update

src/HHbbVV/postprocessing/PostProcessRes.ipynb

@@ -77,7 +77,12 @@
    "outputs": [],
    "source": [
     "# del nonres_samples[\"VBFHHbbVV\"]\n",
-    "nonres_sig_keys = [\"HHbbVV\", \"VBFHHbbVV\"]\n",
+    "nonres_sig_keys = [\n",
+    "    \"HHbbVV\",\n",
+    "    \"VBFHHbbVV\",\n",
+    "    \"qqHH_CV_1_C2V_1_kl_2_HHbbVV\",\n",
+    "    \"qqHH_CV_1_C2V_2_kl_1_HHbbVV\",\n",
+    "]\n",
     "nonres_samples = {key: nonres_samples[key] for key in nonres_sig_keys}\n",
     "\n",
     "# bg_keys = [\"QCD\", \"TT\", \"ST\", \"V+Jets\", \"Diboson\"]\n",
@@ -91,7 +96,7 @@
    "outputs": [],
    "source": [
     "MAIN_DIR = Path(\"../../../\")\n",
-    "samples_dir = MAIN_DIR / \"../data/skimmer/24Mar6AllYearsBDTVars\"\n",
+    "samples_dir = MAIN_DIR / \"../data/skimmer/24Mar14UpdateData\"\n",
     "# samples_dir = f\"{MAIN_DIR}/../data/skimmer/Feb24\"\n",
     "# nonres_signal_samples_dir = f\"{MAIN_DIR}/../data/skimmer/Jun10\"\n",
     "# res_signal_samples_dir = f\"{MAIN_DIR}/../data/skimmer/Apr11\"\n",
@@ -100,7 +105,7 @@
     "# res_signal_samples_dir = \"/eos/uscms/store/user/rkansal/bbVV/skimmer/Apr11/\"\n",
     "year = \"2016APV\"\n",
     "\n",
-    "date = \"24Mar6\"\n",
+    "date = \"24Mar18\"\n",
     "plot_dir = MAIN_DIR / f\"plots/PostProcessing/{date}/\"\n",
     "templates_dir = Path(f\"templates/{date}/\")\n",
     "\n",
@@ -138,6 +143,7 @@
     "    {**nonres_samples},  # , **res_samples, **samples},\n",
     "    year,\n",
     "    load_filters,\n",
+    "    variations=False,\n",
     ")\n",
     "\n",
     "utils.add_to_cutflow(events_dict, \"Preselection\", \"finalWeight\", cutflow)\n",
@@ -150,44 +156,34 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "events = events_dict[\"HHbbVV\"]"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "list(events.columns)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "events[\"DijetMass\"]"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "events = pd.read_parquet(f\"{samples_dir}/{year}/GluGluToHHTobbVV_node_cHHH1/parquet\")\n",
-    "events"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "list(events.columns)"
+    "import matplotlib as mpl\n",
+    "import matplotlib.pyplot as plt\n",
+    "import matplotlib.ticker as mticker\n",
+    "import mplhep as hep\n",
+    "\n",
+    "plt.style.use(hep.style.CMS)\n",
+    "hep.style.use(\"CMS\")\n",
+    "formatter = mticker.ScalarFormatter(useMathText=True)\n",
+    "formatter.set_powerlimits((-3, 3))\n",
+    "\n",
+    "# this is needed for some reason to update the font size for the first plot\n",
+    "fig, ax = plt.subplots(1, 1, figsize=(12, 12))\n",
+    "plt.rcParams.update({\"font.size\": 24})\n",
+    "plt.close()\n",
+    "\n",
+    "fig, ax = plt.subplots(1, 1, figsize=(8, 8))\n",
+    "for sample, events in events_dict.items():\n",
+    "    plt.hist(\n",
+    "        events[\"ak8FatJetPt\"].to_numpy().reshape(-1),\n",
+    "        bins=np.arange(300, 1200, 50),\n",
+    "        label=sample,\n",
+    "        histtype=\"step\",\n",
+    "        density=True,\n",
+    "    )\n",
+    "\n",
+    "plt.legend()\n",
+    "plt.xlabel(\"AK8 Jet pT (GeV)\")\n",
+    "plt.show()"
    ]
   },
   {
@@ -218,16 +214,6 @@
     "Control Plots"
    ]
   },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "events = events_dict[\"HHbbVV\"]\n",
-    "events['VBFJetPt'][]\n"
-   ]
-  },
   {
    "cell_type": "code",
    "execution_count": null,

src/HHbbVV/postprocessing/bash_scripts/LPSFYearCheck.sh

@@ -0,0 +1,51 @@
+#!/bin/bash
+# shellcheck disable=SC2086,SC2043
+
+####################################################################################################
+# Checking LP SF for each year
+# Author: Raghav Kansal
+####################################################################################################
+
+MAIN_DIR="../../.."
+data_dir="$MAIN_DIR/../data/skimmer/24Mar14UpdateData"
+TAG=""
+
+
+options=$(getopt -o "" --long "tag:" -- "$@")
+eval set -- "$options"
+
+while true; do
+    case "$1" in
+        --tag)
+            shift
+            TAG=$1
+            ;;
+        --)
+            shift
+            break;;
+        \?)
+            echo "Invalid option: -$OPTARG" >&2
+            exit 1
+            ;;
+        :)
+            echo "Option -$OPTARG requires an argument." >&2
+            exit 1
+            ;;
+    esac
+    shift
+done
+
+if [[ -z $TAG ]]; then
+  echo "Tag required using the --tag option. Exiting"
+  exit 1
+fi
+
+for year in 2016APV 2016 2017 2018
+do
+    # --sig-samples qqHH_CV_1_C2V_1_kl_2_HHbbVV --bg-keys "" --no-data \
+    python -u postprocessing.py --year $year --data-dir "$data_dir" --templates \
+    --no-lp-sf-all-years \
+    --bdt-preds-dir "$data_dir/24_03_07_new_samples_max_depth_5/inferences" \
+    --plot-dir "${MAIN_DIR}/plots/PostProcessing/$TAG" \
+    --template-dir "test_templates/$TAG$year" --no-do-jshifts --vbf
+done

src/HHbbVV/postprocessing/bash_scripts/MassPlots.sh

@@ -13,15 +13,20 @@
 ####################################################################################################
 
 MAIN_DIR="../../.."
+data_dir="$MAIN_DIR/../data/skimmer/24Mar14UpdateData"
 TAG=""
 samples="HHbbVV VBFHHbbVV NMSSM_XToYHTo2W2BTo4Q2B_MX-900_MY-80 NMSSM_XToYHTo2W2BTo4Q2B_MX-1200_MY-190 NMSSM_XToYHTo2W2BTo4Q2B_MX-2000_MY-125 NMSSM_XToYHTo2W2BTo4Q2B_MX-3000_MY-250 NMSSM_XToYHTo2W2BTo4Q2B_MX-4000_MY-150"
-# samples="HHbbVV VBFHHbbVV NMSSM_XToYHTo2W2BTo4Q2B_MX-900_MY-80"
+resonant="--resonant"
 
-options=$(getopt -o "" --long "tag:" -- "$@")
+options=$(getopt -o "" --long "nonresonant,tag:" -- "$@")
 eval set -- "$options"
 
 while true; do
     case "$1" in
+        --nonresonant)
+            resonant=""
+            samples="HHbbVV VBFHHbbVV qqHH_CV_1_C2V_0_kl_1_HHbbVV qqHH_CV_1_C2V_2_kl_1_HHbbVV"
+            ;;
         --tag)
             shift
             TAG=$1
@@ -46,11 +51,13 @@ if [[ -z $TAG ]]; then
   exit 1
 fi
 
-for year in 2016APV 2016 2017 2018
+# for year in 2016APV 2016 2017 2018
+for year in 2016APV 2016 2017
 do
-    python -u postprocessing.py --control-plots --year $year --resonant \
-    --data-dir "${MAIN_DIR}/../data/skimmer/24Mar5AllYears" \
+    python -u postprocessing.py --control-plots --year $year $resonant \
+    --data-dir $data_dir \
     --sig-samples $samples \
     --plot-dir "${MAIN_DIR}/plots/PostProcessing/$TAG" \
+    --bdt-preds-dir "$data_dir/24_03_07_new_samples_max_depth_5/inferences" \
     --mass-plots
 done

src/HHbbVV/postprocessing/bash_scripts/NonresTemplates.sh

@@ -40,8 +40,7 @@ if [[ -z $TAG ]]; then
   exit 1
 fi
 
-# for year in 2016APV 2016 2017 2018
-for year in 2017 2018 2016APV 2016
+for year in 2016APV 2016 2017 2018
 do
     python -u postprocessing.py --year $year --data-dir "$data_dir" --templates \
     --bdt-preds-dir "$data_dir/24_03_07_new_samples_max_depth_5/inferences" \

src/HHbbVV/postprocessing/postprocessing.py

@@ -424,7 +424,7 @@ def main(args):
     derive_variables(
         events_dict,
         bb_masks,
-        nonres_vars=args.vbf or args.control_plots,
+        nonres_vars=args.vbf or (args.control_plots and not args.mass_plots),
         # nonres_vars=args.vbf,
         vbf_vars=args.vbf,
         do_jshifts=args.vbf,  # only need shifts for BDT pre-processing
@@ -437,7 +437,7 @@ def main(args):
     print("\nCutflow", cutflow)
 
     # Load BDT Scores
-    if not args.resonant and not args.vbf:
+    if not args.resonant and not args.vbf and not args.mass_plots:
         print("\nLoading BDT predictions")
         load_bdt_preds(
             events_dict,
@@ -466,10 +466,10 @@ def main(args):
         else:
             p_sig_keys = plot_sig_keys_nonres
             sig_scale_dict = {
-                "HHbbVV": 1e5,
-                "VBFHHbbVV": 2e5,
-                "qqHH_CV_1_C2V_0_kl_1_HHbbVV": 2e3,
-                "qqHH_CV_1_C2V_2_kl_1_HHbbVV": 2e3,
+                "HHbbVV": 3e5,
+                "VBFHHbbVV": 3e6,
+                "qqHH_CV_1_C2V_0_kl_1_HHbbVV": 6e3,
+                "qqHH_CV_1_C2V_2_kl_1_HHbbVV": 6e3,
             }
 
         control_plots(
@@ -483,6 +483,7 @@ def main(args):
             sig_scale_dict=sig_scale_dict,
             # sig_splits=sig_splits,
             HEM2d=args.HEM2d,
+            plot_ratio=not args.mass_plots,  # don't need data/MC ratio for mreg vs msd comparison
             same_ylim=args.mass_plots,
             show=False,
         )
@@ -1069,6 +1070,9 @@ def apply_trigger_weights(events_dict: dict[str, pd.DataFrame], year: str, cutfl
 
 def qcd_sf(events_dict: dict[str, pd.DataFrame], cutflow: pd.DataFrame):
     """Applies a QCD scale factor."""
+    if qcd_key not in events_dict:
+        return
+
     trig_yields = cutflow.iloc[:, -1]
     non_qcd_bgs_yield = np.sum(
         [
@@ -1440,6 +1444,7 @@ def control_plots(
     sig_scale_dict: dict[str, float] = None,
     combine_pdf: bool = True,
     HEM2d: bool = False,
+    plot_ratio: bool = True,
     plot_significance: bool = False,
     same_ylim: bool = False,
     show: bool = False,
@@ -1453,6 +1458,7 @@ def control_plots(
           {var1: ([num bins, min, max], label), var2...}.
         sig_splits: split up signals into different plots (in case there are too many for one)
         HEM2d: whether to plot 2D hists of FatJet phi vs eta for bb and VV jets as a check for HEM cleaning.
+        plot_ratio: whether to plot the data/MC ratio.
         plot_significance: whether to plot the significance as well as the ratio plot.
         same_ylim: whether to use the same y-axis limits for all plots.
         log: True or False if plot on log scale or not - or "both" if both.
@@ -1516,6 +1522,7 @@ def control_plots(
                     show=show,
                     log=log,
                     ylim=ylim if not log else 1e15,
+                    plot_ratio=plot_ratio,
                 )
                 merger_control_plots.append(name)
 

src/HHbbVV/postprocessing/templates/24Mar18LPSFCheck/cutflows/2016/pass_cutflow.csv

@@ -0,0 +1,2 @@
+,Pre-selection,LP SF,bbFatJetPt >= 300,VVFatJetPt >= 300,BDTScore >= 0.998,bbFatJetParticleNetMD_Txbb >= 0.9735,0 ≤ nGoodElectronsHbb < 0.9,0 ≤ nGoodMuonsHbb < 0.9
+qqHH_CV_1_C2V_1_kl_2_HHbbVV,0.0173959132140592,0.0173959132140592,0.017299990201845,0.017214419089199705,0.00042044328198753,0.0002600195116698255,0.0002558856684327143,0.0002558856684327143

src/HHbbVV/postprocessing/templates/24Mar18LPSFCheck/cutflows/2016APV/pass_cutflow.csv

@@ -0,0 +1,2 @@
+,Pre-selection,LP SF,bbFatJetPt >= 300,VVFatJetPt >= 300,BDTScore >= 0.998,bbFatJetParticleNetMD_Txbb >= 0.9737,0 ≤ nGoodElectronsHbb < 0.9,0 ≤ nGoodMuonsHbb < 0.9
+qqHH_CV_1_C2V_1_kl_2_HHbbVV,0.021501518954477026,0.021501518954477026,0.021387774292197015,0.02128691499756289,0.0005347383241229343,0.0003420753830479197,0.0003420753830479197,0.0003420753830479197

src/HHbbVV/postprocessing/templates/24Mar18LPSFCheck/cutflows/2017/pass_cutflow.csv

@@ -0,0 +1,2 @@
+,Pre-selection,LP SF,bbFatJetPt >= 300,VVFatJetPt >= 300,BDTScore >= 0.998,bbFatJetParticleNetMD_Txbb >= 0.9714,0 ≤ nGoodElectronsHbb < 0.9,0 ≤ nGoodMuonsHbb < 0.9
+qqHH_CV_1_C2V_1_kl_2_HHbbVV,0.03499528349928152,0.03499528349928152,0.034883938631222695,0.034828787476897716,0.0011217067019144162,0.0007799503782162023,0.0007798089829283552,0.0007711790124800327

src/HHbbVV/postprocessing/templates/24Mar18LPSFCheck/cutflows/2018/pass_cutflow.csv

@@ -0,0 +1,2 @@
+,Pre-selection,LP SF,bbFatJetPt >= 300,VVFatJetPt >= 300,BDTScore >= 0.998,bbFatJetParticleNetMD_Txbb >= 0.9734,0 ≤ nGoodElectronsHbb < 0.9,0 ≤ nGoodMuonsHbb < 0.9
+qqHH_CV_1_C2V_1_kl_2_HHbbVV,0.04105557177629756,0.04105557177629756,0.04090476204464955,0.0407915993104881,0.0015553429231248339,0.0011102815028561314,0.0011071970472941277,0.0011060993774949018

src/HHbbVV/postprocessing/templates/24Mar18LPSFCheck/lpsfs.csv

@@ -0,0 +1,2 @@
+,SF,sj_matching_unc,sj_pt_unc,stat_unc,syst_unc
+qqHH_CV_1_C2V_1_kl_2_HHbbVV,0.86 ± 0.59,0.14285714285714285,0.010252321392099382,0.12946452895544497,0.6522582240401613

src/HHbbVV/postprocessing/templates/24Mar18LPSFCheck/systematics.json

@@ -0,0 +1 @@
+{"2016APV": {}, "qqHH_CV_1_C2V_1_kl_2_HHbbVV": {"lp_sf": 0.8604309258381978, "lp_sf_unc": 0.6802316726269908, "lp_sf_uncs": {"syst_unc": 0.6522582240401613, "stat_unc": 0.12946452895544497, "sj_pt_unc": 0.010252321392099382, "sj_matching_unc": 0.14285714285714285}}}

src/HHbbVV/postprocessing/test_templates/24Mar18LPSFCheck2016/cutflows/2016/pass_cutflow.csv

@@ -0,0 +1,2 @@
+,Pre-selection,LP SF,bbFatJetPt >= 300,VVFatJetPt >= 300,BDTScore >= 0.998,bbFatJetParticleNetMD_Txbb >= 0.9735,0 ≤ nGoodElectronsHbb < 0.9,0 ≤ nGoodMuonsHbb < 0.9
+qqHH_CV_1_C2V_1_kl_2_HHbbVV,0.0173959132140592,0.0173959132140592,0.017299990201845,0.017214419089199705,0.00042044328198753,0.0002600195116698255,0.0002558856684327143,0.0002558856684327143

src/HHbbVV/postprocessing/test_templates/24Mar18LPSFCheck2016/lpsfs.csv

@@ -0,0 +1,2 @@
+,SF,sj_matching_unc,sj_pt_unc,stat_unc,syst_unc
+qqHH_CV_1_C2V_1_kl_2_HHbbVV,1.24 ± 0.67,0.1399108138238573,0.008987658498025358,0.1001598592479952,0.5155430106030157

src/HHbbVV/postprocessing/test_templates/24Mar18LPSFCheck2016/systematics.json

@@ -0,0 +1 @@
+{"2016": {}, "qqHH_CV_1_C2V_1_kl_2_HHbbVV": {"lp_sf": 1.2411820701630025, "lp_sf_unc": 0.5435737365034577, "lp_sf_uncs": {"syst_unc": 0.5155430106030157, "stat_unc": 0.1001598592479952, "sj_pt_unc": 0.008987658498025358, "sj_matching_unc": 0.1399108138238573}}}

src/HHbbVV/postprocessing/test_templates/24Mar18LPSFCheck2016APV/cutflows/2016APV/pass_cutflow.csv

@@ -0,0 +1,2 @@
+,Pre-selection,LP SF,bbFatJetPt >= 300,VVFatJetPt >= 300,BDTScore >= 0.998,bbFatJetParticleNetMD_Txbb >= 0.9737,0 ≤ nGoodElectronsHbb < 0.9,0 ≤ nGoodMuonsHbb < 0.9
+qqHH_CV_1_C2V_1_kl_2_HHbbVV,0.021501518954477026,0.021501518954477026,0.021387774292197015,0.02128691499756289,0.0005347383241229343,0.0003420753830479197,0.0003420753830479197,0.0003420753830479197

src/HHbbVV/postprocessing/test_templates/24Mar18LPSFCheck2016APV/lpsfs.csv

@@ -0,0 +1,2 @@
+,SF,sj_matching_unc,sj_pt_unc,stat_unc,syst_unc
+qqHH_CV_1_C2V_1_kl_2_HHbbVV,0.86 ± 0.59,0.14285714285714285,0.010252321392099382,0.12946452895544497,0.6522582240401613

src/HHbbVV/postprocessing/test_templates/24Mar18LPSFCheck2016APV/systematics.json

@@ -0,0 +1 @@
+{"2016APV": {}, "qqHH_CV_1_C2V_1_kl_2_HHbbVV": {"lp_sf": 0.8604309258381978, "lp_sf_unc": 0.6802316726269908, "lp_sf_uncs": {"syst_unc": 0.6522582240401613, "stat_unc": 0.12946452895544497, "sj_pt_unc": 0.010252321392099382, "sj_matching_unc": 0.14285714285714285}}}

src/HHbbVV/postprocessing/test_templates/24Mar18LPSFCheck2017/cutflows/2017/pass_cutflow.csv

@@ -0,0 +1,2 @@
+,Pre-selection,LP SF,bbFatJetPt >= 300,VVFatJetPt >= 300,BDTScore >= 0.998,bbFatJetParticleNetMD_Txbb >= 0.9714,0 ≤ nGoodElectronsHbb < 0.9,0 ≤ nGoodMuonsHbb < 0.9
+qqHH_CV_1_C2V_1_kl_2_HHbbVV,0.03499528349928152,0.03499528349928152,0.034883938631222695,0.034828787476897716,0.0011217067019144162,0.0007799503782162023,0.0007798089829283552,0.0007711790124800327

src/HHbbVV/postprocessing/test_templates/24Mar18LPSFCheck2017/lpsfs.csv

@@ -0,0 +1,2 @@
+,SF,sj_matching_unc,sj_pt_unc,stat_unc,syst_unc
+qqHH_CV_1_C2V_1_kl_2_HHbbVV,1.13 ± 0.57,0.12784248474764282,0.007395479581722196,0.06913171449227407,0.48201850196491147

src/HHbbVV/postprocessing/test_templates/24Mar18LPSFCheck2017/systematics.json

@@ -0,0 +1 @@
+{"2017": {}, "qqHH_CV_1_C2V_1_kl_2_HHbbVV": {"lp_sf": 1.1269971767791451, "lp_sf_unc": 0.5035071242890548, "lp_sf_uncs": {"syst_unc": 0.48201850196491147, "stat_unc": 0.06913171449227407, "sj_pt_unc": 0.007395479581722196, "sj_matching_unc": 0.12784248474764282}}}

src/HHbbVV/postprocessing/test_templates/24Mar18LPSFCheck2018/cutflows/2018/pass_cutflow.csv

@@ -0,0 +1,2 @@
+,Pre-selection,LP SF,bbFatJetPt >= 300,VVFatJetPt >= 300,BDTScore >= 0.998,bbFatJetParticleNetMD_Txbb >= 0.9734,0 ≤ nGoodElectronsHbb < 0.9,0 ≤ nGoodMuonsHbb < 0.9
+qqHH_CV_1_C2V_1_kl_2_HHbbVV,0.04105557177629756,0.04105557177629756,0.04090476204464955,0.0407915993104881,0.0015553429231248339,0.0011102815028561314,0.0011071970472941277,0.0011060993774949018

src/HHbbVV/postprocessing/test_templates/24Mar18LPSFCheck2018/lpsfs.csv

@@ -0,0 +1,2 @@
+,SF,sj_matching_unc,sj_pt_unc,stat_unc,syst_unc
+qqHH_CV_1_C2V_1_kl_2_HHbbVV,1.11 ± 0.64,0.11687725631768953,0.004125334945509687,0.07621464039715414,0.5657408058437927

src/HHbbVV/postprocessing/test_templates/24Mar18LPSFCheck2018/systematics.json

@@ -0,0 +1 @@
+{"2018": {}, "qqHH_CV_1_C2V_1_kl_2_HHbbVV": {"lp_sf": 1.1068573499629724, "lp_sf_unc": 0.5827080248635804, "lp_sf_uncs": {"syst_unc": 0.5657408058437927, "stat_unc": 0.07621464039715414, "sj_pt_unc": 0.004125334945509687, "sj_matching_unc": 0.11687725631768953}}}