def __init__(self):
super(SamplesBase, self).__init__()
self.config = jsonTools.JsonDict({})
self.postfit_scales = None
self.expressions = expressions.ExpressionsDict()
self.exclude_cuts = []
self.period = "run"
def create_input_root_files(datacards, args):
''' Configuring Harry plotter according to the samples and creating input root files according to the args.'''
plot_configs = []
output_files = []
merged_output_files = []
hadd_commands = []
sample_settings = samples.Samples()
expression_settings = expressions.ExpressionsDict()
binnings_settings = binnings.BinningsDict()
systematics_factory = systematics.SystematicsFactory()
datacards.configs._mapping_process2sample = {
"data_obs": "data",
"EWKZ": "ewkz",
"QCD": "qcd",
"TT": "ttj",
"TTT": "ttt",
"TTJ": "ttjj",
"VV": "vv",
"VVT": "vvt",
"VVJ": "vvj",
"W": "wj",
"ZJ": "zj",
"ZL": "zl",
"ZLL": "zll",
"ZTTPOSPOL": "zttpospol",
"ZTTNEGPOL": "zttnegpol",
"ZTT_GEN_DM_ZERO": "ztt_gen_dm_zero",
"ZTT_GEN_DM_ONE": "ztt_gen_dm_one",
"ZTT_GEN_DM_TWO": "ztt_gen_dm_two",
"ZTT_GEN_DM_TEN": "ztt_gen_dm_ten",
"ZTT_GEN_DM_ELEVEN": "ztt_gen_dm_eleven",
}
for index, (channel,
categories) in enumerate(zip(args.channel, args.categories)):
for category in categories:
datacards_per_channel_category = datacardsbase.Datacards(
cb=datacards.cb.cp().channel([channel]).bin([category]))
higgs_masses = [
mass for mass in datacards_per_channel_category.cb.mass_set()
if mass != "*"
]
output_file = os.path.join(
args.output_dir,
"input/{ANALYSIS}_{CHANNEL}_{BIN}_{ERA}.root".format(
ANALYSIS="ztt", CHANNEL=channel, BIN=category,
ERA="13TeV"))
output_files.append(output_file)
tmp_output_files = []
for shape_systematic, list_of_samples in datacards_per_channel_category.get_samples_per_shape_systematic(
).iteritems():
nominal = (shape_systematic == "nominal")
list_of_samples = [
datacards.configs.process2sample(process)
for process in list_of_samples
]
if ("wj"
in list_of_samples) and not ("qcd" in list_of_samples):
list_of_samples.append("qcd")
elif ("qcd"
in list_of_samples) and not ("wj" in list_of_samples):
list_of_samples.append("wj")
asimov_nicks = []
if args.use_asimov_dataset:
asimov_nicks = [
nick.replace("zttpospol", "zttpospol_noplot").replace(
"zttnegpol", "zttnegpol_noplot")
for nick in list_of_samples
]
if "data" in asimov_nicks:
asimov_nicks.remove("data")
for shift_up in ([True] if nominal else [True, False]):
systematic = "nominal" if nominal else (
shape_systematic + ("Up" if shift_up else "Down"))
log.debug(
"Create inputs for (samples, systematic) = ([\"{samples}\"], {systematic}), (channel, category) = ({channel}, {category})."
.format(samples="\", \"".join(list_of_samples),
channel=channel,
category=category,
systematic=systematic))
tmp_quantity = args.quantity
tmp_omega_version = args.omega_version
if args.fixed_variables == "best_choice":
if channel in ["tt"]:
if category in [
channel + "_" + cat for cat in [
"combined_rho_oneprong",
"combined_oneprong_oneprong"
]
]:
tmp_quantity = "m_vis"
tmp_omega_version = None
elif category in [
channel + "_" + cat
for cat in ["combined_a1_rho"]
]:
tmp_quantity = None
tmp_omega_version = None
elif category in [
channel + "_" + cat for cat in
["combined_a1_a1", "combined_a1_oneprong"]
]:
tmp_quantity = None
tmp_omega_version = "BarSvfitM91"
elif category in [
channel + "_" + cat
for cat in ["combined_rho_rho", "rho"]
]:
tmp_quantity = None
tmp_omega_version = "VisibleSvfit"
elif channel in ["mt", "et"]:
if category in [
channel + "_" + cat
for cat in ["combined_a1_oneprong"]
]:
tmp_quantity = "m_vis"
tmp_omega_version = None
elif category in [
channel + "_" + cat
for cat in ["combined_rho_oneprong"]
]:
tmp_quantity = None
tmp_omega_version = None
elif category in [
channel + "_" + cat for cat in [
"combined_oneprong_oneprong", "a1",
"oneprong"
]
]:
tmp_quantity = None
tmp_omega_version = "BarSvfitM91"
elif category in [
channel + "_" + cat for cat in ["rho"]
]:
tmp_quantity = None
tmp_omega_version = "VisibleSvfit"
elif channel in ["em"]:
if category in [
channel + "_" + cat
for cat in ["combined_oneprong_oneprong"]
]:
tmp_quantity = "m_vis"
tmp_omega_version = None
elif args.fixed_variables == "best_choice_no_svfit":
tmp_quantity = "m_vis"
tmp_omega_version = None
if channel in ["tt", "mt", "et"]:
if category in [
channel + "_" + cat
for cat in ["combined_rho_rho", "rho"]
]:
tmp_quantity = None
tmp_omega_version = "VisibleSvfit"
x_expression = None
if tmp_quantity:
x_expression = tmp_quantity
else:
x_expression = "testZttPol13TeV_" + category
if tmp_omega_version:
x_expression = expression_settings.expressions_dict[
x_expression].replace("BarSvfit",
tmp_omega_version)
x_expression = expression_settings.expressions_dict.get(
x_expression, x_expression)
# prepare plotting configs for retrieving the input histograms
config = sample_settings.get_config(
samples=[
getattr(samples.Samples, sample)
for sample in list_of_samples
],
channel=channel,
category="catZttPol13TeV_" + category,
weight=args.weight,
lumi=args.lumi * 1000,
higgs_masses=higgs_masses,
estimationMethod="new",
polarisation_bias_correction=True,
cut_type="low_mvis_smhtt2016",
exclude_cuts=(["m_vis"]
if x_expression == "m_vis" else []),
no_ewk_samples=args.no_ewk_samples,
no_ewkz_as_dy=True,
asimov_nicks=asimov_nicks)
systematics_settings = systematics_factory.get(
shape_systematic)(config)
# TODO: evaluate shift from datacards_per_channel_category.cb
config = systematics_settings.get_config(shift=(
0.0 if nominal else (1.0 if shift_up else -1.0)))
#config["qcd_subtract_shape"] =[args.qcd_subtract_shapes]
config["x_expressions"] = [
("0" if (("gen_zttpospol" in nick) or
("gen_zttnegpol" in nick)) else x_expression)
for nick in config["nicks"]
]
binnings_key = "binningZttPol13TeV_" + category + "_" + x_expression
if not (binnings_key in binnings_settings.binnings_dict):
binnings_key = "binningZttPol13TeV_" + category + (
("_" + tmp_quantity) if tmp_quantity else "")
if binnings_key in binnings_settings.binnings_dict:
config["x_bins"] = [
("1,-1,1" if
(("gen_zttpospol" in nick) or
("gen_zttnegpol" in nick)) else binnings_key)
for nick in config["nicks"]
]
if args.fixed_binning:
if args.fixed_variables:
if channel == "tt":
config["x_bins"] = [
args.fixed_binning.split(",")[0] +
",-1.0001,1.0001"
for nick in config["nicks"]
]
else:
config["x_bins"] = [
args.fixed_binning
for nick in config["nicks"]
]
else:
config["x_bins"] = [
args.fixed_binning for nick in config["nicks"]
]
config["directories"] = [args.input_dir]
histogram_name_template = "${BIN}/${PROCESS}" if nominal else "${BIN}/${PROCESS}_${SYSTEMATIC}"
config["labels"] = [
histogram_name_template.replace("$", "").format(
PROCESS=datacards.configs.sample2process(
sample.replace("asimov", "data")),
BIN=category,
SYSTEMATIC=systematic)
for sample in config["labels"]
]
tmp_output_file = os.path.join(
args.output_dir,
"input/{ANALYSIS}_{CHANNEL}_{BIN}_{SYSTEMATIC}_{ERA}.root"
.format(ANALYSIS="ztt",
CHANNEL=channel,
BIN=category,
SYSTEMATIC=systematic,
ERA="13TeV"))
tmp_output_files.append(tmp_output_file)
config["output_dir"] = os.path.dirname(tmp_output_file)
config["filename"] = os.path.splitext(
os.path.basename(tmp_output_file))[0]
config["plot_modules"] = ["ExportRoot"]
config["file_mode"] = "UPDATE"
if "legend_markers" in config:
config.pop("legend_markers")
plot_configs.append(config)
hadd_commands.append("hadd -f {DST} {SRC} && rm {SRC}".format(
DST=output_file, SRC=" ".join(tmp_output_files)))
tmp_output_files = list(
set([
os.path.join(config["output_dir"], config["filename"] + ".root")
for config in plot_configs[:args.n_plots[0]]
]))
for output_file in tmp_output_files:
if os.path.exists(output_file):
os.remove(output_file)
log.debug("Removed file \"" + output_file +
"\" before it is recreated again.")
output_files = list(set(output_files))
higgsplot.HiggsPlotter(list_of_config_dicts=plot_configs,
list_of_args_strings=[args.args],
n_processes=args.n_processes,
n_plots=args.n_plots[0],
batch=args.batch)
if args.n_plots[0] != 0:
tools.parallelize(_call_command,
hadd_commands,
n_processes=args.n_processes)
debug_plot_configs = []
for output_file in (output_files):
debug_plot_configs.extend(plotconfigs.PlotConfigs().all_histograms(
output_file,
plot_config_template={
"markers": ["E"],
"colors": ["#FF0000"]
}))
if args.www:
for debug_plot_config in debug_plot_configs:
debug_plot_config["www"] = debug_plot_config["output_dir"].replace(
args.output_dir, args.www)
#higgsplot.HiggsPlotter(list_of_config_dicts=debug_plot_configs, list_of_args_strings=[args.args], n_processes=args.n_processes, n_plots=args.n_plots[0])
return None
help="Channel. [Default: %(default)s]")
parser.add_argument("--title",
default="Category Overlap",
help="Title [Default: %(default)s]")
parser.add_argument("-e", "--exclude-cuts", nargs="+", default=[],
help="Exclude (default) selection cuts. [Default: %(default)s]")
parser.add_argument("-r", "--ratio", nargs="+", default=["qqh"],
help="List of Numerator samples. [Default: %(default)s]")
parser.add_argument("--ratio-title",
default="VBF/ggh",
help="Title for ratio [Default: %(default)s]")
args = parser.parse_args()
logger.initLogger(args)
exp_dict = expressions.ExpressionsDict()
plot_configs = []
list_of_samples = [getattr(samples.Samples, sample) for sample in args.Samples]
sample_settings = samples.Samples()
config = sample_settings.get_config(
samples=list_of_samples,
channel=args.channel,
category="1.0",
weight="1.0",
higgs_masses=args.higgs_masses,
normalise_signal_to_one_pb=False,
exclude_cuts=args.exclude_cuts
)
firsts = [exp_dict.replace_expressions(s) for s in args.first_category]
seconds = [exp_dict.replace_expressions(s) for s in args.second_category]
