Devel updates to LST

[ariostas]

This PR adds the updates to LST/LSTCore that were developed while the CMSSW integration PR was being reviewed.

(We'll think of a better title and description when submitting to the upstream CMSSW repo.)

[slava77]

/run all

[github-actions]

The PR was built and ran successfully in standalone mode. Here are some of the comparison plots.

The full set of validation and comparison plots can be found here.

Here is a timing comparison:

   Evt    Hits       MD       LS      T3       T5       pLS       pT5      pT3      TC       Reset    Event     Short             Rate
   avg     39.7    396.0    194.0    127.9    135.0    503.7    117.1    247.0    101.7      2.7    1864.9    1321.6+/- 364.1     491.5   explicit[s=4] (target branch)
   avg     41.0    383.5    183.6    151.1    163.1    641.3    120.0    231.0    163.4      3.0    2081.1    1398.8+/- 393.4     552.1   explicit[s=4] (this PR)

[ariostas]

Here's a draft for the CMSSW PR. @slava77 @VourMa

Title: LST algorithm developments accumulated during integration PR

Description:

This PR integrates accumulated LST algorithm developments held back during the review of the LST integration (cms-sw#45117). The full list of PRs that are included here can be found in #117. This is our first step in our list of follow-up tasks (cms-sw#46746).

The changes include bug fixes and algorithmic improvements, but they also start to address follow-up comments from the integration PR. In particular, the LST readme was updated (addressing cms-sw#45117 (comment) and cms-sw#45117 (comment)) and the pT threshold is now configurable (addressing cms-sw#45117 (comment)).

[VourMa]

Here's a draft for the CMSSW PR. @slava77 @VourMa

Title: LST algorithm developments accumulated during integration PR

Description:

This PR integrates accumulated LST algorithm developments held back during the review of the LST integration (cms-sw#45117). The full list of PRs that are included here can be found in #117. This is our first step in our list of follow-up tasks (cms-sw#46746).

The changes include bug fixes and algorithmic improvements, but they also start to address follow-up comments from the integration PR. In particular, the LST readme was updated (addressing cms-sw#45117 (comment) and cms-sw#45117 (comment)) and the pT threshold is now configurable (addressing cms-sw#45117 (comment)).

Looks good to me! And it also fixes the first bullet point of this cms-sw#45117 (comment).

[github-actions]

The PR was built and ran successfully with CMSSW. Here are some plots.

OOTB All Tracks

The full set of validation and comparison plots can be found here.

[slava77]

Here's a draft for the CMSSW PR. @slava77 @VourMa

I'd add a plot from MTV to show the combined impact.

A slightly more expanded summary beyond the list of titles of other PRs would also be more self-contained and clear. Something like

• configurable and improved support of minimal pt for track reconstruction (expanding also to 0.6 GeV)
• several updates to T5 DNN selection now replacing most of the algorithmic/rectangular cuts including notebooks for documentation
• helix-based chi2 in rz improvement for pT5 and introduction for T3 objects
• other improvements and fixes, including developments in the standalone code

Most of these were described in TRK POG presentation on Sep 9

[slava77]

I'd add ...

@ariostas
if the text looks reasonable, please update the CMSSW PR description (it's usually better than a comment deeper in the thread).

Also, should this be made non-draft for our own bookkeeping ?