
xValues,yValues,qReal,intensity,
initialGuess.getCenterX()['val'],
initialGuess.getCenterX()['fixed'],
initialGuess.getCenterY()['val'],
initialGuess.getCenterY()['fixed'],
initialGuess.getDistance()['val'],
initialGuess.getDistance()['fixed'],
initialGuess.getEnergy()['val'],
initialGuess.getEnergy()['fixed'],
initialGuess.getAlpha()['val'],
initialGuess.getAlpha()['fixed'],
initialGuess.getBeta()['val'],
initialGuess.getBeta()['fixed'],
initialGuess.getRotation()['val'],
initialGuess.getRotation()['fixed'],
initialGuess.getPixelLength()['val'],
initialGuess.getPixelHeight()['val'])
# add the fit data to a new calibration file
bestGuess = CalibrationData.CalibrationData()
bestGuess.setCenterX(centerX, fixed=initialGuess.getCenterX()['fixed'])
bestGuess.setCenterY(centerY, fixed=initialGuess.getCenterY()['fixed'])
bestGuess.setDistance(distance, fixed=initialGuess.getDistance()['fixed'])
bestGuess.setEnergy(energy, fixed=initialGuess.getEnergy()['fixed'])
bestGuess.setAlpha(alpha, fixed=initialGuess.getAlpha()['fixed'])
bestGuess.setBeta(beta, fixed=initialGuess.getBeta()['fixed'])
bestGuess.setRotation(rotation, fixed=initialGuess.getRotation()['fixed'])
# pixel length & height don't change
bestGuess.setPixelLength(initialGuess.getPixelLength()['val'])
bestGuess.setPixelHeight(initialGuess.getPixelHeight()['val'])
return bestGuess, peakList, covariance, initialResidual, finalResidual, reasonForQuitting
def fit(data, initialGuess, peakList, verbose=1):
print
print 'Performing the image calibration'
# create a nice Numeric data strucure to store all of the peak stuff in.
# This lets us read our data easily once we get into C
length = len(peakList)
xValues = Numeric.zeros((length), Numeric.Float64)
yValues = Numeric.zeros((length), Numeric.Float64)
qReal = Numeric.zeros((length), Numeric.Float64)
intensity = Numeric.zeros((length), Numeric.Float64)
for peakIndex in range(len(peakList)):
xPeak, yPeak, Qreal, Qfit, chi, width = peakList[peakIndex]
xValues[peakIndex] = xPeak
yValues[peakIndex] = yPeak
qReal[peakIndex] = Qreal
intensity[peakIndex] = data[int(xPeak)][int(yPeak)]
# make C do the hard work :)
centerX, centerY, distance, energy, alpha, beta, rotation, covariance = FitWrap.fitCalibrationParameters(
xValues, yValues, qReal, intensity,
initialGuess.getCenterX()['val'],
initialGuess.getCenterX()['fixed'],
initialGuess.getCenterY()['val'],
initialGuess.getCenterY()['fixed'],
initialGuess.getDistance()['val'],
initialGuess.getDistance()['fixed'],
initialGuess.getEnergy()['val'],
initialGuess.getEnergy()['fixed'],
initialGuess.getAlpha()['val'],
initialGuess.getAlpha()['fixed'],
initialGuess.getBeta()['val'],
initialGuess.getBeta()['fixed'],
initialGuess.getRotation()['val'],
initialGuess.getRotation()['fixed'],
initialGuess.getPixelLength()['val'],
initialGuess.getPixelHeight()['val'])
print 'Covariance Matrix'
print Numeric.array2string(covariance, precision=2)
print 'Root of the diagonal of the covariance matrix (I think these are uncertanties)'
print 'xc: ', sqrt(covariance[0][0])
print 'yc: ', sqrt(covariance[1][1])
print 'd: ', sqrt(covariance[2][2])
print 'E: ', sqrt(covariance[3][3])
print 'alpha: ', sqrt(covariance[4][4])
print 'beta: ', sqrt(covariance[5][5])
print 'rotation: ', sqrt(covariance[6][6])
# add the fit data to a new calibration file
ret = CalibrationData.CalibrationData()
ret.setCenterX(centerX, fixed=initialGuess.getCenterX()['fixed'])
ret.setCenterY(centerY, fixed=initialGuess.getCenterY()['fixed'])
ret.setDistance(distance, fixed=initialGuess.getDistance()['fixed'])
ret.setEnergy(energy, fixed=initialGuess.getEnergy()['fixed'])
ret.setAlpha(alpha, fixed=initialGuess.getAlpha()['fixed'])
ret.setBeta(beta, fixed=initialGuess.getBeta()['fixed'])
ret.setRotation(rotation, fixed=initialGuess.getRotation()['fixed'])
# pixel length & height don't change
ret.setPixelLength(initialGuess.getPixelLength()['val'])
ret.setPixelHeight(initialGuess.getPixelHeight()['val'])
return ret, peakList
def fit(data,initialGuess,peakList,verbose=1):
print
print 'Performing the image calibration'
# create a nice Numeric data strucure to store all of the peak stuff in.
# This lets us read our data easily once we get into C
length = len(peakList)
xValues = Numeric.zeros( (length), Numeric.Float64 )
yValues = Numeric.zeros( (length), Numeric.Float64 )
qReal = Numeric.zeros( (length), Numeric.Float64 )
intensity = Numeric.zeros( (length), Numeric.Float64 )
for peakIndex in range(len(peakList)):
xPeak,yPeak,Qreal,Qfit,chi,width = peakList[peakIndex]
xValues[peakIndex] = xPeak
yValues[peakIndex] = yPeak
qReal[peakIndex] = Qreal
intensity[peakIndex] = data[int(xPeak)][int(yPeak)]
# make C do the hard work :)
centerX,centerY,distance,energy,alpha,beta,rotation,covariance = FitWrap.fitCalibrationParameters(
xValues,yValues,qReal,intensity,
initialGuess.getCenterX()['val'],
initialGuess.getCenterX()['fixed'],
initialGuess.getCenterY()['val'],
initialGuess.getCenterY()['fixed'],
initialGuess.getDistance()['val'],
initialGuess.getDistance()['fixed'],
initialGuess.getEnergy()['val'],
initialGuess.getEnergy()['fixed'],
initialGuess.getAlpha()['val'],
initialGuess.getAlpha()['fixed'],
initialGuess.getBeta()['val'],
initialGuess.getBeta()['fixed'],
initialGuess.getRotation()['val'],
initialGuess.getRotation()['fixed'],
initialGuess.getPixelLength()['val'],
initialGuess.getPixelHeight()['val'])
print 'Covariance Matrix'
print Numeric.array2string(covariance,precision=2)
print 'Root of the diagonal of the covariance matrix (I think these are uncertanties)'
print 'xc: ',sqrt(covariance[0][0])
print 'yc: ',sqrt(covariance[1][1])
print 'd: ',sqrt(covariance[2][2])
print 'E: ',sqrt(covariance[3][3])
print 'alpha: ',sqrt(covariance[4][4])
print 'beta: ',sqrt(covariance[5][5])
print 'rotation: ',sqrt(covariance[6][6])
# add the fit data to a new calibration file
ret = CalibrationData.CalibrationData()
ret.setCenterX(centerX, fixed = initialGuess.getCenterX()['fixed'] )
ret.setCenterY(centerY, fixed = initialGuess.getCenterY()['fixed'] )
ret.setDistance(distance, fixed = initialGuess.getDistance()['fixed'] )
ret.setEnergy(energy, fixed = initialGuess.getEnergy()['fixed'] )
ret.setAlpha(alpha, fixed = initialGuess.getAlpha()['fixed'] )
ret.setBeta(beta, fixed = initialGuess.getBeta()['fixed'] )
ret.setRotation(rotation, fixed = initialGuess.getRotation()['fixed'] )
# pixel length & height don't change
ret.setPixelLength(initialGuess.getPixelLength()['val'])
ret.setPixelHeight(initialGuess.getPixelHeight()['val'])
return ret,peakList
