Reimplement ATLAS Z0 7 TeV Low mass

nnpdf_data/nnpdf_data/commondata/ATLAS_Z0_7TEV_LOMASS/filter.py

@@ -0,0 +1,86 @@
+"""
+filter.py module for ATLAS_Z0_7TEV_LOMASS dataset
+
+When running `python filter.py` the relevant uncertainties , data and kinematics yaml
+file will be created in the `nnpdf_data/commondata/ATLAS_Z0_7TEV_LOMASS` directory.
+"""
+
+import yaml
+from filter_utils import get_kinematics, get_data_values, get_systematics
+
+
+def filter_ATLAS_Z0_7TEV_LOMASS_data_kinetic():
+    """
+    This function writes the central values and kinematics to yaml files.
+    """
+
+    kin = get_kinematics()
+    central_values = list(get_data_values())
+
+    data_central_yaml = {"data_central": central_values}
+
+    kinematics_yaml = {"bins": kin}
+
+    # write central values and kinematics to yaml file
+    with open("data.yaml", "w") as file:
+        yaml.dump(data_central_yaml, file, sort_keys=False)
+
+    with open("kinematics.yaml", "w") as file:
+        yaml.dump(kinematics_yaml, file, sort_keys=False)
+
+
+def filter_ATLAS_Z0_7TEV_LOMASS_systematics():
+    """
+    This function writes the systematics to a yaml file.
+    """
+
+    with open("metadata.yaml", "r") as file:
+        metadata = yaml.safe_load(file)
+
+    systematics = get_systematics()
+
+    # error definition
+    error_definitions = {}
+    errors = []
+
+    for sys in systematics:
+        if sys[0]['name'] == 'stat':
+            error_definitions[sys[0]['name']] = {
+                "description": f"{sys[0]['name']}",
+                "treatment": "ADD",
+                "type": "UNCORR",
+            }
+
+        elif (sys[0]['name'] == 'sys_res') or (sys[0]['name'] == 'sys_MC'):
+            error_definitions[sys[0]['name']] = {
+                "description": f"{sys[0]['name']}",
+                "treatment": "MULT",
+                "type": "UNCORR",
+            }
+
+        else:
+            error_definitions[sys[0]['name']] = {
+                "description": f"{sys[0]['name']}",
+                "treatment": "MULT",
+                "type": "CORR",
+            }
+
+    #
+    for i in range(metadata['implemented_observables'][0]['ndata']):
+        error_value = {}
+
+        for sys in systematics:
+            error_value[sys[0]['name']] = float(sys[0]['values'][i])
+
+        errors.append(error_value)
+
+    uncertainties_yaml = {"definitions": error_definitions, "bins": errors}
+
+    # write uncertainties
+    with open(f"uncertainties.yaml", 'w') as file:
+        yaml.dump(uncertainties_yaml, file, sort_keys=False)
+
+
+if __name__ == "__main__":
+    filter_ATLAS_Z0_7TEV_LOMASS_data_kinetic()
+    filter_ATLAS_Z0_7TEV_LOMASS_systematics()

nnpdf_data/nnpdf_data/commondata/ATLAS_Z0_7TEV_LOMASS/filter_utils.py

@@ -0,0 +1,81 @@
+"""
+This module contains helper functions that are used to extract the uncertainties, kinematics and data values 
+from the rawdata files.
+"""
+
+import yaml
+import pandas as pd
+import numpy as np
+
+
+def get_kinematics():
+    """
+    returns the kinematics in the form of a list of dictionaries.
+    """
+    kin = []
+
+    hepdata_table = f"rawdata/HEPData-ins1288706-v1-Table_6.yaml"
+
+    with open(hepdata_table, 'r') as file:
+        input = yaml.safe_load(file)
+
+    for i, M in enumerate(input["independent_variables"][0]['values']):
+
+        kin_value = {
+            'm_ll': {'min': M['low'], 'mid': (0.5 * (M['low'] + M['high'])), 'max': M['high']},
+            'sqrts': {'min': None, 'mid': 7000.0, 'max': None},
+        }
+
+        kin.append(kin_value)
+
+    return kin
+
+
+def get_data_values():
+    """
+    returns the central data values in the form of a list.
+    """
+
+    data_central = []
+
+    hepdata_table = f"rawdata/HEPData-ins1288706-v1-Table_6.yaml"
+
+    with open(hepdata_table, 'r') as file:
+        input = yaml.safe_load(file)
+
+    values = input['dependent_variables'][0]['values']
+
+    for value in values:
+        # store data central and convert the units
+        data_central.append(value['value'] * 1000)
+
+    return data_central
+
+
+def get_systematics_dataframe():
+    """
+    returns the absolute systematic uncertainties in the form of a pandas dataframe.
+    """
+    sys_rawdata_path = "rawdata/ATLASLOMASSDY11EXT.csv"
+
+    df = pd.read_csv(sys_rawdata_path)
+    data_central = np.array(get_data_values())
+
+    # convert (MULT) percentage unc to absolute unc
+    abs_unc_df = (df.T[3:] * data_central).T / 100
+
+    return abs_unc_df
+
+
+def get_systematics():
+    """ """
+    abs_unc_df = get_systematics_dataframe()
+
+    uncertainties = []
+
+    for i, unc_dp in enumerate(abs_unc_df.values.T):
+        name = f"{abs_unc_df.columns[i]}"
+        values = [unc_dp[j] for j in range(len(unc_dp))]
+        uncertainties.append([{"name": name, "values": values}])
+
+    return uncertainties

nnpdf_data/nnpdf_data/commondata/ATLAS_Z0_7TEV_LOMASS/kinematics.yaml

@@ -0,0 +1,49 @@
+bins:
+- m_ll:
+    min: 12.0
+    mid: 14.5
+    max: 17.0
+  sqrts:
+    min: null
+    mid: 7000.0
+    max: null
+- m_ll:
+    min: 17.0
+    mid: 19.5
+    max: 22.0
+  sqrts:
+    min: null
+    mid: 7000.0
+    max: null
+- m_ll:
+    min: 22.0
+    mid: 25.0
+    max: 28.0
+  sqrts:
+    min: null
+    mid: 7000.0
+    max: null
+- m_ll:
+    min: 28.0
+    mid: 32.0
+    max: 36.0
+  sqrts:
+    min: null
+    mid: 7000.0
+    max: null
+- m_ll:
+    min: 36.0
+    mid: 41.0
+    max: 46.0
+  sqrts:
+    min: null
+    mid: 7000.0
+    max: null
+- m_ll:
+    min: 46.0
+    mid: 56.0
+    max: 66.0
+  sqrts:
+    min: null
+    mid: 7000.0
+    max: null

nnpdf_data/nnpdf_data/commondata/ATLAS_Z0_7TEV_LOMASS/metadata.yaml

@@ -1,14 +1,14 @@
 setname: ATLAS_Z0_7TEV_LOMASS
 version: 1
-version_comment: Port of old commondata
+version_comment: "First implementation"
 nnpdf_metadata:
   nnpdf31_process: DY NC
   experiment: ATLAS
 arXiv:
   url: https://arxiv.org/abs/1404.1212
   journal: JHEP 06 (2014) 112
 iNSPIRE:
-  url: ''
+  url: 'https://inspirehep.net/literature/1288706'
 hepdata:
   url: 10.17182/hepdata.64183.v1/t6
   version: -1
@@ -18,43 +18,39 @@ implemented_observables:
     description: Drell-Yan Mass Distribution
     label: ATLAS low-mass DY 2011
     units: ''
-  process_type: EWK_MLL
-  tables: []
-  npoints: []
+  process_type: DY_MLL
+  tables: [6]
+  npoints: [6]
   ndata: 6
   plotting:
-    kinematics_override: ewk_mll_sqrt_scale
+    kinematics_override: identity
     dataset_label: ATLAS low-mass DY 2011
     y_label: $d\sigma_{Z/\gamma^{*}}/dM_{ll}$ (fb)
-    plot_x: k2
+    plot_x: m_ll
   kinematic_coverage:
-  - k1
-  - k2
-  - k3
+  - m_ll
+  - sqrts
   kinematics:
     variables:
-      k1:
-        description: Variable k1
-        label: k1
-        units: ''
-      k2:
-        description: Variable k2
-        label: k2
-        units: ''
-      k3:
-        description: Variable k3
-        label: k3
-        units: ''
-    file: kinematics_M.yaml
+      m_ll:
+        description: squared mass
+        label: m_ll
+        units: 'GeV'
+      sqrts:
+        description: center of mass energy
+        label: sqrts
+        units: 'GeV'
+    file: kinematics.yaml
   theory:
     conversion_factor: 1000.0
     operation: 'null'
     FK_tables:
     - - ATLAS_DY_7TEV_LOMASS_EXT
-  data_uncertainties: []
+  data_uncertainties:
+    - uncertainties.yaml
   variants:
     legacy:
       data_uncertainties:
       - uncertainties_legacy_M.yaml
-  data_central: data_legacy_M.yaml
+  data_central: data.yaml
   ported_from: ATLASLOMASSDY11EXT

nnpdf_data/nnpdf_data/commondata/ATLAS_Z0_7TEV_LOMASS/rawdata/ATLASLOMASSDY11EXT.csv

@@ -0,0 +1,7 @@
+Mlow, Mhigh,y,stat,sys_reco,sys_trig,sys_iso,sys_mjet,sys_pTsc,sys_res,sys_MC,lumi
+12.0,17,12.41,4.2,2.5,4.0,11.3,-3.0,-0.2,0.5,0.6,3.5
+17.0,22.0,22.57,3.1,1.4,3.7,11.3,-2.8,0.1,0.3,0.3,3.5
+22.0,28.0,14.64,3.3,0.9,3.6,8.5,-1.8,0.0,0.1,0.4,3.5
+28.0,36.0,6.73,4.0,0.7,3.6,6.2,-1.6,-0.1,0.2,0.4,3.5
+36.0,46.0,2.81,5.2,0.7,3.6,4.2,-1.3,-0.1,0.1,0.5,3.5
+46.0,66.0,1.27,4.7,0.6,3.6,3.6,-0.7,0.0,0.1,0.5,3.5

nnpdf_data/nnpdf_data/commondata/ATLAS_Z0_7TEV_LOMASS/rawdata/ATLASLOMASSDY11EXT.data

@@ -0,0 +1,16 @@
+ Mlow   Mhigh     y       stat+     stat-    sys_reco   sys_trig   sys_iso   sys_mjet   sys_pTsc   sys_res   sys_MC    Lumi 
+----------------------------------------------------------------------------------------------------------------------------- 
+ 12.0	17.0	12.41	+0.52122  -0.52122     2.5         4.0      11.3      -3.0       -0.2        0.5      0.6      3.5
+ 17.0	22.0	22.57	+0.69967  -0.69967     1.4         3.7      11.3      -2.8	  0.1	     0.3      0.3      3.5
+ 22.0	28.0	14.64	+0.48312  -0.48312     0.9         3.6       8.5      -1.8	  0.0	     0.1      0.4      3.5
+ 28.0	36.0	 6.73	+0.26920  -0.26920     0.7         3.6       6.2      -1.6	 -0.1	     0.2      0.4      3.5
+ 36.0	46.0	 2.81	+0.14612  -0.14612     0.7         3.6	     4.2      -1.3	 -0.1	     0.1      0.5      3.5
+ 46.0	66.0	 1.27	+0.05969  -0.05969     0.6         3.6	     3.6      -0.7	 -0.0	     0.1      0.5      3.5
+
+
+Path: /HepData/8644/d6-x1-y1
+Born-level fiducial differential cross section dsigma/dm_ll [pb/GeV], with statistical, systematic and Lumi uncertainties 
+Location: Table 6-7, Page 22 of preprint
+RE : P P --> MU+ MU- X
+SQRT(S) : 7000.0 GeV
+