From 975f20d44a790fed79596302f31308755afbc327 Mon Sep 17 00:00:00 2001
From: Frank Sauerburger <f.sauerburger@cern.ch>
Date: Thu, 28 Nov 2019 14:38:24 +0100
Subject: [PATCH] Add density argument to hist()
---
examples/Histogram.ipynb | 17 +++++++++++++++++
nnfwtbn/plot.py | 19 ++++++++++++++++---
2 files changed, 33 insertions(+), 3 deletions(-)
diff --git a/examples/Histogram.ipynb b/examples/Histogram.ipynb
index 215bd2f..2a2718f 100644
--- a/examples/Histogram.ipynb
+++ b/examples/Histogram.ipynb
@@ -167,6 +167,23 @@
"None"
]
},
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "colors = [\"windows blue\", \"amber\", \"greyish\", \"faded green\", \"dusty purple\"]\n",
+ "colors = sns.xkcd_palette(colors)\n",
+ "\n",
+ "s_sig = McStack(p_sig, color=colors[0], histtype='step')\n",
+ "s_ztt = McStack(p_ztt, color=colors[1], histtype='step')\n",
+ "\n",
+ "hist(df, v_higgs_m, 20, [s_sig, s_ztt], range=(0, 200), selection=None,\n",
+ " weight=\"weight\", numerator=None, density=True)\n",
+ "None"
+ ]
+ },
{
"cell_type": "code",
"execution_count": null,
diff --git a/nnfwtbn/plot.py b/nnfwtbn/plot.py
index 1814a6d..01a8b90 100644
--- a/nnfwtbn/plot.py
+++ b/nnfwtbn/plot.py
@@ -60,7 +60,7 @@ def hist(dataframe, variable, bins, stacks, selection=None,
weight=None, y_log=False, y_min=None, vlines=[],
denominator=0, numerator=-1, ratio_label=None, diff=False,
ratio_range=None, atlas=None, info=None, enlarge=1.3,
- **kwds):
+ density=False, **kwds):
"""
Creates a histogram of stacked processes. The first argument is the
dataframe to operate on. The 'variable' argument defines the x-axis. The
@@ -124,6 +124,9 @@ def hist(dataframe, variable, bins, stacks, selection=None,
The module constants ATLAS and INFO are passed to atlasify. Overwrite them
to change the badges.
+
+ If the density argument is True, the area of each stack is normalized to
+ unity.
"""
# Wrap column string by variable
if isinstance(variable, str):
@@ -219,17 +222,26 @@ def hist(dataframe, variable, bins, stacks, selection=None,
c_blind = variable.blinding(variable, bins)
else:
c_blind = lambda d: d
+
+ density_norm = 1.0
+ if density:
+ density_norm = stack.get_total(dataframe,
+ [float('-inf'), float('inf')],
+ variable, weight).sum()
for i_process in range_(len(stack)):
process = stack.processes[i_process]
histogram = stack.get_hist(c_blind(dataframe), i_process, bins,
variable, weight)
+ histogram = histogram / density_norm
+
# Draw uncertainty iff this is the last processes of the stack
if i_process == len(stack) - 1:
uncertainty = stack.get_total_uncertainty(c_blind(dataframe),
bins, variable,
weight)
+ uncertainty = uncertainty / density_norm
# Draw process
histtype = stack.get_histtype(i_process)
@@ -317,11 +329,12 @@ def hist(dataframe, variable, bins, stacks, selection=None,
horizontalalignment='right', x=0.95)
if equidistant_bins:
+ subject = "Fraction" if density else "Events"
if variable.unit is not None:
- axes.set_ylabel("Events / %g %s" % (bins[1] - bins[0], variable.unit),
+ axes.set_ylabel(f"{subject} / %g %s" % (bins[1] - bins[0], variable.unit),
horizontalalignment='right', y=0.95)
else:
- axes.set_ylabel("Events / %g" % (bins[1] - bins[0]),
+ axes.set_ylabel(f"{subject} / %g" % (bins[1] - bins[0]),
horizontalalignment='right', y=0.95)
else:
axes.set_ylabel("Events / Bin", horizontalalignment='right', y=0.95)
--
GitLab