RooFit in hadron analyzer #955
Conversation
DelloStritto
commented
Sep 29, 2024
DelloStritto
commented
- Extend roofit fit to hadron analyzer
- implement Significance and S/B calculation in the fitter
|
|
||
| if level == "data": | ||
| signal_pdf = ws.pdf("sigrefl") | ||
| if not signal_pdf: | ||
| print('"sigrefl" PDF not found, using "sig" instead.') | ||
| signal_pdf = ws.pdf("sig") | ||
| if not signal_pdf: | ||
| print('"sig" PDF not found.') | ||
| background_pdf = ws.pdf("bkg") | ||
| model = RooAddPdf("model", | ||
| "Total model", | ||
| RooArgList(signal_pdf, background_pdf), | ||
| RooArgList(n_signal, n_background)) | ||
|
|
There was a problem hiding this comment.
This is based on assumptions about what names are used in the DB and I am not convinced that this is the best way of dealing with this. If we decide to rely on names from the DB, we should use sig and bkg there. But I would actually prefer to pass the names of the pdfs to be used for signal and background as parameters in the function call such that the roofitter stays free of assumptions on naming. In addition, we should avoid print in parts that behave like a library and, instead, throw an exception if there is an unrecoverable error.
There was a problem hiding this comment.
This is the main db for the Lc hadron analysis. Maybe We can delete the old format dbs and rename this one
| histonorm.SetBinContent(2, neventsafterevtsel) | ||
| histonorm.GetXaxis().SetBinLabel(2, "tot events after evt sel") |
There was a problem hiding this comment.
This changes the normalisation. Why is it needed?
