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
Related to https://github.com/nsmith-/correctionlib/issues/8 : if many different working points, scalefactor types, JEC levels etc. end up being put in the same JSON, it would be very useful for the evaluator if the evaluator would only load the parts that are needed (e.g. for b-tagging only loose WP, or only light jet mistag for the loose WP).
The text was updated successfully, but these errors were encountered:
I had some plan to provide a pure python command-line tool to manipulate the json, allowing to slice various dimensions or remove a subset of corrections. For the evaluator, I am a bit reluctant to provide too much customization, but at least it seems straightfoward to have an alternative factory to https://github.com/nsmith-/correctionlib/blob/c3710d48bb0dca6334bf03c41b79130f09c925fb/include/correction.h#L152-L153 that takes a list of correction names to load, skipping others present in the file. More fine-grained control seems better left to the pure python tool, since it is easier to develop.
Related to https://github.com/nsmith-/correctionlib/issues/8 : if many different working points, scalefactor types, JEC levels etc. end up being put in the same JSON, it would be very useful for the evaluator if the evaluator would only load the parts that are needed (e.g. for b-tagging only loose WP, or only light jet mistag for the loose WP).
The text was updated successfully, but these errors were encountered: