Benifit of specifying the default branches when initializing RDatafrme?

ROOT
1

Hi All,
I have a question, whether declaring default branches help improve running speed?
Since I want to analyze nanoAOD but it has lots of branches.
Does it really reduce running time?

// code example
// case 1
RDataFrame d1(*myTree, {"var1"});
d1.Filter("passMyFunc(var1)").Snapshot("newTree1","output1.root");
// case 2
RDataFrame d2(*myTree);
d2.Filter("passMyFunc(var1)").Snapshot("newTree2","output2.root");
2

Hi,
default branches are a convenience feature and do not improve running speed w.r.t. passing the default branches as argument to each method call.

Cheers,
Enrico

P.S.
with that said, please report any usecase that results in unreasonably long runtimes, it might be a bug

3

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.