Code review comment for lp:~maddevelopers/mg5amcnlo/2.5.5_NLOandLOmerging

Hi Olivier,

Yes, that's indeed the syntax.

All tests with kind of low statistics have been performed. However, we wanted to give the code to Josh Bendavid (who requested this feature) to perform some high-statistics tests and to confirm that's working within the CMS framework. He prefers that the code is part of the main branch, as he's loosing track of all separate contributions. That's why Marco wrote that some more testing is needed. Hence, we should keep this functionality somewhat hidden (or in beta) for the moment.

Indeed, it's not working with UNLOPS. However, with UNLOPS one can always generate separate samples and put them together at the end (at the cost of introducing large fractions of negative weights).

best,
Rikkert