Example analysisΒΆ
An example analysis is the one run when running the module, that is, the
__main__.py file. For convenience, we report the source code here.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 | # Copyright (C) 2021 Ludovico Massaccesi
#
# This file is part of upsilon_analysis.
#
# upsilon_analysis is free software: you can redistribute it and/or
# modify it under the terms of the GNU General Public License as
# published by the Free Software Foundation, either version 3 of the
# License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program. If not, see <https://www.gnu.org/licenses/>.
"""Example analysis using the upsilon_analysis package."""
import logging
import argparse
import os
import ROOT
from . import *
from .utils import print_fit_results
def args_for(func, kwargs):
"""Strips ``kwargs`` of the names that func does not require."""
return {k: v for k, v in kwargs.items() if k in func.__code__.co_varnames}
def iter_csv(file_path):
"""Simple CSV parser."""
with open(file_path) as ifs:
for ln in ifs:
fields = ln.strip().split(",")
if len(fields) != 0:
try:
yield tuple(float(x.strip()) for x in fields)
except ValueError: # Non-numeric values
pass
if __name__ == "__main__":
parser = argparse.ArgumentParser(description=__doc__)
parser.add_argument("--input-file", "-i", metavar="PATH",
default=("root://eospublic.cern.ch//eos/opendata/cms"
"/derived-data/AOD2NanoAODOutreachTool"
"/Run2012BC_DoubleMuParked_Muons.root"),
help=("Input file to be used, optional; the default "
"file is opened from root://eospublic.cern.ch; "
"any URL supported by RDataFrame can be used."))
parser.add_argument("--pt-min", type=float, default=10., metavar="MIN",
help="Minimum resonance pt (GeV/c); default 10.")
parser.add_argument("--pt-max", type=float, default=100., metavar="MAX",
help="Maximum resonance pt (GeV/c); default 100.")
parser.add_argument("--pt-bin-width", type=float, default=2., metavar="W",
help=("Width of the pt bins (GeV/c); above 40 GeV/c "
"bins will be made larger; default 2."))
parser.add_argument("--y-min", type=float, default=0., metavar="MIN",
help=("Minimum resonance rapidity (absolute value); "
"default 0."))
parser.add_argument("--y-max", type=float, default=1.2, metavar="MAX",
help=("Maximum resonance rapidity (absolute value); "
"default 1.2"))
parser.add_argument("--y-bins", type=float, default=2, metavar="N",
help=("Number of resonance rapidity (absolute value) "
"bins; default 2."))
parser.add_argument("--mass-bins", type=int, default=100, metavar="N",
help=("Number of invariant mass bins; default 100; "
"histograms with too few events are rebinned."))
parser.add_argument("--no-quality", action="store_true",
help="Skip muon quality cuts.")
parser.add_argument("-v", action="store_true", help="Verbose mode.")
parser.add_argument("--vv", action="store_true", help="Very verbose mode.")
parser.add_argument("--threads", "-j", type=int, default=0, metavar="N",
help=("Number of threads, see ROOT::EnableImplicitMT; "
"chosen automatically by ROOT by default; if "
"set to 1, MT is not enabled at all."))
parser.add_argument("--output-dir", "-o", default=".", metavar="DIR",
help="Output directory for the plots; default is cd.")
parser.add_argument("--max-mass-delta", type=float, default=0.025,
help=("Max delta (in GeV) between fitted and known "
"resonance mass to consider the fit good; "
"default 0.025; known masses are from PDG."))
parser.add_argument("--luminosity", type=float, default=1,
help="The integrated luminosity of the sample.")
parser.add_argument("--luminosity-units", default=None,
help="The int. lumi. units for the plots' labels.")
parser.add_argument("--efficiency-table", default=None,
help=("A CSV file with the efficiency per bin, whose "
"columns are y_min, y_max, pt_min, pt_max, "
"eff_y1, eff_y2, eff_y3. If you have a global "
"efficiency for all bins, multiply luminosity "
"by it."))
args = parser.parse_args()
kwargs = vars(args)
logging.basicConfig(level=(logging.DEBUG if args.vv else
(logging.INFO if args.v else logging.WARNING)))
# Data reading, reconstruction, cuts, mass histograms
if args.threads != 1:
logging.debug("Enabling MT")
ROOT.EnableImplicitMT(args.threads)
df = build_dataframe(**args_for(build_dataframe, kwargs))
mass_histos = book_histograms(df, **args_for(book_histograms, kwargs))
logging.info("Actually running the analysis with the RDataFrame")
df.Report().Print() # Here all the booked actions are actually run
# Fitting, plotting, saving to disk plots and fit results
if not os.path.isdir(args.output_dir):
logging.info("Creating output directory %s", args.output_dir)
os.mkdir(args.output_dir)
fits = fit_histograms(mass_histos, **args_for(fit_histograms, kwargs))
# Save fit results to a CSV file, for importing use TTree::ReadFile
out_csv = os.path.join(args.output_dir, "fit_results.csv")
logging.info("Saving fit results to %s", out_csv)
with open(out_csv, "w") as ofs:
print_fit_results(fits, file=ofs)
# Differential cross sections
canvas = ROOT.TCanvas("", "", 640, 480)
ROOT.gStyle.SetOptStat(10)
ROOT.gStyle.SetOptFit(100)
ROOT.gStyle.SetStatX(0.91)
ROOT.gStyle.SetStatY(0.91)
ROOT.gStyle.SetStatH(0.05)
ROOT.gStyle.SetStatW(0.15)
out_pdf = os.path.join(args.output_dir, "cross_section_plots.pdf")
logging.info("Saving xsec plots to %s", out_pdf)
canvas.Print(f"{out_pdf}[") # Open PDF to plot multiple pages
for (y_low, y_high), pt_bins in fits.items():
# Cut the bin with all the data and those with bad fits
del pt_bins[(args.pt_min, args.pt_max)]
ok_bins = {k: (v.y1.a, v.y2.a, v.y3.a) for k, v in pt_bins.items()
if v.y1.a >= 0 and v.y2.a >= 0 and v.y3.a >= 0
and v.y1.a + v.y2.a + v.y3.a < v.nevt
and abs(v.y1.m - 9.4603) <= args.max_mass_delta
and abs(v.y2.m - 10.0232) <= args.max_mass_delta
and abs(v.y3.m - 10.3552) <= args.max_mass_delta}
logging.debug("Discarded %d bins due to wrong fitted mass(es) "
"(xsec plots %g<y<%g)", len(pt_bins) - len(ok_bins),
y_low, y_high)
if len(ok_bins) < 2:
logging.warning("Not enough pt bins for xsec plots (%g<y<%g)",
y_low, y_high)
continue
if args.efficiency_table is None:
eff = [1, 1, 1]
else:
eff = [{}, {}, {}]
for ln in iter_csv(args.efficiency_table):
if ln[0] == y_low and ln[1] == y_high:
eff[0][(ln[2], ln[3])] = ln[4]
eff[1][(ln[2], ln[3])] = ln[5]
eff[2][(ln[2], ln[3])] = ln[6]
# Graphs for the three Ys' cross sections
graphs = [
build_cross_section_graph({k: v[n] for k, v in ok_bins.items()},
args.luminosity, eff[n])
for n in range(3)
]
for n, graph in enumerate(graphs):
graph.SetTitle(f"#Upsilon({n+1}s), |y| #in [{y_low:g},{y_high:g})")
if args.luminosity_units is not None:
graph.GetYaxis().SetTitle("d#sigma/dp_{T}#times#it{Br} "
f"[{args.luminosity_units}/GeV]")
graph.SetLineWidth(2)
graph.Draw("APZ")
canvas.SetGrid()
canvas.SetLogy()
canvas.Print(out_pdf, f"Title:Y({n+1}s), y ({y_low:g},{y_high:g})")
canvas.Print(f"{out_pdf}]") # Close PDF
del canvas
|