Closed slava77 closed 7 years ago
A new Issue was created by @slava77 Slava Krutelyov.
@davidlange6, @smuzaffar, @Dr15Jones can you please review it and eventually sign/assign? Thanks.
cms-bot commands are listed here cms-sw/cmssw#13029
Inclusion of v4 training for the double b tagger for AK8 jets (for CA15 jets staying with v3 training for now). PR does not yet exist but there exists an 80X recipe that can easily be turned into a PR. However, before making a PR will need to propagate the new training file to externals.
Related to the mitigation of the ECAL gain switch problem:
@ferencek regarding https://github.com/cms-sw/cmssw/issues/16177#issuecomment-273555446 changing training from v3 to v4 will lead to changes in both data and MC. The early Feb re-miniAOD is for data. If a re-miniAOD of both data and MC happens before legacy rereco, this feature will be considered
@slava77, ok, so for now I will then postpone backporting https://github.com/cms-sw/cmssw/pull/17294 until we know what the situation will be with the re-miniAOD of both data and MC.
I have updated the PR description with a note on 03Feb2017 re-miniAOD and for now intend to keep the issue open to preserve tracking of other requests that didn't make it to 03Feb2017. https://cms-pdmv.cern.ch/pmp/historical?r=03Feb2017
issue to keep track of PRs that change miniAOD event content and aiming for a potential re-miniAOD in 80X of 2016 data and MC. There remains a possibility for the legacy rereco. The time scale is around March 2017.
Included in data re-miniAOD (03Feb2017) CMSSW_8_0_26_patch1
17296 [signed] ECAL / EGM: Remake ECAL cluster using modified rechits. Remake superclusters. Update electron/photon objects. Update PAT objects.
17308 [under review] Muon+JetMET part of the re-miniAOD fix
In case of full data and MC re-miniAOD
In case of data only re-miniAOD (after 03Feb2017)
17282 [signed] CTPPS: miniAOD