BatchSampler

class BatchSampler : public madspace::FunctionGenerator

Concatenates the per-channel tensors of a multi-channel batch.

Given one tensor set per channel and the per-channel batch sizes, it packs them into a single contiguous batch (the inverse of the split done inside MultiChannelMapping). Used to assemble a mixed batch for the shared integrand evaluation (see Sec. 3.1 of [1]).

batch is the leading batch dimension. i indexes the channels.

Arguments

• channel<i>_in_<key> – the per-channel input tensors.

• batch_sizes – the number of entries in each channel.

Returns

• channel<i>_out_<key> – the concatenated tensors.

References

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

Public Functions

BatchSampler(const std::vector<NamedVector<Type>> &types)

Parameters:

types – One tensor-type set per channel.