L1 emulator of topological trigger for cms-sw #49119
Conversation
|
cms-bot internal usage |
|
-code-checks Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-49119/46392
Code check has found code style and quality issues which could be resolved by applying following patch(s)
|
|
+code-checks Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-49119/46432
|
|
Pull request #49119 was updated. |
LukasEbeling
changed the title
|
To be done by OffSW conveners: a cmsdist entry for TOPO like for AXO https://github.com/cms-sw/cmsdist/blob/IB/CMSSW_16_0_X/master/AXOL1TL.spec |
|
somehow I don't see the usual bot message asking for the L2 review .. @BenjaminRS @quinnanm |
|
A new Pull Request was created by @LukasEbeling for master. It involves the following packages:
@BenjaminRS, @cmsbuild, @quinnanm can you please review it and eventually sign? Thanks. cms-bot commands are listed here |
|
it seems to have just come through now - thanks |
|
please test |
|
+1 Size: This PR adds an extra 16KB to repository Comparison SummarySummary:
|
| theCondition = std::make_unique<TOPOCondition>(itCond.second, this); | ||
| theCondition->setVerbosity(m_verbosity); | ||
|
|
||
| if (m_saveAXOScore and not m_axoScoreConditionName.empty()) { |
There was a problem hiding this comment.
Is it right that this is using m_saveAXOScore and not m_axoScoreConditionName.empty()? should these be m_saveTOPOScore and not m_topoScoreConditionName.empty()
| } | ||
|
|
||
| //setModelVersion - set the model version of the condition | ||
| void TOPOTemplate::setModelVersion(const std::string& modelversion) { m_modelVersion = modelversion; } |
There was a problem hiding this comment.
I'm guessing you tested getting the model version from a custom xml file? can you describe those tests? and do you have a link to the cms-dist model release?
5 participants
PR description
This PR adds the L1 emulator code for the topological trigger targeting the single-muon finale state of the hhbbww process.
PR validation
the final dnn score outputted by the cms-sw L1 emulator agrees well with the keras model training in a python environment. At relevant scores above 0.8, only deviations of the order two percent are expected: