Change BDT cuts from skimming pT binning to analysis pT binning

machine_learning_hep/data/data_run3/database_ml_parameters_D0Jet_pp.yml

@@ -30,8 +30,8 @@ D0jet_pp:
   # obsolete, to be removed
   # var_cand: fCandidateSelFlag
   # # var_swap: fIsCandidateSwapped
-  # bitmap_sel:
-  #   var_name: fFlagMcMatchRec
+  bitmap_sel:
+    var_name: fFlagMcMatchRec
   #   var_name_gen: fFlagMcMatchGen
   #   var_name_origgen: fOriginMcGen
   #   var_name_origrec: fOriginMcRec
@@ -785,7 +785,7 @@ D0jet_pp:
 
       # Additional cuts applied before mass histogram is filled
       use_cuts: True  # systematics
-      cuts: ["mlBkgScore < 0.02", "mlBkgScore < 0.04", "mlBkgScore < 0.06", "mlBkgScore < .08", "mlBkgScore < 0.14", "mlBkgScore < 0.26", "mlBkgScore < 0.46", "mlBkgScore < 0.52"] # (sel_skim_binmin bins) systematics FIXME: Update for new model.
+      cuts: ["mlBkgScore < 0.02", "mlBkgScore < 0.02", "mlBkgScore < 0.04", "mlBkgScore < 0.06", "mlBkgScore < .08", "mlBkgScore < 0.14", "mlBkgScore < 0.14", "mlBkgScore < 0.26", "mlBkgScore < 0.26", "mlBkgScore < 0.46", "mlBkgScore < 0.52", "mlBkgScore < 0.52"]  # (sel_an_binmin bins) systematics FIXME: Update for new model.
 
   systematics:  # used in machine_learning_hep/analysis/systematics.py
     probvariation:

machine_learning_hep/data/data_run3/database_ml_parameters_LcJet_pp.yml

@@ -570,7 +570,7 @@ LcJet_pp:
 
       # Additional cuts applied before mass histogram is filled
       use_cuts: True
-      cuts: ["mlPromptScore > 0.96", "mlPromptScore > 0.97", "mlPromptScore > 0.9", "mlPromptScore > 0.85", "mlPromptScore > 0.8", "mlPromptScore > 0.6", null]
+      cuts: ["mlPromptScore > 0.97", "mlPromptScore > 0.9", "mlPromptScore > 0.9", "mlPromptScore > 0.85", "mlPromptScore > 0.85", "mlPromptScore > 0.8", "mlPromptScore > 0.8", "mlPromptScore > 0.6", "mlPromptScore > 0.6", null]  # (sel_an_binmin bins) systematics FIXME: Update for new model.
 
   systematics:  # used in machine_learning_hep/analysis/systematics.py
     probvariation:

machine_learning_hep/data/data_run3/database_variations_D0Jet_pp_jet_obs.yml

@@ -919,10 +919,10 @@ categories:
                         jet_obs:
                             use_cuts: [True, True, True, True, True, True, True]
                             cuts:
-                                    - ["mlBkgScore < 0.02", "mlBkgScore < 0.04", "mlBkgScore < 0.06", "mlBkgScore < .08", "mlBkgScore < 0.14", "mlBkgScore < 0.26", "mlBkgScore < 0.46", "mlBkgScore < 0.52"]  # default
-                                    - [null,null,null,null,null,null,null,null]
-                                    - ["mlBkgScore < 0.12", "mlBkgScore < 0.16", "mlBkgScore < 0.2", "mlBkgScore < .25", "mlBkgScore < 0.4", "mlBkgScore < 0.6", "mlBkgScore < 0.8", "mlBkgScore < 0.8"] # loosest
-                                    - ["mlBkgScore < 0.06", "mlBkgScore < 0.08", "mlBkgScore < 0.1", "mlBkgScore < .20", "mlBkgScore < 0.3", "mlBkgScore < 0.6", "mlBkgScore < 0.8", "mlBkgScore < 0.8"] # loose
-                                    - ["mlBkgScore < 0.01", "mlBkgScore < 0.02", "mlBkgScore < 0.03", "mlBkgScore < .04", "mlBkgScore < 0.07", "mlBkgScore < 0.13", "mlBkgScore < 0.23", "mlBkgScore < 0.26"] # tighter 2
-                                    - ["mlBkgScore < 0.005", "mlBkgScore < 0.01", "mlBkgScore < 0.015", "mlBkgScore < .02", "mlBkgScore < 0.035", "mlBkgScore < 0.065", "mlBkgScore < 0.125", "mlBkgScore < 0.13"] # tighter 4
-                                    - ["mlBkgScore < 0.015", "mlBkgScore < 0.03", "mlBkgScore < 0.045", "mlBkgScore < .06", "mlBkgScore < 0.105", "mlBkgScore < 0.13", "mlBkgScore < 0.23", "mlBkgScore < 0.26"] # tight
+                                    - ["mlBkgScore < 0.02", "mlBkgScore < 0.02", "mlBkgScore < 0.04", "mlBkgScore < 0.06", "mlBkgScore < .08", "mlBkgScore < 0.14", "mlBkgScore < 0.14", "mlBkgScore < 0.26", "mlBkgScore < 0.26", "mlBkgScore < 0.46", "mlBkgScore < 0.52", "mlBkgScore < 0.52"]  # default
+                                    - [null,null,null,null,null,null,null,null,null,null,null,null]
+                                    - ["mlBkgScore < 0.12", "mlBkgScore < 0.12", "mlBkgScore < 0.16", "mlBkgScore < 0.2", "mlBkgScore < .25", "mlBkgScore < 0.4", "mlBkgScore < 0.4", "mlBkgScore < 0.6", "mlBkgScore < 0.6", "mlBkgScore < 0.8", "mlBkgScore < 0.8", "mlBkgScore < 0.8"]  # loosest
+                                    - ["mlBkgScore < 0.06", "mlBkgScore < 0.06", "mlBkgScore < 0.08", "mlBkgScore < 0.1", "mlBkgScore < .20", "mlBkgScore < 0.3", "mlBkgScore < 0.3", "mlBkgScore < 0.6", "mlBkgScore < 0.6", "mlBkgScore < 0.8", "mlBkgScore < 0.8", "mlBkgScore < 0.8"]  # loose
+                                    - ["mlBkgScore < 0.01", "mlBkgScore < 0.01", "mlBkgScore < 0.02", "mlBkgScore < 0.03", "mlBkgScore < .04", "mlBkgScore < 0.07", "mlBkgScore < 0.07", "mlBkgScore < 0.13", "mlBkgScore < 0.13", "mlBkgScore < 0.23", "mlBkgScore < 0.26", "mlBkgScore < 0.26"]  # tighter 2
+                                    - ["mlBkgScore < 0.005", "mlBkgScore < 0.005", "mlBkgScore < 0.01", "mlBkgScore < 0.015", "mlBkgScore < .02", "mlBkgScore < 0.035", "mlBkgScore < 0.035", "mlBkgScore < 0.065", "mlBkgScore < 0.065", "mlBkgScore < 0.125", "mlBkgScore < 0.13", "mlBkgScore < 0.13"]  # tighter 4
+                                    - ["mlBkgScore < 0.015", "mlBkgScore < 0.015", "mlBkgScore < 0.03", "mlBkgScore < 0.045", "mlBkgScore < .06", "mlBkgScore < 0.105", "mlBkgScore < 0.105", "mlBkgScore < 0.13", "mlBkgScore < 0.13", "mlBkgScore < 0.23", "mlBkgScore < 0.26", "mlBkgScore < 0.26"]  # tight

machine_learning_hep/data/data_run3/database_variations_LcJet_pp_jet_obs.yml

@@ -451,10 +451,10 @@ categories:
                         jet_obs:
                             use_cuts: [True, True, True, True, True, True, True]
                             cuts:
-                                    - ["mlPromptScore > 0.96", "mlPromptScore > 0.97", "mlPromptScore > 0.9", "mlPromptScore > 0.85", "mlPromptScore > 0.8", "mlPromptScore > 0.6", null]  # default
-                                    - [null,null,null,null,null,null,null]
-                                    - ["mlPromptScore > 0.85", "mlPromptScore > 0.85", "mlPromptScore > 0.6", "mlPromptScore > 0.4", "mlPromptScore > 0.4", "mlPromptScore > 0.15", null] # loosest
-                                    - ["mlPromptScore > 0.9", "mlPromptScore > 0.9", "mlPromptScore > 0.7", "mlPromptScore > 0.6", "mlPromptScore > 0.6", "mlPromptScore > 0.3", null] # loose
-                                    - ["mlPromptScore > 0.98", "mlPromptScore > 0.98", "mlPromptScore > 0.9", "mlPromptScore > 0.85", "mlPromptScore > 0.8", "mlPromptScore > 0.6", null] # tight 2
-                                    - ["mlPromptScore > 0.96", "mlPromptScore > 0.97", "mlPromptScore > 0.9", "mlPromptScore > 0.85", "mlPromptScore > 0.8", "mlPromptScore > 0.6", null] # tight 4
-                                    - ["mlPromptScore > 0.98", "mlPromptScore > 0.98", "mlPromptScore > 0.95", "mlPromptScore > 0.9", "mlPromptScore > 0.9", "mlPromptScore > 0.7", null] # tight
+                                    - ["mlPromptScore > 0.97", "mlPromptScore > 0.9", "mlPromptScore > 0.9", "mlPromptScore > 0.85", "mlPromptScore > 0.85", "mlPromptScore > 0.8", "mlPromptScore > 0.8", "mlPromptScore > 0.6", "mlPromptScore > 0.6", null]  # default
+                                    - [null,null,null,null,null,null,null,null,null,null]
+                                    - ["mlPromptScore > 0.85", "mlPromptScore > 0.6", "mlPromptScore > 0.6", "mlPromptScore > 0.4", "mlPromptScore > 0.4", "mlPromptScore > 0.4", "mlPromptScore > 0.4", "mlPromptScore > 0.15", "mlPromptScore > 0.15", null]  # loosest
+                                    - ["mlPromptScore > 0.9", "mlPromptScore > 0.7", "mlPromptScore > 0.7", "mlPromptScore > 0.6", "mlPromptScore > 0.6", "mlPromptScore > 0.6", "mlPromptScore > 0.6", "mlPromptScore > 0.3", "mlPromptScore > 0.3", null]  # loose
+                                    - ["mlPromptScore > 0.98", "mlPromptScore > 0.9", "mlPromptScore > 0.9", "mlPromptScore > 0.85", "mlPromptScore > 0.85", "mlPromptScore > 0.8", "mlPromptScore > 0.8", "mlPromptScore > 0.6", "mlPromptScore > 0.6", null]  # tight 2
+                                    - ["mlPromptScore > 0.97", "mlPromptScore > 0.9", "mlPromptScore > 0.9", "mlPromptScore > 0.85", "mlPromptScore > 0.85", "mlPromptScore > 0.8", "mlPromptScore > 0.8", "mlPromptScore > 0.6", "mlPromptScore > 0.6", null]  # tight 4
+                                    - ["mlPromptScore > 0.98", "mlPromptScore > 0.95", "mlPromptScore > 0.95", "mlPromptScore > 0.9", "mlPromptScore > 0.9", "mlPromptScore > 0.9", "mlPromptScore > 0.9", "mlPromptScore > 0.7", "mlPromptScore > 0.7", null]  # tight

machine_learning_hep/processer.py

@@ -128,7 +128,7 @@ def __init__(self, case, datap, run_param, mcordata, p_maxfiles, # pylint: disab
 
         #variables name
         self.v_train = datap["variables"]["var_training"]
-        self.v_bitvar = datap["bitmap_sel"]["var_name"]
+        self.v_bitvar = datap["bitmap_sel"]["var_name"]  # used in hadrons
         # self.v_bitvar_gen = datap["bitmap_sel"]["var_name_gen"]
         # self.v_bitvar_origgen = datap["bitmap_sel"]["var_name_origgen"]
         # self.v_bitvar_origrec = datap["bitmap_sel"]["var_name_origrec"]
@@ -564,36 +564,36 @@ def process_mergedec(self):
 
 
     def load_cuts(self):
-        """Load cuts from database
+        """Load custom analysis cuts from the database.
         """
         raw_cuts = self.datap["analysis"][self.typean].get("cuts", None)
         if not raw_cuts:
             print("No custom cuts given, hence not cutting...")
-            self.analysis_cuts = [None] * self.p_nptbins
+            self.analysis_cuts = [None] * self.p_nptfinbins
             return
-        if len(raw_cuts) != self.p_nptbins:
-            print(f"You have {self.p_nptbins} but you passed {len(raw_cuts)} cuts. Exit...")
+        if len(raw_cuts) != self.p_nptfinbins:
+            print(f"You have {self.p_nptfinbins} but you passed {len(raw_cuts)} cuts. Exit...")
             sys.exit(1)
         self.analysis_cuts = deepcopy(raw_cuts)
 
 
-    def apply_cuts_ptbin(self, df_, ipt):
-        """Cut dataframe with cuts for a given skimming pT bin"""
+    def apply_cuts_ptbin(self, df_ipt, ipt):
+        """Cut dataframe with cuts for a given analysis pT bin"""
         if not self.analysis_cuts[ipt]:
-            return df_
-        return df_.query(self.analysis_cuts[ipt])
+            return df_ipt
+        return df_ipt.query(self.analysis_cuts[ipt])
 
 
     def apply_cuts_all_ptbins(self, df_):
         """Apply cuts for all skimming pT bins."""
         if not self.do_custom_analysis_cuts or not any(self.analysis_cuts):
             return df_
 
-        def apply_cut_for_ipt(dff_, ipt: int):
-            df_ipt = seldf_singlevar(dff_, self.v_var_binning, self.lpt_anbinmin[ipt], self.lpt_anbinmax[ipt])
+        def apply_cut_for_ipt(df_full, ipt: int):
+            df_ipt = seldf_singlevar(df_full, self.v_var_binning, self.lpt_finbinmin[ipt], self.lpt_finbinmax[ipt])
             return df_ipt.query(self.analysis_cuts[ipt]) if self.analysis_cuts[ipt] else df_ipt
 
-        return pd.concat(apply_cut_for_ipt(df_, ipt) for ipt in range(self.p_nptbins))
+        return pd.concat(apply_cut_for_ipt(df_, ipt) for ipt in range(self.p_nptfinbins))
 
 
     def process_histomass(self):