Reimplement Atlas Z0 8 TeV lowmass

nnpdf_data/nnpdf_data/commondata/ATLAS_Z0_8TEV_LOWMASS/data_legacy_M-Y.yaml → nnpdf_data/nnpdf_data/commondata/ATLAS_Z0_8TEV_LOWMASS/data.yaml

@@ -1,36 +1,36 @@
 data_central:
 - 184.96
-- 186.96
+- 1.86960000e+02
 - 185.78
 - 187.32
 - 185.6
-- 187.64
-- 185.98
+- 1.87640000e+02
+- 1.85980000e+02
 - 181.96
 - 166.32
 - 134.78
 - 85.98
 - 29.3
-- 304.64
+- 3.04640000e+02
 - 301.96
 - 302.34
-- 300.66
+- 3.00660000e+02
 - 294.94
-- 293.32
+- 2.93320000e+02
 - 281.74
 - 254.34
 - 208.32
-- 154.02
-- 92.66
+- 1.54020000e+02
+- 9.26600000e+01
 - 30.5
-- 5996.4
+- 5.99640000e+03
 - 5995.0
 - 5970.0
 - 5925.6
-- 5868.6
-- 5655.4
+- 5.86860000e+03
+- 5.65540000e+03
 - 5185.4
-- 4509.6
+- 4.50960000e+03
 - 3658.4
 - 2674.0
 - 1596.62
@@ -40,10 +40,10 @@ data_central:
 - 7008.6
 - 6978.6
 - 6897.8
-- 6597.4
+- 6.59740000e+03
 - 6017.0
 - 5209.0
-- 4227.6
+- 4.22760000e+03
 - 3069.4
 - 1831.68
 - 595.56
@@ -52,20 +52,20 @@ data_central:
 - 234.6
 - 233.08
 - 228.94
-- 215.76
-- 195.84
+- 2.15760000e+02
+- 1.95840000e+02
 - 166.38
 - 134.9
 - 98.4
 - 57.62
 - 18.48
 - 48.44
 - 47.86
-- 47.38
+- 4.73800000e+01
 - 47.72
 - 45.36
 - 42.26
-- 38.44
+- 3.84400000e+01
 - 32.28
 - 26.58
 - 19.32
@@ -75,11 +75,11 @@ data_central:
 - 11.02
 - 10.82
 - 10.66
-- 9.98
+- 9.98000000e+00
 - 9.22
 - 8.56
-- 7.12
-- 5.72
-- 4.06
+- 7.12000000e+00
+- 5.72000000e+00
+- 4.06000000e+00
 - 2.46
 - 0.82

nnpdf_data/nnpdf_data/commondata/ATLAS_Z0_8TEV_LOWMASS/filter.py

@@ -0,0 +1,93 @@
+"""
+filter.py module for ATLAS_Z0_8TEV_LOWMASS dataset
+When running `python filter.py` the relevant uncertainties , data and kinematics yaml
+file will be created in the `nnpdf_data/commondata/ATLAS_Z0_8TEV_LOWMASS` directory.
+"""
+
+import yaml
+from filter_utils import get_kinematics, get_data_values, get_systematics
+from nnpdf_data.filter_utils.utils import prettify_float
+
+yaml.add_representer(float, prettify_float)
+
+
+def filter_ATLAS_Z0_8TEV_LOWMASS_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_8TEV_LOWMASS_systematics(version=3):
+    """
+    This function writes the systematics to a yaml file.
+    """
+
+    with open("metadata.yaml", "r") as file:
+        metadata = yaml.safe_load(file)
+
+    systematics = get_systematics(version=version)
+
+    # error definition
+    error_definitions = {}
+    errors = []
+
+    for sys in systematics:
+        if (sys[0]['name'] == 'stat') or (sys[0]['name'] == 'sys,uncor'):
+            error_definitions[sys[0]['name']] = {
+                "description": f"{sys[0]['name']}",
+                "treatment": "ADD",
+                "type": "UNCORR",
+            }
+
+        elif (sys[0]['name'] == 'ATLAS_LUMI') or (sys[0]['name'] == 'Lumi:M'):
+            error_definitions[sys[0]['name']] = {
+                "description": f"{sys[0]['name']}",
+                "treatment": "MULT",
+                "type": "ATLASLUMI12",
+            }
+
+        else:
+            error_definitions[sys[0]['name']] = {
+                "description": f"{sys[0]['name']}",
+                "treatment": "ADD",
+                "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
+    if version == 1:
+        with open(f"uncertainties_v1.yaml", 'w') as file:
+            yaml.dump(uncertainties_yaml, file, sort_keys=False)
+    else:
+        with open(f"uncertainties.yaml", 'w') as file:
+            yaml.dump(uncertainties_yaml, file, sort_keys=False)
+
+
+if __name__ == "__main__":
+    filter_ATLAS_Z0_8TEV_LOWMASS_data_kinetic()
+    filter_ATLAS_Z0_8TEV_LOWMASS_systematics(version=3)
+    # filter_ATLAS_Z0_8TEV_LOWMASS_systematics(version=1)

nnpdf_data/nnpdf_data/commondata/ATLAS_Z0_8TEV_LOWMASS/filter_utils.py

@@ -0,0 +1,99 @@
+"""
+This module contains helper functions that are used to extract the uncertainties, kinematics and data values 
+from the rawdata files.
+"""
+
+import yaml
+
+ATLAS_LUMI_UNC = 0.018
+
+def get_kinematics():
+    """
+    returns the kinematics in the form of a list of dictionaries.
+    """
+    kin = []
+
+    hepdata_table = f"rawdata/HEPData-ins1630886-v3-Table_5.yaml"
+
+    with open(hepdata_table, 'r') as file:
+        input = yaml.safe_load(file)
+
+    for indep_var1, indep_var2 in zip(
+        input["independent_variables"][1]['values'], input["independent_variables"][2]['values']
+    ):
+
+        kin_value = {
+            'abs_y': {
+                'min': indep_var1['low'],
+                'mid': 0.5 * (indep_var1['low'] + indep_var1['high']),
+                'max': indep_var1['high'],
+            },
+            'm_Z2': {
+                'min': indep_var2['low']**2,
+                'mid': (0.5 * (indep_var2['low'] + indep_var2['high']))**2,
+                'max': indep_var2['high']**2,
+            },
+            'sqrts': {'min': None, 'mid': 8000.0, '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-ins1630886-v3-Table_5.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(version=3):
+    """ """
+
+    uncertainties = []
+
+    hepdata_table = f"rawdata/HEPData-ins1630886-v{version}-Table_5.yaml"
+
+    with open(hepdata_table, 'r') as file:
+        input = yaml.safe_load(file)
+
+    # loop over systematics
+    for unc_labels in input['dependent_variables'][0]['values'][0]['errors']:
+
+        name = f"{unc_labels['label']}"
+        values = []
+
+        # loop over data points
+        for unc in input['dependent_variables'][0]['values']:
+            err = unc['errors']
+            # convert unc from TeV to GeV
+            for e in err:
+                if e['label'] == name:
+                    if name == 'Lumi:M':
+                        values.append(e['symerror'] * unc['value'] * 1000)
+                    else:
+                        values.append(e['symerror'] * 1000)
+
+        uncertainties.append([{"name": name, "values": values}])
+
+    # # Luminosity uncertainty is 1.8 % of the central value (see https://inspirehep.net/literature/1630886)
+    if version == 3:  # in version 1 Lumi is included in the hepdata file already
+        name = "ATLAS_LUMI"
+        values = [ATLAS_LUMI_UNC * val for val in get_data_values()]
+        uncertainties.append([{"name": name, "values": values}])
+    return uncertainties