Flexible sample database location, column definitions, and Run 3-specific weights and selections

configs/nmssm_boosted/2018/preselection_et.yaml

@@ -69,6 +69,94 @@ processes:
             - "EGamma_Run2018C-UL2018"
             - "EGamma_Run2018D-UL2018"
 
+column_definitions:
+    njets:
+        expression: njets_boosted
+        allow_redefine: True
+    nbtag:
+        expression: nbtag_boosted
+        allow_redefine: True
+    metphi:
+        expression: metphi_boosted
+        allow_redefine: True
+    met:
+        expression: met_boosted
+        allow_redefine: True
+    pt_1:
+        expression: boosted_pt_1
+        allow_redefine: True
+    q_1:
+        expression: boosted_q_1
+        allow_redefine: True
+    pt_2:
+        expression: boosted_pt_2
+        allow_redefine: True
+    q_2:
+        expression: boosted_q_2
+        allow_redefine: True
+    mt_1:
+        expression: boosted_mt_1
+        allow_redefine: True
+    iso_1:
+        expression: boosted_iso_1
+        allow_redefine: True
+    mass_2:
+        expression: boosted_mass_2
+        allow_redefine: True
+    tau_decaymode_2:
+        expression: boosted_tau_decaymode_2
+        allow_redefine: True
+    deltaR_ditaupair:
+        expression: boosted_deltaR_ditaupair
+        allow_redefine: True
+    m_vis:
+        expression: boosted_m_vis
+        allow_redefine: True
+    fj_Xbb_pt:
+        expression: fj_Xbb_pt_boosted
+        allow_redefine: True
+    fj_Xbb_eta:
+        expression: fj_Xbb_eta_boosted
+        allow_redefine: True
+    fj_Xbb_particleNet_XbbvsQCD:
+        expression: fj_Xbb_particleNet_XbbvsQCD_boosted
+        allow_redefine: True
+    bpair_pt_1:
+        expression: bpair_pt_1_boosted
+        allow_redefine: True
+    bpair_pt_2:
+        expression: bpair_pt_2_boosted
+        allow_redefine: True
+    bpair_btag_value_2:
+        expression: bpair_btag_value_2_boosted
+        allow_redefine: True
+    bpair_eta_2:
+        expression: bpair_eta_2_boosted
+        allow_redefine: True
+    extraelec_veto:
+        expression: extraelec_veto_boosted
+        allow_redefine: True
+    gen_match_1:
+        expression: boosted_gen_match_1
+        allow_redefine: True
+        exclude_processes:
+        - data
+    gen_match_2:
+        expression: boosted_gen_match_2
+        allow_redefine: True
+        exclude_processes:
+        - data
+    btag_weight:
+        expression: btag_weight_boosted
+        allow_redefine: True
+        exclude_processes:
+        - data
+    pNet_Xbb_weight:
+        expression: pNet_Xbb_weight_boosted
+        allow_redefine: True
+        exclude_processes:
+        - data
+
 event_selection:
     # lep_pt: "boosted_pt_1 > 120"
     had_tau_pt: "boosted_pt_2 > 40"
@@ -130,4 +218,4 @@ output_features:
     - "bpair_eta_2"
     - "met"
     - "mass_2"
-    - "tau_decaymode_2"
+    - "tau_decaymode_2"

configs/nmssm_boosted/2018/preselection_mt.yaml

@@ -69,6 +69,89 @@ processes:
             - "SingleMuon_Run2018C-UL2018_GT36"
             - "SingleMuon_Run2018D-UL2018_GT36"
 
+column_definitions:
+    njets:
+        expression: njets_boosted
+        allow_redefine: True
+    nbtag:
+        expression: nbtag_boosted
+        allow_redefine: True
+    metphi:
+        expression: metphi_boosted
+        allow_redefine: True
+    met:
+        expression: met_boosted
+        allow_redefine: True
+    pt_1:
+        expression: boosted_pt_1
+        allow_redefine: True
+    q_1:
+        expression: boosted_q_1
+        allow_redefine: True
+    pt_2:
+        expression: boosted_pt_2
+        allow_redefine: True
+    q_2:
+        expression: boosted_q_2
+        allow_redefine: True
+    mt_1:
+        expression: boosted_mt_1
+        allow_redefine: True
+    iso_1:
+        expression: boosted_iso_1
+        allow_redefine: True
+    mass_2:
+        expression: boosted_mass_2
+        allow_redefine: True
+    tau_decaymode_2:
+        expression: boosted_tau_decaymode_2
+        allow_redefine: True
+    deltaR_ditaupair:
+        expression: boosted_deltaR_ditaupair
+        allow_redefine: True
+    m_vis:
+        expression: boosted_m_vis
+        allow_redefine: True
+    fj_Xbb_pt:
+        expression: fj_Xbb_pt_boosted
+        allow_redefine: True
+    fj_Xbb_eta:
+        expression: fj_Xbb_eta_boosted
+        allow_redefine: True
+    fj_Xbb_particleNet_XbbvsQCD:
+        expression: fj_Xbb_particleNet_XbbvsQCD_boosted
+        allow_redefine: True
+    bpair_pt_1:
+        expression: bpair_pt_1_boosted
+        allow_redefine: True
+    bpair_pt_2:
+        expression: bpair_pt_2_boosted
+        allow_redefine: True
+    bpair_btag_value_2:
+        expression: bpair_btag_value_2_boosted
+        allow_redefine: True
+    bpair_eta_2:
+        expression: bpair_eta_2_boosted
+        allow_redefine: True
+    extramuon_veto:
+        expression: extramuon_veto_boosted
+        allow_redefine: True
+    gen_match_2:
+        expression: boosted_gen_match_2
+        allow_redefine: True
+        exclude_processes:
+        - data
+    btag_weight:
+        expression: btag_weight_boosted
+        allow_redefine: True
+        exclude_processes:
+        - data
+    pNet_Xbb_weight:
+        expression: pNet_Xbb_weight_boosted
+        allow_redefine: True
+        exclude_processes:
+        - data
+
 event_selection:
     # lep_pt: "boosted_pt_1 > 55"
     had_tau_pt: "boosted_pt_2 > 40"
@@ -130,4 +213,4 @@ output_features:
     - "bpair_eta_2"
     - "met"
     - "mass_2"
-    - "tau_decaymode_2"
+    - "tau_decaymode_2"

configs/nmssm_boosted/2018/preselection_tt.yaml

@@ -69,6 +69,95 @@ processes:
             - "JetHT_Run2018C-UL2018"
             - "JetHT_Run2018D-UL2018"
 
+column_definitions:
+    njets:
+        expression: njets_boosted
+        allow_redefine: True
+    nbtag:
+        expression: nbtag_boosted
+        allow_redefine: True
+    metphi:
+        expression: metphi_boosted
+        allow_redefine: True
+    met:
+        expression: met_boosted
+        allow_redefine: True
+    pt_1:
+        expression: boosted_pt_1
+        allow_redefine: True
+    q_1:
+        expression: boosted_q_1
+        allow_redefine: True
+    pt_2:
+        expression: boosted_pt_2
+        allow_redefine: True
+    q_2:
+        expression: boosted_q_2
+        allow_redefine: True
+    mt_1:
+        expression: boosted_mt_1
+        allow_redefine: True
+    iso_1:
+        expression: boosted_iso_1
+        allow_redefine: True
+    mass_1:
+        expression: boosted_mass_1
+        allow_redefine: True
+    mass_2:
+        expression: boosted_mass_2
+        allow_redefine: True
+    tau_decaymode_1:
+        expression: boosted_tau_decaymode_1
+        allow_redefine: True
+    tau_decaymode_2:
+        expression: boosted_tau_decaymode_2
+        allow_redefine: True
+    deltaR_ditaupair:
+        expression: boosted_deltaR_ditaupair
+        allow_redefine: True
+    m_vis:
+        expression: boosted_m_vis
+        allow_redefine: True
+    fj_Xbb_pt:
+        expression: fj_Xbb_pt_boosted
+        allow_redefine: True
+    fj_Xbb_eta:
+        expression: fj_Xbb_eta_boosted
+        allow_redefine: True
+    fj_Xbb_particleNet_XbbvsQCD:
+        expression: fj_Xbb_particleNet_XbbvsQCD_boosted
+        allow_redefine: True
+    bpair_pt_1:
+        expression: bpair_pt_1_boosted
+        allow_redefine: True
+    bpair_pt_2:
+        expression: bpair_pt_2_boosted
+        allow_redefine: True
+    bpair_btag_value_2:
+        expression: bpair_btag_value_2_boosted
+        allow_redefine: True
+    bpair_eta_2:
+        expression: bpair_eta_2_boosted
+        allow_redefine: True
+    extramuon_veto:
+        expression: extramuon_veto_boosted
+        allow_redefine: True
+    gen_match_2:
+        expression: boosted_gen_match_2
+        allow_redefine: True
+        exclude_processes:
+        - data
+    btag_weight:
+        expression: btag_weight_boosted
+        allow_redefine: True
+        exclude_processes:
+        - data
+    pNet_Xbb_weight:
+        expression: pNet_Xbb_weight_boosted
+        allow_redefine: True
+        exclude_processes:
+        - data
+
 event_selection:
     # met: "(met_boosted > 120)"
     had_tau_pt: "(boosted_pt_1 > 40) && (boosted_pt_2 > 40)"
@@ -132,4 +221,4 @@ output_features:
     - "mass_1"
     - "tau_decaymode_1"
     - "mass_2"
-    - "tau_decaymode_2"
+    - "tau_decaymode_2"

docs/preselection.md

@@ -8,6 +8,7 @@ The preselection config has the following parameters:
   ---|---|---
   `channel` | `string` | tau pair decay channels ("et", "mt", "tt")
   `processes` | `dict` | process parameters are explained below
+  `column_definitions` | `dict` | in this section, new columns can be defined based on a given `ROOT` expression. <br>The keys of the dictionary correspond to the name of the defined column. The values are dictionaries itself, with the `expression` key defining the `ROOT` expression for defining the column. Optional entries `processes` and `exclude_processes` allow to target specific processes, the entry `allow_redefine` can be used to enable the use of the `ROOT.RDataFrame.Redefine` function for overwriting already existing columns. For a more detailed description, see below.
   `event_selection` | `dict` | with this parameter all selections that should be applied are defined. <br>This is basically a dictionary of cuts where the key is the name of a cut and the value is the cut itself as a string e.g. `had_tau_pt: "pt_2 > 30"`. The name of a cut is not really important, it is only used as an output information in the terminal. A cut can only use variables which are in the ntuples.
   `mc_weights` | `dict` | weight parameter are defined below
   `emb_weights` | `dict` | all weights that should be applied for embedded samples are defined. <br>Like for `event_selection` a weight can directly be specified and is then applied to all samples the same way e.g. `single_trigger: "trg_wgt_single_mu24ormu27"`
@@ -31,6 +32,42 @@ The `tau_gen_modes` have following modes:
   `L` | `string` | lepton misidentified as a tau
   `all` | `string` | if no split should be performed
 
+In `column_definitions`, new columns can be added to the output `ntuples` by
+using `ROOT` expression. An example entry could look like this:
+
+```yaml
+column_definitions:
+    nbtag:
+        expression: n_bjets
+        processes:
+        - ttbar
+        - DY
+    btag_weight:
+        expression: id_wgt_bjet_pnet_shape
+        exclude_processes:
+        - data
+        allow_redefine: True
+    jj_deltaR:
+        expression: ROOT::VecOps::DeltaR(jeta_1, jeta_2, jphi_1, jphi_2)
+```
+
+The key `expression` is required and can contain any valid `ROOT` expression.
+
+The entry `exclude_processes` is optional. Column definitions are performed for
+all processes except the ones given in this list. The entry `processes` is also
+optional. The column definition is performed only for processes in this list.
+The lists `processes` and `exclude_processes` can contain the names from the
+`processes` section of this configuration. By default, the new columns are
+defined for all processes. To write the new columns to the output file, you have
+to explicitly add the columns to the `output_features` list. Note that you can
+only set `processes` or `exclude_processes` for a column, but not both at the
+same time.
+
+If the key `allow_redefine` is set to `True`, the `ROOT.RDataFrame.Redefine`
+function is used if a column with the same name has been found in the
+`RDataFrame`. The values in this column are then overwritten by the expression
+given for the new column.
+
 In `mc_weights` all weights that should be applied for simulated samples are defined. <br>
 There are two types of weights.
 
@@ -53,4 +90,4 @@ python preselection.py --config-file configs/PATH/CONFIG.yaml
 Further there are additional optional parameters: 
 
 1. `--nthreads=SOME_INTEGER` to define the number of threads for the multiprocessing pool to run the sample processing in parallel. Default value is 8 (this should normally cover running all of the samples in parallel).
-2.  `--ncores=SOME_INTEGER` to define the number of cores that should be used for each pool thread to speed up the ROOT dataframe calculation. Default value is 2.
+2.  `--ncores=SOME_INTEGER` to define the number of cores that should be used for each pool thread to speed up the ROOT dataframe calculation. Default value is 2.