Comparing histograms

mplhep provides dedicated comparison plotters in the mh.comp module for creating plots with different comparison panels. The following functions are available:

• mh.comp.hists(): compare two histograms, plot the main plot and a comparison panel.
• mh.comp.data_model(): compare a model made of histograms or functions to data, plot the main plot and a comparison panel.
• mh.comp.comparison(): to only plot the comparison panel given two histograms.
• mh.comp.get_comparison(): to get the [values, lower_uncertainties, upper_uncertainties] for a given comparison type.

Prerequisites

Throughout this guide the following codeblock is assumed.

import matplotlib.pyplot as plt
import numpy as np
import hist
np.random.seed(42)
import mplhep as mh
# mh.style.use('<as appropriate>')

Comparing two histograms

Available methods

To compare two histograms, use mh.comp.hists(). This function takes two histograms as input and creates a figure with a main plot showing both histograms and a comparison panel below it.

The comparison parameter accepts:

• ratio: \(\frac{\text{h1}}{\text{h2}}\)
• split_ratio: same as ratio, but the uncertainties of h1 and h2 are shown separately in the comparison panel (used extensively in data/model comparisons, see below)
• pull: \(\frac{\text{h1} - \text{h2}}{\sqrt{\sigma_{\text{h1}}^2 + \sigma_{\text{h2}}^2}}\)
• difference: \(\text{h1} - \text{h2}\)
• relative_difference: \(\frac{\text{h1} - \text{h2}}{\text{h2}}\)
• asymmetry: \(\frac{\text{h1} - \text{h2}}{\text{h1} + \text{h2}}\)
• efficiency: \(\frac{\text{h1}}{\text{h2}}\) (with uncertainties from eq.19 here)

DefaultplothistCMSATLASLHCbALICEDUNE

RatioSplit ratioPullDifferenceRelative differenceAsymmetryEfficiency

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='ratio',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='split_ratio',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='pull',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='difference',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='relative_difference',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='asymmetry',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)

# Generate sample data and model
x2 = np.random.normal(0, 1, 1000)
x1 = np.random.choice(x2, size=500, replace=False)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1) # h1 needs to be a subset of h2
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='efficiency',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)

RatioSplit ratioPullDifferenceRelative differenceAsymmetryEfficiency

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='ratio',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
txt_obj = mh.add_text('plothist', loc='over left', ax=ax_main)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='split_ratio',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
txt_obj = mh.add_text('plothist', loc='over left', ax=ax_main)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='pull',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
txt_obj = mh.add_text('plothist', loc='over left', ax=ax_main)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='difference',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
txt_obj = mh.add_text('plothist', loc='over left', ax=ax_main)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='relative_difference',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
txt_obj = mh.add_text('plothist', loc='over left', ax=ax_main)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='asymmetry',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
txt_obj = mh.add_text('plothist', loc='over left', ax=ax_main)

# Generate sample data and model
x2 = np.random.normal(0, 1, 1000)
x1 = np.random.choice(x2, size=500, replace=False)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1) # h1 needs to be a subset of h2
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='efficiency',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
txt_obj = mh.add_text('plothist', loc='over left', ax=ax_main)

RatioSplit ratioPullDifferenceRelative differenceAsymmetryEfficiency

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='ratio',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
mh.cms.label(data=True, ax=ax_main)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='split_ratio',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
mh.cms.label(data=True, ax=ax_main)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='pull',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
mh.cms.label(data=True, ax=ax_main)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='difference',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
mh.cms.label(data=True, ax=ax_main)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='relative_difference',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
mh.cms.label(data=True, ax=ax_main)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='asymmetry',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
mh.cms.label(data=True, ax=ax_main)

# Generate sample data and model
x2 = np.random.normal(0, 1, 1000)
x1 = np.random.choice(x2, size=500, replace=False)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1) # h1 needs to be a subset of h2
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='efficiency',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
mh.cms.label(data=True, ax=ax_main)

RatioSplit ratioPullDifferenceRelative differenceAsymmetryEfficiency

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='ratio',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
mh.atlas.label(data=True, ax=ax_main)
mh.mpl_magic(soft_fail=True)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='split_ratio',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
mh.atlas.label(data=True, ax=ax_main)
mh.mpl_magic(soft_fail=True)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='pull',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
mh.atlas.label(data=True, ax=ax_main)
mh.mpl_magic(soft_fail=True)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='difference',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
mh.atlas.label(data=True, ax=ax_main)
mh.mpl_magic(soft_fail=True)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='relative_difference',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
mh.atlas.label(data=True, ax=ax_main)
mh.mpl_magic(soft_fail=True)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='asymmetry',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
mh.atlas.label(data=True, ax=ax_main)
mh.mpl_magic(soft_fail=True)

# Generate sample data and model
x2 = np.random.normal(0, 1, 1000)
x1 = np.random.choice(x2, size=500, replace=False)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1) # h1 needs to be a subset of h2
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='efficiency',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
mh.atlas.label(data=True, ax=ax_main)
mh.mpl_magic(soft_fail=True)

RatioSplit ratioPullDifferenceRelative differenceAsymmetryEfficiency

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='ratio',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
mh.lhcb.label(data=True, ax=ax_main)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='split_ratio',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
mh.lhcb.label(data=True, ax=ax_main)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='pull',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
mh.lhcb.label(data=True, ax=ax_main)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='difference',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
mh.lhcb.label(data=True, ax=ax_main)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='relative_difference',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
mh.lhcb.label(data=True, ax=ax_main)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='asymmetry',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
mh.lhcb.label(data=True, ax=ax_main)

# Generate sample data and model
x2 = np.random.normal(0, 1, 1000)
x1 = np.random.choice(x2, size=500, replace=False)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1) # h1 needs to be a subset of h2
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='efficiency',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
mh.lhcb.label(data=True, ax=ax_main)

RatioSplit ratioPullDifferenceRelative differenceAsymmetryEfficiency

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='ratio',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
mh.alice.label(data=True, ax=ax_main)
mh.mpl_magic(soft_fail=True)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='split_ratio',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
mh.alice.label(data=True, ax=ax_main)
mh.mpl_magic(soft_fail=True)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='pull',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
mh.alice.label(data=True, ax=ax_main)
mh.mpl_magic(soft_fail=True)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='difference',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
mh.alice.label(data=True, ax=ax_main)
mh.mpl_magic(soft_fail=True)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='relative_difference',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
mh.alice.label(data=True, ax=ax_main)
mh.mpl_magic(soft_fail=True)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='asymmetry',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
mh.alice.label(data=True, ax=ax_main)
mh.mpl_magic(soft_fail=True)

# Generate sample data and model
x2 = np.random.normal(0, 1, 1000)
x1 = np.random.choice(x2, size=500, replace=False)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1) # h1 needs to be a subset of h2
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='efficiency',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
mh.alice.label(data=True, ax=ax_main)
mh.mpl_magic(soft_fail=True)

RatioSplit ratioPullDifferenceRelative differenceAsymmetryEfficiency

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='ratio',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
mh.dune.label(data=True, ax=ax_main)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='split_ratio',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
mh.dune.label(data=True, ax=ax_main)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='pull',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
mh.dune.label(data=True, ax=ax_main)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='difference',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
mh.dune.label(data=True, ax=ax_main)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='relative_difference',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
mh.dune.label(data=True, ax=ax_main)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='asymmetry',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
mh.dune.label(data=True, ax=ax_main)

# Generate sample data and model
x2 = np.random.normal(0, 1, 1000)
x1 = np.random.choice(x2, size=500, replace=False)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1) # h1 needs to be a subset of h2
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.hists(
 h1,
 h2,
 comparison='efficiency',
 xlabel='Observable [GeV]',
 ylabel='Events',
 h1_label='h1',
 h2_label='h2'
)
mh.dune.label(data=True, ax=ax_main)

Only plot the comparison panel

To only plot the comparison panel given two histograms, use mh.comp.comparison(). This function takes two histograms as input and creates a figure with only the comparison panel.

DefaultplothistCMSATLASLHCbALICEDUNE

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison panel only
fig, ax = plt.subplots()

mh.comp.comparison(
 h1,
 h2,
 comparison='ratio',
 xlabel='Observable [GeV]',
 h1_label='h1',
 h2_label='h2',
 ax=ax
)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison panel only
fig, ax = plt.subplots()

mh.comp.comparison(
 h1,
 h2,
 comparison='ratio',
 xlabel='Observable [GeV]',
 h1_label='h1',
 h2_label='h2',
 ax=ax
)
txt_obj = mh.add_text('plothist', ax=ax)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison panel only
fig, ax = plt.subplots()

mh.comp.comparison(
 h1,
 h2,
 comparison='ratio',
 xlabel='Observable [GeV]',
 h1_label='h1',
 h2_label='h2',
 ax=ax
)
mh.cms.label(data=True, lumi=100, ax=ax)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison panel only
fig, ax = plt.subplots()

mh.comp.comparison(
 h1,
 h2,
 comparison='ratio',
 xlabel='Observable [GeV]',
 h1_label='h1',
 h2_label='h2',
 ax=ax
)
mh.atlas.label(data=True, lumi=150, ax=ax)
mh.mpl_magic(soft_fail=True)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison panel only
fig, ax = plt.subplots()

mh.comp.comparison(
 h1,
 h2,
 comparison='ratio',
 xlabel='Observable [GeV]',
 h1_label='h1',
 h2_label='h2',
 ax=ax
)
mh.lhcb.label(data=True, lumi=50, ax=ax)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison panel only
fig, ax = plt.subplots()

mh.comp.comparison(
 h1,
 h2,
 comparison='ratio',
 xlabel='Observable [GeV]',
 h1_label='h1',
 h2_label='h2',
 ax=ax
)
mh.alice.label(data=True, lumi=100, ax=ax)
mh.mpl_magic(soft_fail=True)

# Generate sample data and model
x1 = np.random.normal(0, 1, 1000)
x2 = np.random.normal(0, 1.05, 1000)
h1 = hist.new.Reg(25, -4, 4).Weight().fill(x1)
h2 = hist.new.Reg(25, -4, 4).Weight().fill(x2)

# Create comparison panel only
fig, ax = plt.subplots()

mh.comp.comparison(
 h1,
 h2,
 comparison='ratio',
 xlabel='Observable [GeV]',
 h1_label='h1',
 h2_label='h2',
 ax=ax
)
mh.dune.label(data=True, lumi=100, ax=ax)

To get the [values, lower_uncertainties, upper_uncertainties] for a given comparison type without plotting, use mh.comp.get_comparison():

values, lower_uncertainties, upper_uncertainties = mh.comp.get_comparison(
    h1,
    h2,
    comparison='ratio'
)

Data-MC comparison

To compare data to a model made of multiple components (e.g. signal, backgrounds...), use mh.comp.data_model(). The function is very flexible, it can accept any number of stacked and/or unstacked components, either as histograms or functions. It will then compare the sum of the components to the data, with the comparison of your choice. The default comparison is the split_ratio between the model and the data. It can take any comparison method available in mh.comp.hists().

DefaultplothistCMSATLASLHCbALICEDUNE

StackedMixed stacked and unstackedFunctions, stackedFunctions, mixed stacked and unstacked

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1, h_signal],
 stacked_labels=['Bkg2', 'Bkg1', 'Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events'
)

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# The function will automatically sum the stacked and unstacked components
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1],
 stacked_labels=['Bkg2', 'Bkg1'],
 stacked_colors=['grey', 'lightblue'],
 unstacked_components=[h_signal],
 unstacked_labels=['Signal'],
 unstacked_colors=['red'],
 xlabel='Observable [GeV]',
 ylabel='Events'
)

# Define component functions
def f_signal(x):
 return 200 * scipy.stats.norm.pdf(x, loc=0.5, scale=3)
def f_bkg1(x):
 return 500 * scipy.stats.norm.pdf(x, loc=-1.5, scale=4)
def f_bkg2(x):
 return 500 * scipy.stats.norm.pdf(x, loc=-1., scale=1.8)

# Generate data histogram
x_data = np.concatenate([np.random.normal(-1.5, 4, 1500), np.random.normal(-1., 1.8, 2000)])
data_hist = hist.new.Regular(50, -8, 8).Weight().fill(x_data)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=data_hist,
 stacked_components=[f_bkg2, f_bkg1, f_signal],
 stacked_labels=['Bkg 2', 'Bkg 1', 'Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events'
)

# Define component functions
def f_signal(x):
 return 200 * norm.pdf(x, loc=0.5, scale=3)
def f_bkg1(x):
 return 500 * norm.pdf(x, loc=-1.5, scale=4)
def f_bkg2(x):
 return 500 * norm.pdf(x, loc=-1., scale=1.8)

# Generate data histogram
x_data = np.concatenate([np.random.normal(-1.5, 4, 1500), np.random.normal(-1., 1.8, 2000)])
data_hist = hist.new.Regular(50, -8, 8).Weight().fill(x_data)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=data_hist,
 stacked_components=[f_bkg2, f_bkg1,],
 stacked_labels=['Bkg 2', 'Bkg 1'],
 unstacked_components=[f_signal],
 unstacked_labels=['Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events'
)

StackedMixed stacked and unstackedFunctions, stackedFunctions, mixed stacked and unstacked

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1, h_signal],
 stacked_labels=['Bkg2', 'Bkg1', 'Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events'
)
txt_obj = mh.add_text('plothist', loc='over left', ax=ax_main)

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# The function will automatically sum the stacked and unstacked components
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1],
 stacked_labels=['Bkg2', 'Bkg1'],
 stacked_colors=['grey', 'lightblue'],
 unstacked_components=[h_signal],
 unstacked_labels=['Signal'],
 unstacked_colors=['red'],
 xlabel='Observable [GeV]',
 ylabel='Events'
)
txt_obj = mh.add_text('plothist', loc='over left', ax=ax_main)

# Define component functions
def f_signal(x):
 return 200 * scipy.stats.norm.pdf(x, loc=0.5, scale=3)
def f_bkg1(x):
 return 500 * scipy.stats.norm.pdf(x, loc=-1.5, scale=4)
def f_bkg2(x):
 return 500 * scipy.stats.norm.pdf(x, loc=-1., scale=1.8)

# Generate data histogram
x_data = np.concatenate([np.random.normal(-1.5, 4, 1500), np.random.normal(-1., 1.8, 2000)])
data_hist = hist.new.Regular(50, -8, 8).Weight().fill(x_data)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=data_hist,
 stacked_components=[f_bkg2, f_bkg1, f_signal],
 stacked_labels=['Bkg 2', 'Bkg 1', 'Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events'
)
txt_obj = mh.add_text('plothist', loc='over left', ax=ax_main)

# Define component functions
def f_signal(x):
 return 200 * norm.pdf(x, loc=0.5, scale=3)
def f_bkg1(x):
 return 500 * norm.pdf(x, loc=-1.5, scale=4)
def f_bkg2(x):
 return 500 * norm.pdf(x, loc=-1., scale=1.8)

# Generate data histogram
x_data = np.concatenate([np.random.normal(-1.5, 4, 1500), np.random.normal(-1., 1.8, 2000)])
data_hist = hist.new.Regular(50, -8, 8).Weight().fill(x_data)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=data_hist,
 stacked_components=[f_bkg2, f_bkg1,],
 stacked_labels=['Bkg 2', 'Bkg 1'],
 unstacked_components=[f_signal],
 unstacked_labels=['Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events'
)
txt_obj = mh.add_text('plothist', loc='over left', ax=ax_main)

StackedMixed stacked and unstackedFunctions, stackedFunctions, mixed stacked and unstacked

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1, h_signal],
 stacked_labels=['Bkg2', 'Bkg1', 'Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events'
)
mh.cms.label(data=True, lumi=100, ax=ax_main)

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# The function will automatically sum the stacked and unstacked components
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1],
 stacked_labels=['Bkg2', 'Bkg1'],
 stacked_colors=['grey', 'lightblue'],
 unstacked_components=[h_signal],
 unstacked_labels=['Signal'],
 unstacked_colors=['red'],
 xlabel='Observable [GeV]',
 ylabel='Events'
)
mh.cms.label(data=True, lumi=100, ax=ax_main)

# Define component functions
def f_signal(x):
 return 200 * scipy.stats.norm.pdf(x, loc=0.5, scale=3)
def f_bkg1(x):
 return 500 * scipy.stats.norm.pdf(x, loc=-1.5, scale=4)
def f_bkg2(x):
 return 500 * scipy.stats.norm.pdf(x, loc=-1., scale=1.8)

# Generate data histogram
x_data = np.concatenate([np.random.normal(-1.5, 4, 1500), np.random.normal(-1., 1.8, 2000)])
data_hist = hist.new.Regular(50, -8, 8).Weight().fill(x_data)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=data_hist,
 stacked_components=[f_bkg2, f_bkg1, f_signal],
 stacked_labels=['Bkg 2', 'Bkg 1', 'Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events'
)
mh.cms.label(data=True, lumi=100, ax=ax_main)

# Define component functions
def f_signal(x):
 return 200 * norm.pdf(x, loc=0.5, scale=3)
def f_bkg1(x):
 return 500 * norm.pdf(x, loc=-1.5, scale=4)
def f_bkg2(x):
 return 500 * norm.pdf(x, loc=-1., scale=1.8)

# Generate data histogram
x_data = np.concatenate([np.random.normal(-1.5, 4, 1500), np.random.normal(-1., 1.8, 2000)])
data_hist = hist.new.Regular(50, -8, 8).Weight().fill(x_data)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=data_hist,
 stacked_components=[f_bkg2, f_bkg1,],
 stacked_labels=['Bkg 2', 'Bkg 1'],
 unstacked_components=[f_signal],
 unstacked_labels=['Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events'
)
mh.cms.label(data=True, lumi=100, ax=ax_main)

StackedMixed stacked and unstackedFunctions, stackedFunctions, mixed stacked and unstacked

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1, h_signal],
 stacked_labels=['Bkg2', 'Bkg1', 'Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events'
)
mh.atlas.label(data=True, lumi=150, ax=ax_main)
mh.mpl_magic(soft_fail=True)

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# The function will automatically sum the stacked and unstacked components
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1],
 stacked_labels=['Bkg2', 'Bkg1'],
 stacked_colors=['grey', 'lightblue'],
 unstacked_components=[h_signal],
 unstacked_labels=['Signal'],
 unstacked_colors=['red'],
 xlabel='Observable [GeV]',
 ylabel='Events'
)
mh.atlas.label(data=True, lumi=150, ax=ax_main)
mh.mpl_magic(soft_fail=True)

# Define component functions
def f_signal(x):
 return 200 * scipy.stats.norm.pdf(x, loc=0.5, scale=3)
def f_bkg1(x):
 return 500 * scipy.stats.norm.pdf(x, loc=-1.5, scale=4)
def f_bkg2(x):
 return 500 * scipy.stats.norm.pdf(x, loc=-1., scale=1.8)

# Generate data histogram
x_data = np.concatenate([np.random.normal(-1.5, 4, 1500), np.random.normal(-1., 1.8, 2000)])
data_hist = hist.new.Regular(50, -8, 8).Weight().fill(x_data)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=data_hist,
 stacked_components=[f_bkg2, f_bkg1, f_signal],
 stacked_labels=['Bkg 2', 'Bkg 1', 'Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events'
)
mh.atlas.label(data=True, lumi=150, ax=ax_main)
mh.mpl_magic(soft_fail=True)

# Define component functions
def f_signal(x):
 return 200 * norm.pdf(x, loc=0.5, scale=3)
def f_bkg1(x):
 return 500 * norm.pdf(x, loc=-1.5, scale=4)
def f_bkg2(x):
 return 500 * norm.pdf(x, loc=-1., scale=1.8)

# Generate data histogram
x_data = np.concatenate([np.random.normal(-1.5, 4, 1500), np.random.normal(-1., 1.8, 2000)])
data_hist = hist.new.Regular(50, -8, 8).Weight().fill(x_data)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=data_hist,
 stacked_components=[f_bkg2, f_bkg1,],
 stacked_labels=['Bkg 2', 'Bkg 1'],
 unstacked_components=[f_signal],
 unstacked_labels=['Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events'
)
mh.atlas.label(data=True, lumi=150, ax=ax_main)
mh.mpl_magic(soft_fail=True)

StackedMixed stacked and unstackedFunctions, stackedFunctions, mixed stacked and unstacked

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1, h_signal],
 stacked_labels=['Bkg2', 'Bkg1', 'Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events'
)
mh.lhcb.label(data=True, lumi=50, ax=ax_main)

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# The function will automatically sum the stacked and unstacked components
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1],
 stacked_labels=['Bkg2', 'Bkg1'],
 stacked_colors=['grey', 'lightblue'],
 unstacked_components=[h_signal],
 unstacked_labels=['Signal'],
 unstacked_colors=['red'],
 xlabel='Observable [GeV]',
 ylabel='Events'
)
mh.lhcb.label(data=True, lumi=50, ax=ax_main)

# Define component functions
def f_signal(x):
 return 200 * scipy.stats.norm.pdf(x, loc=0.5, scale=3)
def f_bkg1(x):
 return 500 * scipy.stats.norm.pdf(x, loc=-1.5, scale=4)
def f_bkg2(x):
 return 500 * scipy.stats.norm.pdf(x, loc=-1., scale=1.8)

# Generate data histogram
x_data = np.concatenate([np.random.normal(-1.5, 4, 1500), np.random.normal(-1., 1.8, 2000)])
data_hist = hist.new.Regular(50, -8, 8).Weight().fill(x_data)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=data_hist,
 stacked_components=[f_bkg2, f_bkg1, f_signal],
 stacked_labels=['Bkg 2', 'Bkg 1', 'Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events'
)
mh.lhcb.label(data=True, lumi=50, ax=ax_main)

# Define component functions
def f_signal(x):
 return 200 * norm.pdf(x, loc=0.5, scale=3)
def f_bkg1(x):
 return 500 * norm.pdf(x, loc=-1.5, scale=4)
def f_bkg2(x):
 return 500 * norm.pdf(x, loc=-1., scale=1.8)

# Generate data histogram
x_data = np.concatenate([np.random.normal(-1.5, 4, 1500), np.random.normal(-1., 1.8, 2000)])
data_hist = hist.new.Regular(50, -8, 8).Weight().fill(x_data)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=data_hist,
 stacked_components=[f_bkg2, f_bkg1,],
 stacked_labels=['Bkg 2', 'Bkg 1'],
 unstacked_components=[f_signal],
 unstacked_labels=['Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events'
)
mh.lhcb.label(data=True, lumi=50, ax=ax_main)

StackedMixed stacked and unstackedFunctions, stackedFunctions, mixed stacked and unstacked

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1, h_signal],
 stacked_labels=['Bkg2', 'Bkg1', 'Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events'
)
mh.alice.label(data=True, lumi=100, ax=ax_main)
mh.mpl_magic(soft_fail=True)

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# The function will automatically sum the stacked and unstacked components
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1],
 stacked_labels=['Bkg2', 'Bkg1'],
 stacked_colors=['grey', 'lightblue'],
 unstacked_components=[h_signal],
 unstacked_labels=['Signal'],
 unstacked_colors=['red'],
 xlabel='Observable [GeV]',
 ylabel='Events'
)
mh.alice.label(data=True, lumi=100, ax=ax_main)
mh.mpl_magic(soft_fail=True)

# Define component functions
def f_signal(x):
 return 200 * scipy.stats.norm.pdf(x, loc=0.5, scale=3)
def f_bkg1(x):
 return 500 * scipy.stats.norm.pdf(x, loc=-1.5, scale=4)
def f_bkg2(x):
 return 500 * scipy.stats.norm.pdf(x, loc=-1., scale=1.8)

# Generate data histogram
x_data = np.concatenate([np.random.normal(-1.5, 4, 1500), np.random.normal(-1., 1.8, 2000)])
data_hist = hist.new.Regular(50, -8, 8).Weight().fill(x_data)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=data_hist,
 stacked_components=[f_bkg2, f_bkg1, f_signal],
 stacked_labels=['Bkg 2', 'Bkg 1', 'Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events'
)
mh.alice.label(data=True, lumi=100, ax=ax_main)
mh.mpl_magic(soft_fail=True)

# Define component functions
def f_signal(x):
 return 200 * norm.pdf(x, loc=0.5, scale=3)
def f_bkg1(x):
 return 500 * norm.pdf(x, loc=-1.5, scale=4)
def f_bkg2(x):
 return 500 * norm.pdf(x, loc=-1., scale=1.8)

# Generate data histogram
x_data = np.concatenate([np.random.normal(-1.5, 4, 1500), np.random.normal(-1., 1.8, 2000)])
data_hist = hist.new.Regular(50, -8, 8).Weight().fill(x_data)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=data_hist,
 stacked_components=[f_bkg2, f_bkg1,],
 stacked_labels=['Bkg 2', 'Bkg 1'],
 unstacked_components=[f_signal],
 unstacked_labels=['Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events'
)
mh.alice.label(data=True, lumi=100, ax=ax_main)
mh.mpl_magic(soft_fail=True)

StackedMixed stacked and unstackedFunctions, stackedFunctions, mixed stacked and unstacked

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1, h_signal],
 stacked_labels=['Bkg2', 'Bkg1', 'Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events'
)
mh.dune.label(data=True, lumi=100, ax=ax_main)

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# The function will automatically sum the stacked and unstacked components
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1],
 stacked_labels=['Bkg2', 'Bkg1'],
 stacked_colors=['grey', 'lightblue'],
 unstacked_components=[h_signal],
 unstacked_labels=['Signal'],
 unstacked_colors=['red'],
 xlabel='Observable [GeV]',
 ylabel='Events'
)
mh.dune.label(data=True, lumi=100, ax=ax_main)

# Define component functions
def f_signal(x):
 return 200 * scipy.stats.norm.pdf(x, loc=0.5, scale=3)
def f_bkg1(x):
 return 500 * scipy.stats.norm.pdf(x, loc=-1.5, scale=4)
def f_bkg2(x):
 return 500 * scipy.stats.norm.pdf(x, loc=-1., scale=1.8)

# Generate data histogram
x_data = np.concatenate([np.random.normal(-1.5, 4, 1500), np.random.normal(-1., 1.8, 2000)])
data_hist = hist.new.Regular(50, -8, 8).Weight().fill(x_data)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=data_hist,
 stacked_components=[f_bkg2, f_bkg1, f_signal],
 stacked_labels=['Bkg 2', 'Bkg 1', 'Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events'
)
mh.dune.label(data=True, lumi=100, ax=ax_main)

# Define component functions
def f_signal(x):
 return 200 * norm.pdf(x, loc=0.5, scale=3)
def f_bkg1(x):
 return 500 * norm.pdf(x, loc=-1.5, scale=4)
def f_bkg2(x):
 return 500 * norm.pdf(x, loc=-1., scale=1.8)

# Generate data histogram
x_data = np.concatenate([np.random.normal(-1.5, 4, 1500), np.random.normal(-1., 1.8, 2000)])
data_hist = hist.new.Regular(50, -8, 8).Weight().fill(x_data)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=data_hist,
 stacked_components=[f_bkg2, f_bkg1,],
 stacked_labels=['Bkg 2', 'Bkg 1'],
 unstacked_components=[f_signal],
 unstacked_labels=['Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events'
)
mh.dune.label(data=True, lumi=100, ax=ax_main)

Tip

To only plot the model, without data or comparison panel, use mh.model(). It takes the same arguments as mh.comp.data_model(), except for the data_hist and comparison parameters.

Showcasing more options

mh.comp.data_model() is very flexible and can be customized further. For more examples and details, see the API Reference and the Gallery. Here is a selection of additional examples showcasing some of the available options.

DefaultplothistCMSATLASLHCbALICEDUNE

Different comparisonRemove MC uncertaintiesModel kwargsPlot only 1 ax

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1, h_signal],
 stacked_labels=['Bkg2', 'Bkg1', 'Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events',
 comparison='pull' # Accept any comparison from mh.comp.comparison()
)

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1, h_signal],
 stacked_labels=['Bkg2', 'Bkg1', 'Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events',
 model_uncertainty=False, # Set the model uncertainty to zero
 comparison='pull', # The pull formula is updated automatically
)

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1],
 stacked_labels=['Bkg2', 'Bkg1'],
 stacked_colors=['grey', 'lightblue'],
 unstacked_components=[h_signal],
 unstacked_labels=['Signal'],
 unstacked_colors=['red'],
 model_sum_kwargs={"show": True, "label": "Total Model", "color": "violet"}, # Just like stacked_components_kwargs and unstacked_components_kwargs, you can pass kwargs for the model sum line
 xlabel='Observable [GeV]',
 ylabel='Events'
)

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# Create comparison plot
fig, ax_main, ax_comparison = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1, h_signal],
 stacked_labels=['Bkg2', 'Bkg1', 'Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events',
 plot_only='ax_main' # Only plot the comparison axis (works with 'ax_comparison' as well).
)

Different comparisonRemove MC uncertaintiesModel kwargsPlot only 1 ax

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1, h_signal],
 stacked_labels=['Bkg2', 'Bkg1', 'Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events',
 comparison='pull' # Accept any comparison from mh.comp.comparison()
)
txt_obj = mh.add_text('plothist', loc='over left', ax=ax_main)

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1, h_signal],
 stacked_labels=['Bkg2', 'Bkg1', 'Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events',
 model_uncertainty=False, # Set the model uncertainty to zero
 comparison='pull', # The pull formula is updated automatically
)
txt_obj = mh.add_text('plothist', loc='over left', ax=ax_main)

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1],
 stacked_labels=['Bkg2', 'Bkg1'],
 stacked_colors=['grey', 'lightblue'],
 unstacked_components=[h_signal],
 unstacked_labels=['Signal'],
 unstacked_colors=['red'],
 model_sum_kwargs={"show": True, "label": "Total Model", "color": "violet"}, # Just like stacked_components_kwargs and unstacked_components_kwargs, you can pass kwargs for the model sum line
 xlabel='Observable [GeV]',
 ylabel='Events'
)
txt_obj = mh.add_text('plothist', loc='over left', ax=ax_main)

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# Create comparison plot
fig, ax_main, ax_comparison = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1, h_signal],
 stacked_labels=['Bkg2', 'Bkg1', 'Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events',
 plot_only='ax_main' # Only plot the comparison axis (works with 'ax_comparison' as well).
)
txt_obj = mh.add_text('plothist', loc='over left', ax=ax_main)

Different comparisonRemove MC uncertaintiesModel kwargsPlot only 1 ax

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1, h_signal],
 stacked_labels=['Bkg2', 'Bkg1', 'Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events',
 comparison='pull' # Accept any comparison from mh.comp.comparison()
)
mh.cms.label(data=True, lumi=100, ax=ax_main)

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1, h_signal],
 stacked_labels=['Bkg2', 'Bkg1', 'Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events',
 model_uncertainty=False, # Set the model uncertainty to zero
 comparison='pull', # The pull formula is updated automatically
)
mh.cms.label(data=True, lumi=100, ax=ax_main)

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1],
 stacked_labels=['Bkg2', 'Bkg1'],
 stacked_colors=['grey', 'lightblue'],
 unstacked_components=[h_signal],
 unstacked_labels=['Signal'],
 unstacked_colors=['red'],
 model_sum_kwargs={"show": True, "label": "Total Model", "color": "violet"}, # Just like stacked_components_kwargs and unstacked_components_kwargs, you can pass kwargs for the model sum line
 xlabel='Observable [GeV]',
 ylabel='Events'
)
mh.cms.label(data=True, lumi=100, ax=ax_main)

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# Create comparison plot
fig, ax_main, ax_comparison = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1, h_signal],
 stacked_labels=['Bkg2', 'Bkg1', 'Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events',
 plot_only='ax_main' # Only plot the comparison axis (works with 'ax_comparison' as well).
)
mh.cms.label(data=True, lumi=100, ax=ax_main)

Different comparisonRemove MC uncertaintiesModel kwargsPlot only 1 ax

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1, h_signal],
 stacked_labels=['Bkg2', 'Bkg1', 'Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events',
 comparison='pull' # Accept any comparison from mh.comp.comparison()
)
mh.atlas.label(data=True, lumi=150, ax=ax_main)
mh.mpl_magic(soft_fail=True)

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1, h_signal],
 stacked_labels=['Bkg2', 'Bkg1', 'Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events',
 model_uncertainty=False, # Set the model uncertainty to zero
 comparison='pull', # The pull formula is updated automatically
)
mh.atlas.label(data=True, lumi=150, ax=ax_main)
mh.mpl_magic(soft_fail=True)

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1],
 stacked_labels=['Bkg2', 'Bkg1'],
 stacked_colors=['grey', 'lightblue'],
 unstacked_components=[h_signal],
 unstacked_labels=['Signal'],
 unstacked_colors=['red'],
 model_sum_kwargs={"show": True, "label": "Total Model", "color": "violet"}, # Just like stacked_components_kwargs and unstacked_components_kwargs, you can pass kwargs for the model sum line
 xlabel='Observable [GeV]',
 ylabel='Events'
)
mh.atlas.label(data=True, lumi=150, ax=ax_main)
mh.mpl_magic(soft_fail=True)

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# Create comparison plot
fig, ax_main, ax_comparison = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1, h_signal],
 stacked_labels=['Bkg2', 'Bkg1', 'Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events',
 plot_only='ax_main' # Only plot the comparison axis (works with 'ax_comparison' as well).
)
mh.atlas.label(data=True, lumi=150, ax=ax_main)
mh.mpl_magic(soft_fail=True)

Different comparisonRemove MC uncertaintiesModel kwargsPlot only 1 ax

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1, h_signal],
 stacked_labels=['Bkg2', 'Bkg1', 'Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events',
 comparison='pull' # Accept any comparison from mh.comp.comparison()
)
mh.lhcb.label(data=True, lumi=50, ax=ax_main)

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1, h_signal],
 stacked_labels=['Bkg2', 'Bkg1', 'Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events',
 model_uncertainty=False, # Set the model uncertainty to zero
 comparison='pull', # The pull formula is updated automatically
)
mh.lhcb.label(data=True, lumi=50, ax=ax_main)

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1],
 stacked_labels=['Bkg2', 'Bkg1'],
 stacked_colors=['grey', 'lightblue'],
 unstacked_components=[h_signal],
 unstacked_labels=['Signal'],
 unstacked_colors=['red'],
 model_sum_kwargs={"show": True, "label": "Total Model", "color": "violet"}, # Just like stacked_components_kwargs and unstacked_components_kwargs, you can pass kwargs for the model sum line
 xlabel='Observable [GeV]',
 ylabel='Events'
)
mh.lhcb.label(data=True, lumi=50, ax=ax_main)

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# Create comparison plot
fig, ax_main, ax_comparison = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1, h_signal],
 stacked_labels=['Bkg2', 'Bkg1', 'Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events',
 plot_only='ax_main' # Only plot the comparison axis (works with 'ax_comparison' as well).
)
mh.lhcb.label(data=True, lumi=50, ax=ax_main)

Different comparisonRemove MC uncertaintiesModel kwargsPlot only 1 ax

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1, h_signal],
 stacked_labels=['Bkg2', 'Bkg1', 'Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events',
 comparison='pull' # Accept any comparison from mh.comp.comparison()
)
mh.alice.label(data=True, lumi=100, ax=ax_main)
mh.mpl_magic(soft_fail=True)

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1, h_signal],
 stacked_labels=['Bkg2', 'Bkg1', 'Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events',
 model_uncertainty=False, # Set the model uncertainty to zero
 comparison='pull', # The pull formula is updated automatically
)
mh.alice.label(data=True, lumi=100, ax=ax_main)
mh.mpl_magic(soft_fail=True)

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1],
 stacked_labels=['Bkg2', 'Bkg1'],
 stacked_colors=['grey', 'lightblue'],
 unstacked_components=[h_signal],
 unstacked_labels=['Signal'],
 unstacked_colors=['red'],
 model_sum_kwargs={"show": True, "label": "Total Model", "color": "violet"}, # Just like stacked_components_kwargs and unstacked_components_kwargs, you can pass kwargs for the model sum line
 xlabel='Observable [GeV]',
 ylabel='Events'
)
mh.alice.label(data=True, lumi=100, ax=ax_main)
mh.mpl_magic(soft_fail=True)

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# Create comparison plot
fig, ax_main, ax_comparison = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1, h_signal],
 stacked_labels=['Bkg2', 'Bkg1', 'Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events',
 plot_only='ax_main' # Only plot the comparison axis (works with 'ax_comparison' as well).
)
mh.alice.label(data=True, lumi=100, ax=ax_main)
mh.mpl_magic(soft_fail=True)

Different comparisonRemove MC uncertaintiesModel kwargsPlot only 1 ax

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1, h_signal],
 stacked_labels=['Bkg2', 'Bkg1', 'Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events',
 comparison='pull' # Accept any comparison from mh.comp.comparison()
)
mh.dune.label(data=True, lumi=100, ax=ax_main)

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1, h_signal],
 stacked_labels=['Bkg2', 'Bkg1', 'Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events',
 model_uncertainty=False, # Set the model uncertainty to zero
 comparison='pull', # The pull formula is updated automatically
)
mh.dune.label(data=True, lumi=100, ax=ax_main)

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# Create comparison plot
fig, ax_main, ax_comp = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1],
 stacked_labels=['Bkg2', 'Bkg1'],
 stacked_colors=['grey', 'lightblue'],
 unstacked_components=[h_signal],
 unstacked_labels=['Signal'],
 unstacked_colors=['red'],
 model_sum_kwargs={"show": True, "label": "Total Model", "color": "violet"}, # Just like stacked_components_kwargs and unstacked_components_kwargs, you can pass kwargs for the model sum line
 xlabel='Observable [GeV]',
 ylabel='Events'
)
mh.dune.label(data=True, lumi=100, ax=ax_main)

# Generate signal, backgrounds, and data
x1 = np.random.normal(0, 1, 1400)
x2 = np.random.normal(1, 0.5, 1000)
x3 = np.random.exponential(0.5, 900) - 1.5
data_x = np.random.choice([*x1, *x2, *x3], size=3500, replace=True)
h_bkg1 = hist.new.Reg(25, -3, 5).Weight().fill(x1)
h_bkg2 = hist.new.Reg(25, -3, 5).Weight().fill(x2)
h_signal = hist.new.Reg(25, -3, 5).Weight().fill(x3)
h_data = hist.new.Reg(25, -3, 5).Weight().fill(data_x)

# Create comparison plot
fig, ax_main, ax_comparison = mh.comp.data_model(
 data_hist=h_data,
 stacked_components=[h_bkg2, h_bkg1, h_signal],
 stacked_labels=['Bkg2', 'Bkg1', 'Signal'],
 xlabel='Observable [GeV]',
 ylabel='Events',
 plot_only='ax_main' # Only plot the comparison axis (works with 'ax_comparison' as well).
)
mh.dune.label(data=True, lumi=100, ax=ax_main)

Additional examples

The functions are very flexible, here is some additional examples showcasing more options and use cases. For more details, see the API Reference and the Gallery.

DefaultplothistCMSATLASLHCbALICEDUNE

# Generate training and testing histograms for two classes
x1_train = np.random.normal(0, 0.3, 2000)
x1_test = np.random.normal(0, 0.3, 1000)
x2_train = np.random.normal(1, 0.3, 2000)
x2_test = np.random.normal(1, 0.3, 1000)
h1_train = hist.new.Regular(30, 0, 1).Weight().fill(x1_train)
h2_train = hist.new.Regular(30, 0, 1).Weight().fill(x2_train)
h1_test = hist.new.Regular(30, 0, 1).Weight().fill(x1_test)
h2_test = hist.new.Regular(30, 0, 1).Weight().fill(x2_test)

# Since we have the histogram objects, we can normalize them directly
h1_train /= h1_train.sum().value
h2_train /= h2_train.sum().value
h1_test /= h1_test.sum().value
h2_test /= h2_test.sum().value

# Using the wrapper around subplots
fig, axes = mh.subplots(nrows=3)

colors = ["C0", "C1"]
labels = ["Bkg", "Signal"]

# Plotting on the main axis
for k_hist, h in enumerate([h1_train, h2_train, h1_test, h2_test]):
 color = colors[k_hist % 2]
 label = labels[k_hist % 2]
 yerr = False if k_hist < 2 else True # To make the plot less busy

 mh.histplot(
 h,
 color=color,
 label=label + (" $_{Train}$" if k_hist <2 else " $_{Test}$"),
 ax=axes[0],
 yerr=yerr,
 histtype="step" if k_hist < 2 else "errorbar",
 )

# Plotting comparisons on the other axes
mh.comp.comparison(
 h1_train,
 h1_test,
 ax=axes[1],
 comparison="ratio",
 h1_label="Train",
 h2_label="Test",
 color=colors[0],
)

mh.comp.comparison(
 h2_train,
 h2_test,
 ax=axes[2],
 comparison="pull",
 h1_label="Train",
 h2_label="Test",
 color=colors[1],
)

for ax in axes:
 ax.set_xlim(0, 1)

axes[0].set_ylabel("Normalized Entries")
axes[0].legend()
axes[1].set_ylabel("Ratio")
axes[2].set_ylabel("Pull")
axes[-1].set_xlabel("Feature")

fig.align_ylabels() # Align y labels

# Generate training and testing histograms for two classes
x1_train = np.random.normal(0, 0.3, 2000)
x1_test = np.random.normal(0, 0.3, 1000)
x2_train = np.random.normal(1, 0.3, 2000)
x2_test = np.random.normal(1, 0.3, 1000)
h1_train = hist.new.Regular(30, 0, 1).Weight().fill(x1_train)
h2_train = hist.new.Regular(30, 0, 1).Weight().fill(x2_train)
h1_test = hist.new.Regular(30, 0, 1).Weight().fill(x1_test)
h2_test = hist.new.Regular(30, 0, 1).Weight().fill(x2_test)

# Since we have the histogram objects, we can normalize them directly
h1_train /= h1_train.sum().value
h2_train /= h2_train.sum().value
h1_test /= h1_test.sum().value
h2_test /= h2_test.sum().value

# Using the wrapper around subplots
fig, axes = mh.subplots(nrows=3)

colors = ["C0", "C1"]
labels = ["Bkg", "Signal"]

# Plotting on the main axis
for k_hist, h in enumerate([h1_train, h2_train, h1_test, h2_test]):
 color = colors[k_hist % 2]
 label = labels[k_hist % 2]
 yerr = False if k_hist < 2 else True # To make the plot less busy

 mh.histplot(
 h,
 color=color,
 label=label + (" $_{Train}$" if k_hist <2 else " $_{Test}$"),
 ax=axes[0],
 yerr=yerr,
 histtype="step" if k_hist < 2 else "errorbar",
 )

# Plotting comparisons on the other axes
mh.comp.comparison(
 h1_train,
 h1_test,
 ax=axes[1],
 comparison="ratio",
 h1_label="Train",
 h2_label="Test",
 color=colors[0],
)

mh.comp.comparison(
 h2_train,
 h2_test,
 ax=axes[2],
 comparison="pull",
 h1_label="Train",
 h2_label="Test",
 color=colors[1],
)

for ax in axes:
 ax.set_xlim(0, 1)

axes[0].set_ylabel("Normalized Entries")
axes[0].legend()
axes[1].set_ylabel("Ratio")
axes[2].set_ylabel("Pull")
axes[-1].set_xlabel("Feature")

fig.align_ylabels() # Align y labels

txt_obj = mh.add_text('plothist', loc='over left', ax=ax_main)

# Generate training and testing histograms for two classes
x1_train = np.random.normal(0, 0.3, 2000)
x1_test = np.random.normal(0, 0.3, 1000)
x2_train = np.random.normal(1, 0.3, 2000)
x2_test = np.random.normal(1, 0.3, 1000)
h1_train = hist.new.Regular(30, 0, 1).Weight().fill(x1_train)
h2_train = hist.new.Regular(30, 0, 1).Weight().fill(x2_train)
h1_test = hist.new.Regular(30, 0, 1).Weight().fill(x1_test)
h2_test = hist.new.Regular(30, 0, 1).Weight().fill(x2_test)

# Since we have the histogram objects, we can normalize them directly
h1_train /= h1_train.sum().value
h2_train /= h2_train.sum().value
h1_test /= h1_test.sum().value
h2_test /= h2_test.sum().value

# Using the wrapper around subplots
fig, axes = mh.subplots(nrows=3)

colors = ["C0", "C1"]
labels = ["Bkg", "Signal"]

# Plotting on the main axis
for k_hist, h in enumerate([h1_train, h2_train, h1_test, h2_test]):
 color = colors[k_hist % 2]
 label = labels[k_hist % 2]
 yerr = False if k_hist < 2 else True # To make the plot less busy

 mh.histplot(
 h,
 color=color,
 label=label + (" $_{Train}$" if k_hist <2 else " $_{Test}$"),
 ax=axes[0],
 yerr=yerr,
 histtype="step" if k_hist < 2 else "errorbar",
 )

# Plotting comparisons on the other axes
mh.comp.comparison(
 h1_train,
 h1_test,
 ax=axes[1],
 comparison="ratio",
 h1_label="Train",
 h2_label="Test",
 color=colors[0],
)

mh.comp.comparison(
 h2_train,
 h2_test,
 ax=axes[2],
 comparison="pull",
 h1_label="Train",
 h2_label="Test",
 color=colors[1],
)

for ax in axes:
 ax.set_xlim(0, 1)

axes[0].set_ylabel("Normalized Entries")
axes[0].legend()
axes[1].set_ylabel("Ratio")
axes[2].set_ylabel("Pull")
axes[-1].set_xlabel("Feature")

fig.align_ylabels() # Align y labels

mh.cms.label(data=True, ax=ax_main)

# Generate training and testing histograms for two classes
x1_train = np.random.normal(0, 0.3, 2000)
x1_test = np.random.normal(0, 0.3, 1000)
x2_train = np.random.normal(1, 0.3, 2000)
x2_test = np.random.normal(1, 0.3, 1000)
h1_train = hist.new.Regular(30, 0, 1).Weight().fill(x1_train)
h2_train = hist.new.Regular(30, 0, 1).Weight().fill(x2_train)
h1_test = hist.new.Regular(30, 0, 1).Weight().fill(x1_test)
h2_test = hist.new.Regular(30, 0, 1).Weight().fill(x2_test)

# Since we have the histogram objects, we can normalize them directly
h1_train /= h1_train.sum().value
h2_train /= h2_train.sum().value
h1_test /= h1_test.sum().value
h2_test /= h2_test.sum().value

# Using the wrapper around subplots
fig, axes = mh.subplots(nrows=3)

colors = ["C0", "C1"]
labels = ["Bkg", "Signal"]

# Plotting on the main axis
for k_hist, h in enumerate([h1_train, h2_train, h1_test, h2_test]):
 color = colors[k_hist % 2]
 label = labels[k_hist % 2]
 yerr = False if k_hist < 2 else True # To make the plot less busy

 mh.histplot(
 h,
 color=color,
 label=label + (" $_{Train}$" if k_hist <2 else " $_{Test}$"),
 ax=axes[0],
 yerr=yerr,
 histtype="step" if k_hist < 2 else "errorbar",
 )

# Plotting comparisons on the other axes
mh.comp.comparison(
 h1_train,
 h1_test,
 ax=axes[1],
 comparison="ratio",
 h1_label="Train",
 h2_label="Test",
 color=colors[0],
)

mh.comp.comparison(
 h2_train,
 h2_test,
 ax=axes[2],
 comparison="pull",
 h1_label="Train",
 h2_label="Test",
 color=colors[1],
)

for ax in axes:
 ax.set_xlim(0, 1)

axes[0].set_ylabel("Normalized Entries")
axes[0].legend()
axes[1].set_ylabel("Ratio")
axes[2].set_ylabel("Pull")
axes[-1].set_xlabel("Feature")

fig.align_ylabels() # Align y labels

mh.atlas.label(data=True, ax=ax_main)
mh.mpl_magic(soft_fail=True)

# Generate training and testing histograms for two classes
x1_train = np.random.normal(0, 0.3, 2000)
x1_test = np.random.normal(0, 0.3, 1000)
x2_train = np.random.normal(1, 0.3, 2000)
x2_test = np.random.normal(1, 0.3, 1000)
h1_train = hist.new.Regular(30, 0, 1).Weight().fill(x1_train)
h2_train = hist.new.Regular(30, 0, 1).Weight().fill(x2_train)
h1_test = hist.new.Regular(30, 0, 1).Weight().fill(x1_test)
h2_test = hist.new.Regular(30, 0, 1).Weight().fill(x2_test)

# Since we have the histogram objects, we can normalize them directly
h1_train /= h1_train.sum().value
h2_train /= h2_train.sum().value
h1_test /= h1_test.sum().value
h2_test /= h2_test.sum().value

# Using the wrapper around subplots
fig, axes = mh.subplots(nrows=3)

colors = ["C0", "C1"]
labels = ["Bkg", "Signal"]

# Plotting on the main axis
for k_hist, h in enumerate([h1_train, h2_train, h1_test, h2_test]):
 color = colors[k_hist % 2]
 label = labels[k_hist % 2]
 yerr = False if k_hist < 2 else True # To make the plot less busy

 mh.histplot(
 h,
 color=color,
 label=label + (" $_{Train}$" if k_hist <2 else " $_{Test}$"),
 ax=axes[0],
 yerr=yerr,
 histtype="step" if k_hist < 2 else "errorbar",
 )

# Plotting comparisons on the other axes
mh.comp.comparison(
 h1_train,
 h1_test,
 ax=axes[1],
 comparison="ratio",
 h1_label="Train",
 h2_label="Test",
 color=colors[0],
)

mh.comp.comparison(
 h2_train,
 h2_test,
 ax=axes[2],
 comparison="pull",
 h1_label="Train",
 h2_label="Test",
 color=colors[1],
)

for ax in axes:
 ax.set_xlim(0, 1)

axes[0].set_ylabel("Normalized Entries")
axes[0].legend()
axes[1].set_ylabel("Ratio")
axes[2].set_ylabel("Pull")
axes[-1].set_xlabel("Feature")

fig.align_ylabels() # Align y labels

mh.lhcb.label(data=True, ax=ax_main)

# Generate training and testing histograms for two classes
x1_train = np.random.normal(0, 0.3, 2000)
x1_test = np.random.normal(0, 0.3, 1000)
x2_train = np.random.normal(1, 0.3, 2000)
x2_test = np.random.normal(1, 0.3, 1000)
h1_train = hist.new.Regular(30, 0, 1).Weight().fill(x1_train)
h2_train = hist.new.Regular(30, 0, 1).Weight().fill(x2_train)
h1_test = hist.new.Regular(30, 0, 1).Weight().fill(x1_test)
h2_test = hist.new.Regular(30, 0, 1).Weight().fill(x2_test)

# Since we have the histogram objects, we can normalize them directly
h1_train /= h1_train.sum().value
h2_train /= h2_train.sum().value
h1_test /= h1_test.sum().value
h2_test /= h2_test.sum().value

# Using the wrapper around subplots
fig, axes = mh.subplots(nrows=3)

colors = ["C0", "C1"]
labels = ["Bkg", "Signal"]

# Plotting on the main axis
for k_hist, h in enumerate([h1_train, h2_train, h1_test, h2_test]):
 color = colors[k_hist % 2]
 label = labels[k_hist % 2]
 yerr = False if k_hist < 2 else True # To make the plot less busy

 mh.histplot(
 h,
 color=color,
 label=label + (" $_{Train}$" if k_hist <2 else " $_{Test}$"),
 ax=axes[0],
 yerr=yerr,
 histtype="step" if k_hist < 2 else "errorbar",
 )

# Plotting comparisons on the other axes
mh.comp.comparison(
 h1_train,
 h1_test,
 ax=axes[1],
 comparison="ratio",
 h1_label="Train",
 h2_label="Test",
 color=colors[0],
)

mh.comp.comparison(
 h2_train,
 h2_test,
 ax=axes[2],
 comparison="pull",
 h1_label="Train",
 h2_label="Test",
 color=colors[1],
)

for ax in axes:
 ax.set_xlim(0, 1)

axes[0].set_ylabel("Normalized Entries")
axes[0].legend()
axes[1].set_ylabel("Ratio")
axes[2].set_ylabel("Pull")
axes[-1].set_xlabel("Feature")

fig.align_ylabels() # Align y labels

mh.alice.label(data=True, ax=ax_main)
mh.mpl_magic(soft_fail=True)

# Generate training and testing histograms for two classes
x1_train = np.random.normal(0, 0.3, 2000)
x1_test = np.random.normal(0, 0.3, 1000)
x2_train = np.random.normal(1, 0.3, 2000)
x2_test = np.random.normal(1, 0.3, 1000)
h1_train = hist.new.Regular(30, 0, 1).Weight().fill(x1_train)
h2_train = hist.new.Regular(30, 0, 1).Weight().fill(x2_train)
h1_test = hist.new.Regular(30, 0, 1).Weight().fill(x1_test)
h2_test = hist.new.Regular(30, 0, 1).Weight().fill(x2_test)

# Since we have the histogram objects, we can normalize them directly
h1_train /= h1_train.sum().value
h2_train /= h2_train.sum().value
h1_test /= h1_test.sum().value
h2_test /= h2_test.sum().value

# Using the wrapper around subplots
fig, axes = mh.subplots(nrows=3)

colors = ["C0", "C1"]
labels = ["Bkg", "Signal"]

# Plotting on the main axis
for k_hist, h in enumerate([h1_train, h2_train, h1_test, h2_test]):
 color = colors[k_hist % 2]
 label = labels[k_hist % 2]
 yerr = False if k_hist < 2 else True # To make the plot less busy

 mh.histplot(
 h,
 color=color,
 label=label + (" $_{Train}$" if k_hist <2 else " $_{Test}$"),
 ax=axes[0],
 yerr=yerr,
 histtype="step" if k_hist < 2 else "errorbar",
 )

# Plotting comparisons on the other axes
mh.comp.comparison(
 h1_train,
 h1_test,
 ax=axes[1],
 comparison="ratio",
 h1_label="Train",
 h2_label="Test",
 color=colors[0],
)

mh.comp.comparison(
 h2_train,
 h2_test,
 ax=axes[2],
 comparison="pull",
 h1_label="Train",
 h2_label="Test",
 color=colors[1],
)

for ax in axes:
 ax.set_xlim(0, 1)

axes[0].set_ylabel("Normalized Entries")
axes[0].legend()
axes[1].set_ylabel("Ratio")
axes[2].set_ylabel("Pull")
axes[-1].set_xlabel("Feature")

fig.align_ylabels() # Align y labels

mh.dune.label(data=True, ax=ax_main)