My first plots

The goal of this exercise is to create PDF files with plots of pt and η of electrons from the My first histogrammer exercise. If you haven’t already, complete this exercise first, as we will need files created there as an input for this exercise.

Prepare config file

Start by copying an example plotter config:

cp tea/configs/examples/minimal_plotter_config.py configs/my_plotter_config.py

Open the file and apply modifications listed in the following subsections.

Samples, input, output

First, we need to specify samples to include in the samples dictionary - in our case it’s just one DY sample. Names of parameters in the Sample initializer should be quite self-explanatory. One should also provide the output_path where plots will be stored.

An example of this part of the config:

samples = (
  Sample(
    name="DY",
    file_path="../samples/histograms/background_dy.root",
    type=SampleType.background,
    cross_section=0.4,
    line_alpha=0.0,
    fill_color=ROOT.kRed-2,
    fill_alpha=0.7,
    marker_size=0.0,
    legend_description="DY"
  ),
)

output_path = "../plots"

Defining histograms

Next, we need to define histograms to plot. In our case, this will be Electron_pt and Electron_eta - if not sure which histograms can be plotted, you can check directly in the histograms file. There are a few options to normalize histograms - we will normalize to luminosity * cross-section.

In this case, it’s also important that we provide the luminosity we want to scale to:

histograms = (
#            name            title             logy    norm_type                  rebin xmin xmax ymin ymax  xlabel         ylabel
  Histogram("Electron_pt" , "Electron p_{T}",  True,   NormalizationType.to_lumi, 5,   0  , 150,  1,   1e3 , "p_{T} [GeV]", "# events (2018)"),
  Histogram("Electron_eta", "Electron #eta",   False,  NormalizationType.to_lumi, 5,  -2.4, 2.4,  0,   70  , "#eta"       , "# events (2018)"),
)

luminosity = 63670. # pb^-1 (2018)

Legends and other visuals

Finally, let’s specify:

  • Where to put the legend: we only have one background, but in principle, you could specify positions of signals and data legends here as well.
  • The canvas size.
  • We should turn off ratio plotting (since we only have one sample).

Here’s an example of this part of the config:

canvas_size = (800, 600)
show_ratio_plots = False

legends = {
  SampleType.background: Legend(0.7, 0.8, 0.85, 0.85, "f"),
}

Complete config example

Putting everything together, here’s a complete config:

import ROOT
from Sample import Sample, SampleType
from Legend import Legend
from Histogram import Histogram
from HistogramNormalizer import NormalizationType

samples = (
  Sample(
    name="DY",
    file_path="../samples/histograms/background_dy.root",
    type=SampleType.background,
    cross_section=0.4,
    line_alpha=0.0,
    fill_color=ROOT.kRed-2,
    fill_alpha=0.7,
    marker_size=0.0,
    legend_description="DY"
  ),
)
output_path = "../plots"

histograms = (
#            name            title             logx   logy    norm_type                  rebin xmin xmax ymin ymax  xlabel         ylabel
  Histogram("Electron_pt" , "Electron p_{T}",  False, True,   NormalizationType.to_lumi, 5,   0  , 150,  1,   1e3 , "p_{T} [GeV]", "# events (2018)"),
  Histogram("Electron_eta", "Electron #eta",   False, False,  NormalizationType.to_lumi, 5,  -2.4, 2.4,  0,   70  , "#eta"       , "# events (2018)"),
)
luminosity = 63670. # pb^-1 (2018)

legends = {
  SampleType.background: Legend(0.7, 0.8, 0.85, 0.85, "f"),
}

canvas_size = (800, 600)
show_ratio_plots = False

plotting_options = {
    SampleType.background: "hist",
    SampleType.signal: "nostack hist",
    SampleType.data: "nostack e",
}

Build & run

Now we are ready to build and run:

source tea/build.sh
cd bin
python plotter.py my_plotter_config.py

Have a look at plots directory - you should get nice PDF files with electron pt and η there!