tdl/Modules/CtrData

Important Modules / Packages

modules/xray/ana/ctr_data.py
modules/xray/ana/scan_data.py
modules/xtal
See the Ana Data page for additional information on reading scans

Notes

The CtrData class holds all the scandata and other parameters for a CTR data set. The class instance can be created via:

 >>ctr = scandata.CtrData(scans=[],I='I',Inorm='io',Ierr='Ierr',
                          Ibgr='Ibgr',corr_params={},scan_type='image')

scans = list of scan data instances
I = string label corresponding to the intensity array, ie let y be the intensity, then
```
y = scan[I]
```
Inorm=string label corresponding to the normalization intensity array, ie, norm = scan[Inorm], where the normalized intensity is taken as
```
 ynorm= y/norm
```

Ierr = string label corresponding to the intensity error array, ie y_err = scan[Ierr]. We assume these are standard deviations of the intensities (y).

  Note when the data are normalized we assume the statistics of norm 
  go as norm_err = sqrt(norm).  Therefore the error of normalized
  data is
    ynorm_err = ynorm * sqrt( (y_err/y)^2 + (norm_err/(norm))^2 )
              = ynorm * sqrt( (y_err/y)^2 + 1/norm )
  If its assumed that y_err = sqrt(y) then the expression could be
  simplified futher, but we wont make that assumption since there
  may be other factors that determine how yerr was calculated.

Ibgr = string label corresponding to background intensity array

corr_params = a dictionary containing the necessary information for data corrections.

 corr_params['geom'] = Type of goniometer geometry ('psic' is default)
 corr_params['beam_slits'] = dictionary describing the beam slits,e.g.
                             {'horz':.6,'vert':.8}
 corr_params['det_slits'] = dictionary describing the beam slits,e.g.
                            {'horz':.6,'vert':.8}
 corr_params['sample'] = a dictionary describing the sample shape.
                         {'dia':0,'angles':{},'polygon':[]}
                         if dia >=0 then assume a round sample.
                         Otherwise use polygon/angles
                         If all are None, then no sample correction is
                         computed
 corr_params['scale'] = scale factor to multiply by all the intensity
                        values. e.g.  if Io ~ 1million cps
                        then using 1e6 as the scale makes the normalized
                        intensity close to cps.  ie y = scale*y/norm

scan_type = Type of scans (e.g. 'image', 'phi', etc..)

An example session: reading/parsing spec files and integrating images

The following script reads several scans into a list of scandata objects
Note that the images will be cached in an hdf file with the specified path (rather than cached in memory). See the note further below about editing the path if you move the files.

All the read in images will have the specified rois

########################################
# script to read spec files and images 
# into a list of scan objects 
########################################

# create an instance of the reader object
spec_file = ana.Reader(spec_path = '../',spec = 'caco3_feb09j.spc')
spec_file.spec_params['image'] = True
spec_file.image_params['archive'] = {'path':'d:\\tmp'}
spec_file.image_params['rois']   = [180,70,277,126]
print spec_file

# Make a list of selected scans, and plot
S = []
first_scan = 1
last_scan = 8
pyplot.clf()
j = 0
for i in range(first_scan, last_scan+1):
  S.append(spec_file.spec_scan(i, image = True))
  pyplot.plotter(S[j]['L'],S[j]['IROI']/S[j]['io'],ylog=True)
  j = j+1

#end of script

Now create a CtrData object

>>corr_params = {}
>>corr_params['geom'] = 'psic' 
>>corr_params['beam_slits'] = {'horz':.6,'vert':.8}
>>corr_params['det_slits']  = None
>>corr_params['sample'] = 10.
>>corr_params['scale'] = 1.e6
>>ctr = ana.CtrData(S,I='I',Inorm='io',Ierr='Ierr',Ibgr='Ibgr',corr_params=corr_params)

make plots and write files

>>ctr.plot()
>>ctr.plot_I()
>>ctr.write_HKL(fname='data.lst')

If you want to add some new scan data (in this case from a new file - note all other parameters will stay as above)

Also note in the example, we are adding multiple scans as a list, you can also just add a single scan

>>S2 = []
>>S2.append(spec_file.spec_scan(23,file='caco3_feb09x.spc'))
>>S2.append(spec_file.spec_scan(24))
>>S2.append(spec_file.spec_scan(25))
# 
>>ctr.append_scans(S2)

You can save the ctr data set in a pickle (see tdl/Pds/Files) using:

>>save ctr my_ctr_data.sav

- to restore

>>>restore my_ctr_data.sav

If you have archived your images and later move them you'll need to update the path. this can be done by editing the ctr object as:
```
>>for j in range(len(ctr.scan)):
...   ctr.scan[j].image.image.path = 'c:\\my_path\\to\\images'
>>
```
(this will be improved at a later date!)

Another Example Session

Creat a new reader
```
>>r = ana.Reader(spec_path='spec/')
```

Set image parameters - archive directory and integration params

>>r.image_params['archive'] = {'path':'d:\\tmp'}
>>r.image_params['rois']    = [180,70,277,126]

Create some correction params, including a sample polygon

corr_params = {}
corr_params['geom'] = 'psic' 
corr_params['beam_slits'] = {'horz':.6,'vert':.8}
corr_params['det_slits']  = None
polygon = [[1.,1.], [.5,1.5], [-1.,1.], [-1.,-1.],[0.,.5],[1.,-1.]]
angles = {'phi':108.0007,'chi':0.4831}
sample = {'polygon':polygon,'angles':angles}
corr_params['sample'] = sample
corr_params['scale'] = 1.e6

Now read some data

>>f = 'caco3_feb09f.spc'
>>j = 'caco3_feb09j.spc'
>>s1 = []
>># 00L
>>s1.append(r.spec_scan(1, file=j, image = True))
>>s1.append(r.spec_scan(2))
>>s1.append(r.spec_scan(3))

Create a CtrData object ad plot it

>>ctr = ana.CtrData(s1,I='I_r',Inorm='io',Ierr='Ierr_r',Ibgr='Ibgr_r',
                    corr_params=corr_params)

>>ctr.plot(num_col=1)

Now add some more scans and plot

>># 21L
>>s2 = []
>>s2.append(r.spec_scan(23,file=f))
>>s2.append(r.spec_scan(24))
>>s2.append(r.spec_scan(25))
>>ctr.append_scans(s2)
>>ctr.plot(num_col=2)