# timeSeriesFromCSV¶

This macro illustrates the use of the time axis on a TGraph with data read from a text file containing the SWAN usage statistics during July 2017.

Author: Danilo Piparo, Olivier Couet
This notebook tutorial was automatically generated with ROOTBOOK-izer from the macro found in the ROOT repository on Tuesday, November 29, 2022 at 11:17 AM.

In [1]:
import ROOT

Welcome to JupyROOT 6.27/01


Open the data file. This csv contains the usage statistics of a CERN IT service, SWAN, during two weeks. We would like to plot this data with ROOT to draw some conclusions from it.

In [2]:
dirName = str(ROOT.gROOT.GetTutorialDir())
dirName += "/graphs/"
dirName= dirName.replace("/./", "/")
inputFileName = "%s/SWAN2017.dat" %dirName


Create the time graph

In [3]:
g = ROOT.TGraph()
g.SetTitle("SWAN Users during July 2017;Time;Number of Sessions")


Read the data and fill the graph with time along the X axis and number of users along the Y axis

In [4]:
lines = open(inputFileName, "r").readlines()

for i, line in enumerate(lines):
d, h, value = line.split()
g.SetPoint(i, ROOT.TDatime("%s %s" %(d,h)).Convert(), float(value))


Draw the graph

In [5]:
c = ROOT.TCanvas("c", "c", 950, 500)
c.SetLeftMargin(0.07)
c.SetRightMargin(0.04)
c.SetGrid()
g.SetLineWidth(3)
g.SetLineColor(ROOT.kBlue)
g.Draw("al")
g.GetYaxis().CenterTitle()


Make the X axis labelled with time

In [6]:
xaxis = g.GetXaxis()
xaxis.SetTimeDisplay(1)
xaxis.CenterTitle()
xaxis.SetTimeFormat("%a %d")
xaxis.SetTimeOffset(0)
xaxis.SetNdivisions(-219)
xaxis.SetLimits(ROOT.TDatime(2017, 7, 3, 0, 0, 0).Convert(), ROOT.TDatime(2017, 7, 22, 0, 0, 0).Convert())
xaxis.SetLabelSize(0.025)
xaxis.CenterLabels()


Draw all canvases

In [7]:
%jsroot on
from ROOT import gROOT
gROOT.GetListOfCanvases().Draw()