solution from particle added by add_tracking_particle is not obtainable with get_solution()

[lpyras]

I tried to calculate the contribution of D+ and D0 to the muon flux but adding the particle tracking did not work for me.
Here is the code snipped:
mceq_run.pman.add_tracking_particle([(411, 0), (421, 0)], (-13, 0), 'blabla', from_interactions=False) mceq_run.solve() blabla = mceq_run.get_solution('blabla', mag)

File "/Users/username/Software/MCEq/MCEq/core.py", line 244, in get_solution res = sum_lr(particle_name, prefix='') File "/Users/username/Software/MCEq/MCEq/core.py", line 203, in sum_lr result += sol[ref[prefix + ls].lidx:ref[prefix + ls].uidx] ValueError: operands could not be broadcast together with shapes (121,) (0,) (121,)

Can you tell me what I did wrong?

[afedynitch]

Hi! Which interaction model are you using? Any more details?

[afedynitch]

OK, I figure out what's not working. Since you're adding a new particle after everything has been reinitialized, the system size changed, so the dimensions need updating. After adding particles, you need to:

...add_tracking_particle...
mceq_run._resize_vectors_and_restore()
mceq_run.regenerate_matrices()

mceq_run.solve()
# etc...

How did you find the tracking_particle features? Was it documented somewhere?

[lpyras]

Thanks for your fast response. It works for me now for Sybill 2.3c and GSF.

I think I found it when I searched for the obs_particle function, and found that it has been replaced by the add_tracking_particle function. I don't think it was documented, the doc string is a bit outdated.
I think it would be helpful to add something like this to the examples file Partial_hadron_contribution.ipynb.

mceq_run = MCEqRun(
interaction_model='SIBYLL2.3c',
primary_model = (pm.GlobalSplineFitBeta, None),
theta_deg=60.0,
**config)

mceq_run.pman.add_tracking_particle([(411, 0), (421, 0)], (13, 0), 'Dpm0_mu-', from_interactions=False)
mceq_run.pman.add_tracking_particle([(-411, 0), (-421, 0)], (-13, 0), 'Dpm0_mu+')
mceq_run.pman.add_tracking_particle([(431, 0), (-431, 0)], (-13, 0), 'Ds_mu+')
mceq_run.pman.add_tracking_particle([(-431, 0), (431, 0)], (13, 0), 'Ds_mu-')
mceq_run.pman.add_tracking_particle([(310, 0)], (-13, 0), 'K0s_mu+')
mceq_run.pman.add_tracking_particle([(130, 0)], (-13, 0), 'K0L_mu+')
mceq_run._resize_vectors_and_restore()
mceq_run.regenerate_matrices()
mceq_run.solve()

mu_conv = mceq_run.get_solution('conv_mu+', mag) + \
          mceq_run.get_solution('conv_mu-', mag)

mu_prompt = mceq_run.get_solution('pr_mu+', mag) + \
          mceq_run.get_solution('pr_mu-', mag)
....

[afedynitch]

Thanks. I’ll add it to the ToDO

[The-Ludwig]

@lpyras I played a little bit with this undocumented feature, but never got it to work. Really happy about this issue, this helps me a lot! :)

[The-Ludwig]

Ah, I rediscovered what didn't work for me earlier.
Is is possible that the tracking of e.g. a short-lifed omega meson is not supported anymore since 1.0? @afedynitch

[afedynitch]

It is technically possible but a different data file is needed.