Cannot reweight decay processes

[Cen Zhang]

Hi,

I cannot reweight simple decay processes with MG2.6.0. For example z>ee:

generate z > e- e+
output

using a reweight_card.dat with only one line:

launch

crashes with the following error:

Command "reweight /Users/tywf/YG/MG5260/decayzee/Events/run_02/unweighted_events.lhe.gz_0.lhe -from_cards --multicore=create" interrupted with error:
TypeError : 'numpy.int32' object is not iterable
Please report this bug on https://bugs.launchpad.net/mg5amcnlo
More information is found in '/Users/tywf/YG/MG5260/decayzee/Events/run_02/unweighted_events.lhe.gz_0.lhe_tag_1_debug.log'.
Please attach this file to your report.

The log file is attached. t>bW has the same problem. Production processes seem to work fine.

Thank you,
Cen

[Olivier Mattelaer (olivier-mattelaer)]

Hi Cen,

Thanks for the report,

Note that using the reweighting package for 1>2 seems a huge overkill to me.
For each hypothesis, the matrix-element is independent of the kinematic. So evaluating a single PS point is enough for doing a re-weighting. (fact which is obviously not used by the code).

Otherwise here is the patch to make it work:

=== modified file 'madgraph/interface/reweight_interface.py'
--- madgraph/interface/reweight_interface.py 2018-01-18 12:34:20 +0000
+++ madgraph/interface/reweight_interface.py 2018-02-08 09:44:27 +0000
@@ -1741,11 +1741,12 @@

             for i,pdg in enumerate(all_pdgs):
                 if self.is_decay:
- incoming = pdg[0]
+ incoming = [pdg[0]]
                     outgoing = pdg[1:]
                 else:
                     incoming = pdg[0:2]
                     outgoing = pdg[2:]

[Cen Zhang (czhang24)]

Hi Olivier,

I wanted to reweight a 1->4 decay which has some kinematics, but the problem was independent of the process, so I ended up testing the most trivial case.

The patch does fix the problem. Thank you for looking into this!

Cen