Photon peak tutorial for sector alignment

[kutnyakhov]

Added photon peak tutorial and new Photon_peak dataset (available on Zenodo) to fix #565

[rettigl]

I fixed your configs etc. now, but honestly I don't understand what you are doing here. You don't actually aligne the DLD sectors, as along as I can tell, nor generate the parameters for it...
Ideally, add a routine to the processor to generate those parameters (semi) automatically.

[coveralls]

Pull Request Test Coverage Report for Build 14502539411

Details

• 0 of 0 changed or added relevant lines in 0 files are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage remained the same at 92.063%

---

Totals
Change from base Build 14445750250:	0.0%
Covered Lines:	8166
Relevant Lines:	8870

---

💛 - Coveralls

[rettigl]

@zain-sohail @kutnyakhov The pulserSignAdc channel seems to cause problems for these files. What is it good for anyways? It is not being used anywhere in the tutorials. @steinnymir added it here: 0ef68d5

[kutnyakhov]

@zain-sohail @kutnyakhov The pulserSignAdc channel seems to cause problems for these files. What is it good for anyways? It is not being used anywhere in the tutorials. @steinnymir added it here: 0ef68d5

I agree that it is not needed for tutorials, but we have it from time to time in the datastream, e.g. last time it was used for the magnetic pulser during spin-resolved beamtime. A long time ago, that channel was used for the I0 monitor of an optical laser, but now it has its own channel opticalDiode.

[kutnyakhov]

I fixed your configs etc. now, but honestly I don't understand what you are doing here. You don't actually aligne the DLD sectors, as along as I can tell, nor generate the parameters for it... Ideally, add a routine to the processor to generate those parameters (semi) automatically.

This is basically my problem, that I was not able to understand what align_dld_sectors() is doing. Even if I was trying to feed in some crazy values, like sector_delays= [3050., 10000., 5000., 4000., 15000., 0., 0., 0.], but I could not see any changes. For me, it seems that something is missing in the code. We discussed this with @zain-sohail and I guess he had some ideas what could it be

[rettigl]

I fixed your configs etc. now, but honestly I don't understand what you are doing here. You don't actually aligne the DLD sectors, as along as I can tell, nor generate the parameters for it... Ideally, add a routine to the processor to generate those parameters (semi) automatically.

This is basically my problem, that I was not able to understand what align_dld_sectors() is doing. Even if I was trying to feed in some crazy values, like sector_delays= [3050., 10000., 5000., 4000., 15000., 0., 0., 0.], but I could not see any changes. For me, it seems that something is missing in the code. We discussed this with @zain-sohail and I guess he had some ideas what could it be

It works quite correctly, however it is being applied to the dldTimeSteps, i.e. you have to call it before conversion to ns (actually not needed at all). See updated example.

[kutnyakhov]

I fixed your configs etc. now, but honestly I don't understand what you are doing here. You don't actually aligne the DLD sectors, as along as I can tell, nor generate the parameters for it... Ideally, add a routine to the processor to generate those parameters (semi) automatically.

This is basically my problem, that I was not able to understand what align_dld_sectors() is doing. Even if I was trying to feed in some crazy values, like sector_delays= [3050., 10000., 5000., 4000., 15000., 0., 0., 0.], but I could not see any changes. For me, it seems that something is missing in the code. We discussed this with @zain-sohail and I guess he had some ideas what could it be

It works quite correctly, however it is being applied to the dldTimeSteps, i.e. you have to call it before conversion to ns (actually not needed at all). See updated example.

Thanks for the details and updates. Now I see that it is doing something. The point with the ns conversion was to show some meaningful values, which everyone can understand and have direct comparison to, e.g. DLD time resolution.
Probably we have to add one more explanation line in the config file related to this part, that sector_delays should be configured/filled with proper numbers (in steps) before applying this function - somehow, from the explanation in tutorial 4, it sounded that this is done automatically.

[rettigl]

If you don't provide values, it takes those from the config file, if they exist. Otherwise it does nothing.

[kutnyakhov]

I guess it can be merged.
P.s. @rettigl, why don't I see a tutorial for Moritz in the docs?

[rettigl]

P.s. @rettigl, why don't I see a tutorial for Moritz in the docs?

See the logs:

Notebook error:
NoSuchKernel in tutorial/12_momentum_correction_of_binned_data.ipynb:
No such kernel named sed_kernel_v1

Your latest changes overwrote the kernel line, and I overlooked it. Will fix it here.

[rettigl]

I agree that it is not needed for tutorials, but we have it from time to time in the datastream, e.g. last time it was used for the magnetic pulser during spin-resolved beamtime.

I disabled this now in the general config, if it is needed it can be undocumented per experiment

[rettigl]

Good to go now