The JetMET Payload Inspector

The JetMET Payload Inspector includes Jet Resolution and Jet Corrector that allows detector experts to easily inspect, monitor and share the calibration constants stored in the CMS conditions database. The source code of the plugins can be found in !CondCore/JetMETPlugin . Interactive plots can be generated from user selected data.

The visualization part of this plugin is integrated with the cmsDbBrowser web-based application, which is the main entry point accessible for all CMS members to browse and manipulate conditions data and metadata. By visiting this page, one can see the full list of plugins in CMSSW.

How to check the plots?

Jet resolution object

• JetResolutionVsEta (Histo1D)
• JetResolutionVsPt (Histo1D)
  • The plots are simple 1D histogram that can quickly passed through JSON. User can choose proper bin values like jet pt, eta, rho. Then a dialog box will prompt you to select an iov to draw a plot. By clicking the "show plot" button, one can get the distribution of jet resolution with the selected parameters. The resolution is specified with the tag.

• JetResolutionSummary (Image)
  • For the plots in image format, they are generated with ROOT package with complicated contents and would take some time to see the result. User can choose proper bin values like jet pt, eta, rho. Then a dialog box will prompt you to select an iov to draw a plot. By clicking the "show plot" button, one can get the distribution of jet resolution with the selected parameters.

• JetResolutionCompare (Image)
  • For the comparison plot, user need to select two tags for comparison. User can also choose proper bin values like jet pt, eta, rho. Then a dialog box will prompt you to select the iovs to draw a plot. By clicking the "show plot" button, one can get the distribution of jet resolution with the selected parameters. The red dashed line becomes solid line as too fine binning.

• JetScaleFactorVsEtaDOWN (Histo1D)
• JetScaleFactorVsEtaNORM (Histo1D)
• JetScaleFactorVsEtaUP (Histo1D)
  • The plots are simple 1D histogram that can quickly passed through JSON. User can choose proper bin values like jet pt, eta, rho. Then a dialog box will prompt you to select an iov to draw a plot. By clicking the "show plot" button, one can get the distribution of jet resolution with the selected parameters. One need to use tag names with SF.

• JetScaleFactorVsPtDOWN (Histo1D)
• JetScaleFactorVsPtNORM (Histo1D)
• JetScaleFactorVsPtUP (Histo1D)
  • The plots are simple 1D histogram that can quickly passed through JSON. User can choose proper bin values like jet pt, eta, rho. Then a dialog box will prompt you to select an iov to draw a plot. By clicking the "show plot" button, one can get the distribution of jet resolution with the selected parameters. One need to use tag names with SF. Old tags may not have pt bin payload.

• JetScaleFactorSummary (Iamge)
  • For the plots in image format, they are generated with ROOT package with complicated contents and would take some time to see the result. User can choose proper bin values like jet pt, eta, rho. Then a dialog box will prompt you to select an iov to draw a plot. By clicking the "show plot" button, one can get the distribution of jet resolution with the selected parameters.

• JetScaleFactorCompare (Image)
  • For the comparison plot, user need to select two tags for comparison. User can also choose proper bin values like jet pt, eta, rho. Then a dialog box will prompt you to select the iovs to draw a plot. By clicking the "show plot" button, one can get the distribution of jet resolution with the selected parameters. The red dashed line becomes solid line as too fine binning.

Jet Corrector Parameters object

• JetCorrectorVsEtaL1FastJet (Histo1D)
• JetCorrectorVsEtaL1Offset (Histo1D)
• JetCorrectorVsEtaL1RC (Histo1D)
• JetCorrectorVsEtaL2Relative (Histo1D)
• JetCorrectorVsEtaL2L3Residual (Histo1D)
• JetCorrectorVsEtaL3Absolute (Histo1D)
• JetCorrectorVsEtaUncertainty (Histo1D)
  • The plots are simple 1D histogram that can quickly passed through JSON. User can choose proper bin values like jet pt, eta, rho. Then a dialog box will prompt you to select an iov to draw a plot. By clicking the "show plot" button, one can get the distribution of jet resolution with the selected parameters.

• JetCorrectorVsEtaSummary (Image)
  • For the plots in image format, they are generated with ROOT package with complicated contents and would take some time to see the result. User can choose proper bin values like jet pt, eta, rho. Then a dialog box will prompt you to select an iov to draw a plot. By clicking the "show plot" button, one can get the distribution of jet resolution with the selected parameters. The red dashed line becomes solid line in pt plots as too fine binning.

• JetCorrectorVsEtaCompare (Image)
  • For the comparison plot, user need to select two tags for comparison. User can also choose proper bin values like jet pt, eta, rho. Then a dialog box will prompt you to select the iovs to draw a plot. By clicking the "show plot" button, one can get the distribution of jet resolution with the selected parameters. One needs to specify the correction level curves to draw in either name string or ID:
    • L1Offset = 0,
    • L1JPTOffset = 7,
    • L1FastJet = 10,
    • L2Relative = 1,
    • L3Absolute = 2,
    • L2L3Residual = 8,
    • L4EMF = 3,
    • L5Flavor = 4,
    • L6UE = 5,
    • L7Parton = 6,
    • Uncertainty = 9,
    • L1RC = 34,
    • L1Residual = 35,

How to add new plots?

Condition proponents can easily add new custom plots depending on their needs to the Payload Inspector (PI) service by developing plugins in CMSSW framework.

Here are some brief details how to do it:

1. DPG/POGs needs to develop a plugin (with new plots) in CMSSW CondCore package.
2. Make a pull request and add it to a CMSSW pre-release
3. Once the new plots is included into the IB or pre-release, go to the Payload Inspector service and try to use it - it will automatically discover the newly added plots.

How to develop the PI plugin in CMSSW?

• Taking JetResolution PI plugin as an example: https://github.com/cms-sw/cmssw/blob/master/CondCore/JetMETPlugins/plugins/JetResolution_PayloadInspector.cc
• A plugin should be created for a payload object following the naming convention: XXXXX_PayloadInspector.cc .
• For each new plot, a class needs to be created, inherited the desired plot type. e.g.: class JetResolutionVsEta : public cond::payloadInspector::Histogram1D <JetResolutionObject, SINGLE_IOV> The class name is the plot name, which inherited the CondCore plot type and SINGLE_IOV or MULTIPLE_IOV. A third type argument can be given for two tag comparison.
• The plot name should be unique when stating in the constructor.
• A plot can take user inputs by stating cond::payloadInspector::PlotBase::addInputParam("MY_INPUT") .
  • The bool fill() function is to be overwritten to fill the plots with the de-serialized payload data with fillWithBinAndValue(BIN, VALUE);
  • For PlotImage type, one needs to create a ROOT TCanvas and draw desired contents and save it with canvas.SaveAs(m_imageFileName.c_str());
• Finally, this new class should be registered with a macro function like here:

    PAYLOAD_INSPECTOR_MODULE(JetResolutionObject) {
      PAYLOAD_INSPECTOR_CLASS(JetResolutionVsEta);
      ...
    }
  

-- YuanChao - 2023-08-14