EDITED: the post below shows the problem more clear. it takes 0.16 s to read 4 branches (colunms) [image] and 2400 s when 16 branches (columns) are read [image] Would anyone guess of the origin of such behavior I should be able to provide the root file for testing in spirit of that …

Pin @oschulz . Any ideas? Thanks

Just found a typo in the second execution: it should be process.(tr) instead of process.(t). Then it takes 1s [image] What remains unclear why processing the tree directly is so much slower. It contains the only branches that I use [image]

Well, here is the shortest representation of the problem: [image] A few orders of magnitude difference. Interesting.

Conversion of the TTree (in disk) to Table (in memory) is done when getindex is called. It seems that the conversion is happening for every index when broadcast

Probably best if we discuss this on your GibHub issue ( https://github.com/JuliaHEP/UpROOT.jl/issues/12 ), in case we need to make changes to UpROOT.jl

Tables.jl, UpROOT.jl and the speed

General Usage

misha_mikhasenko August 11, 2020, 9:34am 4

Well, here is the shortest representation of the problem:

A few orders of magnitude difference. Interesting.

Topic		Replies	Views
Reading ROOT file with Branches High Energy Physics hep	3	802	March 23, 2024
Problem reading ROOT branch containing Strings with UnROOT.jl High Energy Physics package , hep	7	262	June 18, 2024
Storing and accessing large jagged array with julia General Usage question , data , filesystem , hep	33	4323	October 31, 2023
High Energy Physics (HEP) folks on Julia? Statistics question	100	9035	December 12, 2020
How to unroll tree traversal efficiently? General Usage metaprogramming , tree	1	1078	May 8, 2019

Tables.jl, UpROOT.jl and the speed

Related topics