Leo fits #72

johbackm · 2025-10-14T20:33:28Z

First (beta) version of a working version of LEO fits for PySP2. Work is still needed on the details/tweaks, but the essentials should be there to expand on.

…cattering only particles

…IncandScatNumEnsemble with core and shell diameters.

…ult is Flase. Set to True for leo fits.

johbackm · 2025-10-14T20:40:39Z

Ah. I need to change the tests as well. Now beam_shape test removed for now.

rcjackson

Overall, this is really good. Some more documentation is needed, as well as unit tests for each of the LEO fit types. In addition, I request that commented out code be removed from the final version (comment notes are fine).

rcjackson · 2025-10-21T15:03:47Z

pysp2/util/leo_fit.py

-        my_high_gain_scatterers['PkHt_ch0'].values))/ \
-                            np.sum(my_high_gain_scatterers['PkHt_ch0'].values))
+    #high_gain_split_position = my_high_gain_scatterers['PkSplitPos_ch3'].values
+    #low_gain_split_position = my_low_gain_scatterers['PkSplitPos_ch7'].values


Please remove commented out code.

rcjackson · 2025-10-21T15:04:32Z

pysp2/util/leo_fit.py

+        #normalize the profile to range [~0,1]
+        profile_ch4 = (data_ch4 - base_ch4) / peak_height_ch4
+        #insert the profile as it was recorded (no shifting due to PEAK POSITION or PSD POSITION)
+        #my_high_gain_profiles_[i,:] = profile


More commented out code to remove at this line.

rcjackson · 2025-10-21T15:04:57Z

pysp2/util/leo_fit.py

+    #                                                                 moving_average_window, axis=0)
+    #moving_median_high_gain_base = np.nanpercentile(moving_high_gain_base, 10,axis=1)
+    #moving_median_low_gain_base = np.nanpercentile(moving_low_gain_base, 10,axis=1)
+


Past 4 lines here are commented out code that should not be in the final product if not used.

rcjackson · 2025-10-21T15:05:24Z

pysp2/util/leo_fit.py

+    #leo_Base_ch0[iloc_high_gain[:-moving_average_window+1]] = moving_median_high_gain_base
+    #leo_Base_ch4 = np.zeros(scatter_low_gain_accepted.shape)*np.nan
+    #leo_Base_ch4[iloc_low_gain[:-moving_average_window+1]] = moving_median_low_gain_base
+


Please remove this block of commented code.

rcjackson · 2025-10-21T15:06:37Z

pysp2/util/leo_fit.py

+    * LAST BIN TO USE FOR LEO FIT (from split detector)  (DONE)
+    * do also for LG.
+    * TAKE BASELINE SCAT FROM ACTUAL TRACERS IN LEO_FIT
+    """


I would put this into the docstring section under a new "Notes" section to inform the user about what is done and what needs to be implemented.

I removed this because it has already been implemented and those notes were mostly for my self.

rcjackson · 2025-10-21T15:07:20Z

pysp2/util/leo_fit.py

+    #Low gain
+    for i in range(my_binary.sizes['event_index']):
+        #max_value = data_ch0[i,:].max() - data_ch0[i,:].min()
+        #bins_ = bins[num_base_pts_2_avg:leo_fit_max_pos[i]]


Commented lines of code to remove.

rcjackson · 2025-10-21T15:07:52Z

pysp2/util/particle_properties.py

 from .DMTGlobals import DMTGlobals

-def calc_diams_masses(input_ds, debug=True, factor=1.0, Globals=None):
+def calc_diams_masses(input_ds, debug=True, factor=1.0, Globals=None, leo_fits=False):


Please add documentation for the leo_fits flag in the docstring below.

I added text to the documentation.

rcjackson · 2025-10-21T15:08:42Z

pysp2/util/particle_properties.py

+        print("Number of incandescence particles rejected for min. peak height = %d" % rejectMinIncandTotal)
+        print("Number of incandescence particles rejected for peak width = %d" % rejectIncandWidthTotal)
+        #print("Number of incandescence particles rejected for fat peak = %d" % rejectFatPeakTotal)
+        #print("Number of incandescence particles rejected for peak pos. = %d" % rejectFtPosTotal)


I would keep as much debug output as possible, so if these two new rejection criteria are recorded, I would keep them in.

Those commented out lines are from the scattering channels. So I removed them. They are shown earlier. The lines that are not commented out (98-100) are rejection criteria that have not previously been shown for incandescence.

rcjackson · 2025-10-21T15:08:57Z

pysp2/util/particle_properties.py

+        leo_Scat_not_sat = 1e-18 * (Globals.c0Scat1 + Globals.c1Scat1*leo_PkHt_ch0 + Globals.c2Scat1*leo_PkHt_ch0**2)
+        leo_Scat_sat = 1e-18 * (Globals.c0Scat2 + Globals.c1Scat2*leo_PkHt_ch4 + Globals.c2Scat2*leo_PkHt_ch4**2)
+        leo_Scatter = np.where(PkHt_ch0 < Globals.ScatMaxPeakHt1, leo_Scat_not_sat, leo_Scat_sat)
+        #leo_Scatter = np.where(accepted_incand, leo_Scatter, np.nan)


Please remove if not used.

rcjackson · 2025-10-21T15:10:02Z

tests/test_gfit.py

+    #coeff, beam_profile = pysp2.util.beam_shape(
+    #    my_binary, beam_position_from='peak maximum', Globals=pysp2.util.DMTGlobals())
+    #np.testing.assert_almost_equal(coeff, [9.83851858e-01, 4.64317390e+01, 
+    #                                       1.14337852e+01, 4.46761788e-03])


I would make unit tests for each of your beam_position_from flags.

Good point! Now I've added those as well.

…o match the rest of the code

johbackm · 2025-11-12T10:37:50Z

Now all the requested changes should have been done, including unit tests for the leo_fit() function.

Changed typo.

rcjackson · 2025-11-12T19:54:26Z

This looks good to go. I am merging. I will cut a new PySP2 release here shortly.

johbackm and others added 30 commits April 3, 2024 19:17

added leo_fit() from .leo_fit

ce50860

unfinnished leo fit code. Not working yet.

b23968b

Merge branch 'bugfix' into leo_fits

f281989

Merge branch 'ARM-DOE:main' into leo_fits

289e86b

Merge branch 'ARM-DOE:main' into leo_fits

eea85a5

leo_PkPos_ch0 for all particles using cross-to-centre distance from s…

2b21912

…cattering only particles

small edits and comments

fd3b4d8

added leo_PkFWHM, leo_MaxPos_ch0, leo_PkPos_ch0, etc.

7999b22

changed to leo_EndPos_ch0 for clarity

1ef8e85

changes to leo_fit()

0159dfc

changes to baseline for leo fits...

1b461b9

improvements

eee8d4c

leo amplitude factor for peak height added (testing)

f533374

xr.dims to xr.sizes

ba6f7e8

force 'fork' method with multiprocessing.Pool

0f865c1

bugfix leo peak amplitude factor

67b11a1

tidying up

1196753

adding incand stats

eb91757

started adding low gain stuff

ccc0c11

more low gain stuff

739c71e

low added to beam_shape

0171bc7

bug

ce4727b

bugfixes

c22cc3e

added ch4 to leo_fit

01313aa

bugfix for leo_AmpFactor_ch0 and leo_AmpFactor_ch4

863e1df

Changed description

355d03b

Changed description

f0c13f9

bugfix

ac480da

cleaning the code

69bd57e

cleaning up

40b8cd6

johbackm and others added 8 commits September 17, 2025 11:32

do not use bad split positions for beam profile and no negative values

f879c82

add leo diamteters if data is available

6737df3

added leo fits to process_psds() which returns a 3d array called leo_…

2ff285f

…IncandScatNumEnsemble with core and shell diameters.

added argument leo_fit to calc_diam_masses() and process_psds(). Defa…

df77f41

…ult is Flase. Set to True for leo fits.

commented out calculations that aren't needed

6e444b8

bug fixes

81f6b8d

removed index variables as they are not needed

fefb492

Merge branch 'ARM-DOE:main' into leo_fits

8ec9021

removed beam_shape tests for now

4e5f7f9

rcjackson requested changes Oct 21, 2025

View reviewed changes

johbackm added 7 commits November 10, 2025 14:04

updated documents

4ad8820

removed comments

be23638

added arguments and comments

66b5918

added text on leo_fits to docstring

7e2e09e

removing inf values if they exist

4306b05

added test for leo fits as well

2a5728f

changed variable names 'high_gain' to 'ch0' and 'low_gain' to 'ch4' t…

8f72046

…o match the rest of the code

Update processing_raw_file.rst

7de31cd

Changed typo.

rcjackson merged commit 38498aa into ARM-DOE:main Nov 12, 2025
11 checks passed

Leo fits #72

Leo fits #72

Uh oh!

Conversation

johbackm commented Oct 14, 2025

Uh oh!

johbackm commented Oct 14, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

rcjackson left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

johbackm commented Nov 12, 2025

Uh oh!

rcjackson commented Nov 12, 2025

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

johbackm commented Oct 14, 2025 •

edited

Loading