## \file ## \ingroup tutorial_dataframe ## \notebook -nodraw ## This tutorial shows how to express the concept of ranges when working with the RDataFrame. ## ## \macro_code ## \macro_output ## ## \date March 2017 ## \author Danilo Piparo import ROOT def fill_tree(treeName, fileName): df = ROOT.RDataFrame(100) df.Define("b1", "(int) rdfentry_")\ .Define("b2", "(float) rdfentry_ * rdfentry_").Snapshot(treeName, fileName) # We prepare an input tree to run on fileName = "df006_ranges_py.root" treeName = "myTree" fill_tree(treeName, fileName) # We read the tree from the file and create a RDataFrame. d = ROOT.RDataFrame(treeName, fileName) # ## Usage of ranges # Now we'll count some entries using ranges c_all = d.Count() # This is how you can express a range of the first 30 entries d_0_30 = d.Range(0, 30) c_0_30 = d_0_30.Count() # This is how you pick all entries from 15 onwards d_15_end = d.Range(15, 0) c_15_end = d_15_end.Count() # We can use a stride too, in this case we pick an event every 3 d_15_end_3 = d.Range(15, 0, 3) c_15_end_3 = d_15_end_3.Count() # The Range is a 1st class citizen in the RDataFrame graph: # not only actions (like Count) but also filters and new columns can be added to it. d_0_50 = d.Range(0, 50) c_0_50_odd_b1 = d_0_50.Filter("1 == b1 % 2").Count() # An important thing to notice is that the counts of a filter are relative to the # number of entries a filter "sees". Therefore, if a Range depends on a filter, # the Range will act on the entries passing the filter only. c_0_3_after_even_b1 = d.Filter("0 == b1 % 2").Range(0, 3).Count() # Ok, time to wrap up: let's print all counts! print("Usage of ranges:") print(" - All entries:", c_all.GetValue()) print(" - Entries from 0 to 30:", c_0_30.GetValue()) print(" - Entries from 15 onwards:", c_15_end.GetValue()) print(" - Entries from 15 onwards in steps of 3:", c_15_end_3.GetValue()) print(" - Entries from 0 to 50, odd only:", c_0_50_odd_b1.GetValue()) print(" - First three entries of all even entries:", c_0_3_after_even_b1.GetValue())