search

arogozhnikov / hep_ml

Machine Learning for High Energy Physics.
https://arogozhnikov.github.io/hep_ml/
Other
176 stars 64 forks source link

Multidimensional reweighting #54

Closed Emix26 closed 5 years ago

Emix26 commented 5 years ago

Hi,

I'm using hep_ml to perform a multidimensional reweighting of a MC sample and it is working really well. I have a question, to which I've not been able to find an answer in the paper (https://arxiv.org/pdf/1608.05806.pdf) or in the documentation (https://arogozhnikov.github.io/hep_ml/reweight.html).

The multidimensional space is split in large bins by optimising a symmetrised chi2. But are those bins multidimensional or 1-dimensional ? In other words, is hep_ml reweighting 1D distributions iteratively ? Or is it reweighting multidimensional distributions directly ?

Cheers, Maxime

arogozhnikov commented 5 years ago

Hi,

But are those bins multidimensional or 1-dimensional ?

Bins are miltidimensional. Bin corresponds to leaf of a tree, if tree has depth > 1, bins are multidimensional.

Emix26 commented 5 years ago

Hi Alex,

Oh I see. I was still thinking in terms of 1D variables being cut in 1D bins at each step of the BDT, but all the nodes are then combined to form the region that is reweighted, making it multidimensional.

Thanks a lot for your quick answer. Cheers, Maxime