HDF_Eventutils : Algorithms used for X-ray event processing

tesfdmtools.methods.HDF_Eventutils.crosscplt(ax, hdf, channel, freq, c1, c2, xlabel='c1', ylabel='c2', center=None, ytilt=None)

plot correlation between two parameters and the fitted linear correlation relation

Args:

• ax = a plotting instance for the plot output

• hdf = HDF5 input file object

• channel = channel number being processed

• freq = frequency number (pixel) being processed

• c1 = array of first parameter

• c2 = array of second parameter

Kwargs:

• xlabel = label for x-axis [default: ‘c1’]

• ylabel = label for y-axis [default: ‘c12’]

• center = fitted center of distribution [default: None]

• ytilt = fitted linear slope between parameters [default: ‘None’]

Returns:

none

tesfdmtools.methods.HDF_Eventutils.dellist(delist, nrecs)

make logical array specifying which event records to use from HDF file

Args:

• delist = full file name which hold the records which decribe which events to ignore

• nrecs = size of array to make

Returns:

• evuse = logical array with True values for events to use

tesfdmtools.methods.HDF_Eventutils.eventlist_out(wfile, freq, recordno, optfit, baseline, peaks, hdffilename, channel=None, rtimes=None, ftimes=None, evttim=None, eefit=None, surf=None)

write eventlist to file

Args:

• wfile = write eventlist or not

• freq = frequency number in file

• recordno = list of record numbers

• optfit = list of optimal filtering results for each event

• baseline = list of baseline levels for each event

• peaks = list of pulse peak values for each event

• hdffilename = full filename (including path) of processed HDF file

Kwargs:

• channel = channel number used

• rtimes = pulses rise times

• ftimes = pulses fall times

• evttim = time of the events

• eefit = list of (calibrated) energies for the events

• surf = list of pulse surfaces for each event

tesfdmtools.methods.HDF_Eventutils.eventpars(hdf, channel=None, freq=None, threshold=2000, bsec=0.05, nrecs=None, rotation=0.0, bpt=False, delist=None, flip=False)

get basic event record parameters for further event selection. Only the I-signal is considered

Args:

• hdf = HDF5 input file object

Kwargs:

• channel = channel number to process. If Nont, take first channel [default: None]

• freq = frequency (pixel) number to process. If None, take first frequency [default: None]

• threshold = consider only records which have a max-min larger than threshold [default: None]

• bsec = part of record (start and end) from which baselines are derived. (This part of the record should not contain a pulse or partial pulse) [default: 0.05]

• nrecs = number of records to process. If None, process all records. [default: None]

• rotation = if not zero, rotate events with this amount [default: 0.0]

• bpt = if True, select pretrigger section of event only for baseline [default: False]

• flip = flip record data

Returns:

• channel = channel number which was processed

• freq = frequency (pixel number) which was processed

• blines1 = background levels at start of record

• blines2 = background levels at end of record

• positions = position of pulse minimum (pulse runs negative)

• peaks = value of pulse minimum (value at maximum pulse signal)

• surf = surface of pulse minus background. Bacground is linear fit between blines1 and blines2

• ptp = maximum signal difference within record (max - min)

• evttim = event times (time of event peak value in seconds)

• rsize = length of record (number of samples in record)

tesfdmtools.methods.HDF_Eventutils.fitspectrum(pltdata, iplt, hdf, channel, freq, spectrum, bins, debug=False, afitrange=[5.86, 5.92])

fit the Holzer alpha and beta Mn54 lines to the spectrum

Args:

• pltdata = dictionary for the fit results

• iplt = number of the plot to process

• hdf = HDF5 input file object

• channel = channel number being processed

• freq = frequency number (pixel) being processed

• spectrum = spectrum of the pulse fit parameters

• bins = centers of the spectrum bins

Kwargs:

• debug = if True, plot debug information [default: False]

• afitrange = range (keV) for spectrum fit [default: 5.860,5.920]

Results:

• pltdata[iplot] = spectrum fit results

Returns:

• fwhm = fitted resolution inFWHM

• fitpars = additional fitted parameters

tesfdmtools.methods.HDF_Eventutils.getphase(hdf, channel, freq, indx, nevents=100, debug=False)

Get average phase of first ‘nevents’ events

Args:

• hdf = HDF5 input file object

• channel = channel number being processed

• freq = frequency number (pixel) being processed

• indx = index of selected events, to be processed

Kwargs:

• nevents = number of events to process [default: 100]

• debug = if True, show some events [default: False]

Returns:

• phase = average phase correction for events

tesfdmtools.methods.HDF_Eventutils.getrisetime(xax, pulse, debug=False, plotfit=None)

compute rise time for pulse, based on linear fit of log values

Args:

• xax = X-axis of pulse expressed in sample #

• pulse = pulse shape, with background subtracted

Kwargs:

• debug = if True, plot each pulse and fit [default: False]

• plotfit = plotinstance for log plot of fitted pulse [default: None]

Returns:

• rtime = rise time of pulse in sample units

• ftime = fall time of pulse in sample units

tesfdmtools.methods.HDF_Eventutils.limdist(cc, ext=0.5, bins=2500)

compute range of main data distribution in array of values

Args:

• cc = array of values

Kwargs:

• bins = number of bins to use for histogram

• ext = range extension [default: 0.5]

Returns:

• xmin = minimum of range

• xmax = maximum of range

tesfdmtools.methods.HDF_Eventutils.makeplot(pdf, ps=False)

finish plot

Args:

• pdf = plot output file

Kwargs:

• ps = if False, also put output on screen [default: False]

tesfdmtools.methods.HDF_Eventutils.makespectrum(pulsefits, sbins=5000, debug=False, efrac=0.05)

make spectrum of fitted optimal filtering scale factors, and determine proper range (leave out the outliers)

Args:

• pulsefits = array of pulse fit paramters (scale factors)

Kwargs:

• sbins = number of bins for spectrum [default: 5000]

• debug = if True, plot/print debug output [default: False]

• efrac = fraction of the events at low and hig end defined as outliers [default: 0.05]

Returns:

• spectrum = spectrum of the pulsefits

• bins = centers of the spectral bins

tesfdmtools.methods.HDF_Eventutils.noisefile(hdf, channel, freq, nfile=None)

See if there is an associated noise file, or a noise file is given. If so , open file and retrieve ptp info

Args:

• hdf = HDF5 input xray file

• channel = channel number to use

• freq = frequency number to use

Kwargs:

• nfile = noise file to use, or True to ask for noise file [default: None]

Returns:

• nhdf = HDF5 file object for noise data

• ptp = max-min signal fluctuations for the records

tesfdmtools.methods.HDF_Eventutils.noisespec(hdf, channel, freq, ptp, fraction=0.1, debug=False, absolute=False, prlen=None, flip=False)

Retrieve noise spectra. Use given fraction of I-signal records with least max-min signal fluctuations.

Args:

• hdf = HDF5 input file object

• channel = channel number to use

• freq = frequency (pixel) number to use

• ptp = max-min signal fluctuations for the records

Kwargs:

• fraction = fraction of least signal fluctuations to use [default: 0.1]

• debug = if True, print/plot debug info [default: False]

• prlen = length of record to take [default: None]

• flip = flip record data

Returns:

• noisspec = average noise spectrum

• fax = frequency axis for noise spectrum

tesfdmtools.methods.HDF_Eventutils.noisplot(pltax, spectrum, nspectrum, fax, title)

plot noise spectrum

Args:

• pltax = plot instance for plotting

• spectrum = spectrum data

• fax = frequency axis

• title = title of plot

tesfdmtools.methods.HDF_Eventutils.optfilter(hdf, channel, freq, indx, base1, base2, noisspec, debug=False, freqcutoff=None, rotate=False, risetime=False, bsec=0.05, absolute=False, prlen=None, apulsepos=None, flip=False, showpspec=None, **kwargs)

perform optimal filtering fit of individual pulses

Args:

• hdf = HDF5 input file object

• channel = channel number being processed

• freq = frequency number (pixel) being processed

• indx = index of selected events, to be processed

• base1 = baseline level at start of record

• base2 = baseline level at end of record

• noisspec = noise spectrum

Kwargs:

• debug = if True, plot various fit parameters [default: False]

• freqcutoff = cutoff frequency (Hz) for optimal filtering [default: None]

• rotate = rotation angle (radians) for I/Q pulse record

• risetime = it True, compute rise time of pulses [default: None]

• bsec = section of record to take for background (only used for ‘rotate’ or ‘absolute’) [default: 0.05]

• absolute = use sqrt(I^2+Q^2) signal [default: False]

• prlen = length of record to process (None = entire record) [default: None]

• apulsepos = average position of pulse maximum; used for limited prlen. [default: None]

• flip = flip record data

• showpspec = interactive plot of a given set of pulse spectra

Returns:

• ifit = fitted pulse heigth parameters

• rtimes = computed rise times of fitted pulses

• ftimes = computed fall times of fitted pulses

• avpulse = average pulse profile with baseline subtracted

• avbline = average baseline for the average pulse prfile

tesfdmtools.methods.HDF_Eventutils.plotepar(hdf, channel, freq, bline, positions, peaks, surf, show=None, pdf=None, indx=None, ectpar=None)

Plot basic event record parameters

Args:

• hdf = HDF5 input file object

• channel = channel number

• freq = frequency (pixel) number

• bline = baseline level

• positions = pulse positions

• peaks = pulse minima (pulse peaks)

• surf = pulse surface

Kwargs:

• show = plot mode [default: None] :

  • ‘X’ : both show on screen and print to pdf file

  • ‘screen’ : show on screen only

  • ‘pdf’ : print to pdf file

  • None : do not plot

• pdf = pdf object for plotting [default: None]

• indx = indexnumbers of records used

• ectpar = electric crosstalk parameters (tuple):

  • ectpar[0] = actual list of selected records

  • ectpar[1] = actual list of parameters for the indexed records

  • ectpar[2] = actual threshold applied for the electric crosstalk

Returns:

none

tesfdmtools.methods.HDF_Eventutils.plotptimes(rtimes, ftimes, nfit, avpulse, ptitle, pdf, ps=False, outtxt=False, erange=[None, None], avbline=None)

Make plot of pulse times (rise and fall) and correlations with energy + average pulse profile Also write ASCII data of average pulse profile when outtxt=True

Args:

• rtimes = pulse rise times for the events

• ftimes = pulse fall times for the events

• nfit = optimum filter fits for the events (scales with energy)

• avpulse = the average pulse profile

• ptitle = title of the plot page

• pdf = pdf plot instance

Kwargs:

• ps = if False, also show plot interactively [default: False]

• outtxt = if True, write ASCII data of average pulse profile [default: False]

• erange = range of the nfit parameter to plot in the energy/(rise/fall)time correlation [default: None,None]

• avbline = average baseline level for the average pulse. To be included in the ASCII pulse profile [default: None]

tesfdmtools.methods.HDF_Eventutils.selectdtim(indx, evttim, etprev=[None, None], etnext=[None, None])

Select events based on the delta times to the previous and next events

Args:

• indx = index of valid events

• evttim = time tages of the events

Kwargs:

• etprev = selection window of delta time to the previous event [min,max]

• etnext = selection window of delta time to the next event [min,max]

tesfdmtools.methods.HDF_Eventutils.selectevs(hdf, channel, freq, bline, positions, peaks, surf, rclen, show=None, debug=False, nopos=False, pdf=None, nopulse=False, bsec=0.05)

select the proper events to process from the basic event record parameters

Args:

• hdf = HDF5 file which is being used

• channel = channel number whch was processed

• freq = frequency (pixel) number which was processed

• bline = baseline levels of the records

• positions = pulse positions for the records

• peaks = record minima (pulse peak values)

• surf = pulse surfaces

• rclen = record length (number of samples in record)

• nopulse = do not select on pulse intensity when set True

Kwargs:

• show = plot mode [default: None]:

  • ‘X’ : both show on screen and print to pdf file

  • ‘screen’ : show on screen only

  • ‘pdf’ : print to pdf file

  • None : do not plot

• debug = if True, print/plot some debug info [default: False]

• nopos = if True, do not select pulses on pulse position [default: False]

• bsec = section of record (baseline section) where pulse should not occur when nopos

• pdf = pdf object for plotting [default: None]

Returns:

• sindx = list of selected record numbers to use for further processing

tesfdmtools.methods.HDF_Eventutils.selecthist(data, tlevel=0.01, debug=False, label='')

compute data range which selects the main body of the data, leaving out data values where data become sparse

Args:

• data = array of input values

Kwargs:

• tlevel = select only data where data density is above this level [default: 0.01]

• debug = if True, plot debug info [default: False]

• label = plot label for debug plot [default: “”]

Returns:

• lowb = low range value

• higb = high rang evalue

tesfdmtools.methods.HDF_Eventutils.showcorr(c1, c2, xlabel='X', ylabel='Y', title='', center=None, ytilt=None)

plot correlation between two parameters and the fitted linear correlation relation

Args:

• c1 = array of first parameter

• c2 = array of second parameter

Kwargs:

• xlabel = label for x-axis [default: ‘X’]

• ylabel = label for y-axis [default: ‘Y’]

• title = title for plot [default: ‘’]

• center = fitted center of distribution [default: None]

• ytilt = fitted linear slope between parameters [default: None]

Returns:

none

tesfdmtools.methods.HDF_Eventutils.subsel(indx, par, prange=[None, None])

select a subset within the given parameter range

Args:

• indx = index of the parameters to be considered

• par = list of parameter values

• prange = selection range of the parameters

Returns:

• sindx = new index selecting only the relevant items

tesfdmtools.methods.HDF_Eventutils.timplt(ax, hdf, channel, freq, c1, c2, frange=None, xlabel='c1', ylabel='c2')

time evolution plot of pulse fit parameter

Args:

• ax = a plotting instance for the plot output

• hdf = HDF5 input file object

• channel = channel number being processed

• freq = frequency number (pixel) being processed

• c1 = array of time parameter (event record number)

• c2 = array of pulse fits

Kwargs:

• frange = range of fits parameter values to plot [default: None]

• xlabel = label for x-axis [default: ‘c1’]

• ylabel = label for y-axis [default: ‘c2’]

Returns:

none

tesfdmtools.methods.HDF_Eventutils.xoutl(values)

find reasonable index which excludes outliers

Args:

• values = array of values

Returns:

• rindx = index which excludes outliers