Add Dplus comparison plots

[fcatalan92]

Needs alisw/AliPhysics#16108 (merged), alisw/AliPhysics#16118 (merged) and AliceO2Group/AliceO2#4797 (merged)

[ginnocen]

@vkucera @fcatalan92 is this good to go?

[fcatalan92]

@vkucera @fcatalan92 is this good to go?

Not yet, we need to merge first the O2 one and probably update some histogram names in this PR

[ginnocen]

OK.

[vkucera]

This PR is not compatible with the latest AliPhysics and O2. Is there an update planned?

[fcatalan92]

This PR is not compatible with the latest AliPhysics and O2. Is there an update planned?

Hi @vkucera, yes I planned to update it but I didn't manage yet. I will do in the next days

[vkucera]

This PR is not compatible with the latest AliPhysics and O2. Is there an update planned?

Hi @vkucera, yes I planned to update it but I didn't manage yet. I will do in the next days

OK, converting to a draft until then.

[fcatalan92]

@vkucera I updated the PR and now the code runs fine. However, the comparison plots are in strong disagreement
comparison_histos_2prong.pdf, comparison_histos_3prong.pdf. It looks like the configurations introduced in #117 are not picked-up/used by the O2 software, is this expected?

P.s. I'm running with updated AliPhysics and O2 (today's master)

[vkucera]

Please rebase from the master, fix conflicts and force push to enable the linter checks.

[ginnocen]

hi @fcatalan92, could you rebase and resolve the conflicts so that we can merge this PR?

[fcatalan92]

hi @fcatalan92, could you rebase and resolve the conflicts so that we can merge this PR?

Done.

I have also opened a PR in O2 AliceO2Group/AliceO2#5204 adding the use of pre-selections for the Dplus. Now the comparison plots are in better agreement comparison_histos_dplus.pdf with respect to the old version comparison_histos_dplus_noPreSel_beforePR.pdf even if some discrepancies are still present. The default configuration returns quite different results for Lc comparison_histos_lc.pdf and D0 comparison_histos_d0.pdf between AliPhysics and O2.

An additional minor thing, running the runtest.sh script I get warnings and errors (terminal_out.txt) even if everything works fine and I get the comparison plots. The log files do not show anything strange.

[ginnocen]

hi @fcatalan92, thanks. "The default configuration returns quite different results for Lc comparison_histos_lc.pdf and D0 comparison_histos_d0.pdf between AliPhysics and O2" -> yes this is correct. But It is overall the same issue. It is the contamination of candidates from other decay channels (in the Lc, you see e.g. the Ds and in the D0 you see the Jpsi channel), which is big if you don't apply selections.

[ginnocen]

hi @fcatalan92 could you try to set dcatoprimxymin_3prong to 0 in your JSON config (if not done already?).
There is a known difference in the dca single track selection between run2 and run3. By setting this value to 0 and using your latest O2 and comparison PRs, I get decent results (apart from discrepancies on the error estimation of the decay vertices, which in my understanding is a known issue). https://www.dropbox.com/s/6vo2fw57wtvaulx/2_comparison_histos_dplus.pdf?dl=0

[vkucera]

Hi @fcatalan92

The log files do not show anything strange.

What is in the comparison log file?

[fcatalan92]

What is in the comparison log file?

There is an error related to a histogram binning (log_postprocess.log) indeed. However, it seems to me that spurious warnings and errors, that do not affect the workflow, are caught.

@ginnocen Setting dcatoprimxymin_3prong to 0, I get a result compatible with yours comparison_histos_dplus.pdf.

In my opinion, this PR is ready to be merged.

[ginnocen]

hi @fcatalan92 thanks for checking, it seems all consistent. Vit is looking into those warnings that seem indeed to be not very informative. I will merge this one.

[vkucera]

@fcatalan92 I see. The binning mismatch is coming from the skimming, not your task, so that is fine.

Please remove the reference plots from your PR. They are obsolete now and should be replaced with the per-task plots once we reach a decent AliPhysics/O2 agreement.

[ginnocen]

@fcatalan92 why are you updating the reference plots?

[fcatalan92]

Please remove the reference plots from your PR. They are obsolete now and should be replaced with the per-task plots once we reach a decent AliPhysics/O2 agreement.

Done, they were a remnant of the first version of the PR

[vkucera]

Thanks, looks good to me.
Just a minor fix. Can you please fix these? cm^2 -> cm^{2}

[fcatalan92]

Thanks, looks good to me.
Just a minor fix. Can you please fix these? cm^2 -> cm^{2}

Done.

[vkucera]

Thanks a lot for the quick fix!

[ginnocen]

hi @vkucera just wait for #5204 O2 to be merged.