You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
From my CMS analysis it's becoming clear how useful a feature it would be to help us optimise event selections. A traditional technique has been to produce so called "n-1" plots, where for each cut in the event selection, you remove that cut and plot the corresponding variable with all the other cuts remaining.
To support this in a single stage, the stage needs to have:
The full event selection
Binning parameters for each variable, or at least a default binning scheme
optionally the name of an output mask for the full event selection.
An alternative might be to just have this stage produce many masks, one per "n-1" cut flow and then we add a mask option to binned dataframes so that we use the existing binned dataframe stages to handle the actual plots. This might give more flexibility and allow for more code re-use.
The text was updated successfully, but these errors were encountered:
From my CMS analysis it's becoming clear how useful a feature it would be to help us optimise event selections. A traditional technique has been to produce so called "n-1" plots, where for each cut in the event selection, you remove that cut and plot the corresponding variable with all the other cuts remaining.
To support this in a single stage, the stage needs to have:
An alternative might be to just have this stage produce many masks, one per "n-1" cut flow and then we add a mask option to binned dataframes so that we use the existing binned dataframe stages to handle the actual plots. This might give more flexibility and allow for more code re-use.
The text was updated successfully, but these errors were encountered: