MultiChannelIntegrand

class madspace.MultiChannelIntegrand(self: madspace._madspace_py.MultiChannelIntegrand, integrands: collections.abc.Sequence[madspace._madspace_py.Integrand], return_sizes: bool = False)

Bases: FunctionGenerator

Evaluates several Integrand channels on one merged batch.

Partitions the batch by the per-channel counts, runs each Integrand on its slice and concatenates the results, the same pattern as MultiChannelMapping.

Arguments

• the union of the channel integrands’ arguments, plus the per-channel batch sizes.

Returns

• the concatenated per-event quantities, plus the per-channel sizes when return_sizes is set.

References

• [1] T. Heimel, O. Mattelaer, R. Winterhalder, “MadSpace”, https://arxiv.org/abs/2602.06895 (Sec. 2.1.1)

Parameters:

• integrands – The per-channel integrands.

• return_sizes – If true, also return the per-channel batch sizes.