merged

.github/workflows/tests.yml

@@ -40,7 +40,7 @@ jobs:
     - name: Build recipe
       shell: bash -l {0}
       run: |
-        conda build -q conda-recipe
+        conda build -q conda-recipe --package-format=1
     - name: Upload noarch conda package to NNPDF server
       if: ${{ (github.ref == 'refs/heads/master' || startsWith(github.ref, 'refs/tags/')) && startsWith(matrix.os, 'ubuntu') }}
       shell: bash -l {0}

nnpdf_data/nnpdf_data/commondata/ATLAS_WPWM_7TEV_36PB/data.yaml

@@ -0,0 +1,23 @@
+data_central:
+- 6.13257400e+05
+- 6.13939929e+05
+- 6.31746805e+05
+- 6.26184703e+05
+- 6.52630155e+05
+- 6.59312827e+05
+- 6.42534838e+05
+- 6.40935479e+05
+- 6.60983495e+05
+- 6.39876031e+05
+- 5.89205893e+05
+- 4.54666184e+05
+- 4.48492862e+05
+- 4.63569622e+05
+- 4.48034447e+05
+- 4.36074909e+05
+- 4.26723243e+05
+- 3.94511949e+05
+- 3.91211361e+05
+- 3.82307923e+05
+- 3.64073193e+05
+- 3.37179513e+05

nnpdf_data/nnpdf_data/commondata/ATLAS_WPWM_7TEV_36PB/filter.py

@@ -0,0 +1,100 @@
+"""
+When running `python filter.py` the relevant data yaml
+file will be created in the `nnpdf_data/commondata/ATLAS_WPWM_7TEV_46FB` directory.
+"""
+
+import yaml
+from filter_utils import get_data_values, get_kinematics, get_systematics
+
+from nnpdf_data.filter_utils.utils import prettify_float
+
+yaml.add_representer(float, prettify_float)
+
+
+def filter_ATLAS_WPWM_7TEV_36FB_data_kinematic():
+    """
+    This function writes the systematics to yaml files.
+    """
+
+    central_values = get_data_values()
+
+    kin = get_kinematics()
+
+    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_WPWM_7TEV_36FB_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 = []
+    counter_1 = 1
+    counter_2 = 0
+    for sys in systematics:
+        if sys[0]['name'] == 'stat':
+            error_definitions[sys[0]['name']] = {
+                "description": "Uncorrelated statistical uncertainties",
+                "treatment": "ADD",
+                "type": "UNCORR",
+            }
+
+        elif sys[0]['name'] == 'uncor':
+            error_definitions[sys[0]['name']] = {
+                "description": f"Sys uncertainty idx: {counter_1}",
+                "treatment": "MULT",
+                "type": "UNCORR",
+            }
+            counter_1 += 1
+
+        elif sys[0]['name'] == 'atlaslumi10':
+            error_definitions[sys[0]['name']] = {
+                "description": f"Sys uncertainty idx: {counter_1}",
+                "treatment": "MULT",
+                "type": "ATLASLUMI10",
+            }
+            counter_1 += 1
+
+        else:
+            error_definitions[sys[0]['name']] = {
+                "description": f"Sys uncertainty idx: {counter_1}",
+                "treatment": "MULT",
+                "type": f"ATLASWZRAP36PB_{counter_2}",
+            }
+            counter_1 += 1
+            counter_2 += 1
+
+    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_WPWM_7TEV_36FB_data_kinematic()
+    filter_ATLAS_WPWM_7TEV_36FB_systematics()

nnpdf_data/nnpdf_data/commondata/ATLAS_WPWM_7TEV_36PB/filter_utils.py

@@ -0,0 +1,96 @@
+"""
+This module contains helper functions that are used to extract the data values 
+from the rawdata files.
+"""
+
+import yaml
+import pandas as pd
+import numpy as np
+
+
+def get_data_values():
+    """
+    returns the central data values in the form of a list.
+    """
+
+    data_central = []
+
+    tables = [5, 3]
+
+    for table in tables:
+        hepdata_table = f"rawdata/HEPData-ins928289-v1-Table_{table}.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 and apply the correction factor
+            data_central.append(value['value'] * 1000 * 1.0187)
+
+    return data_central
+
+
+def get_kinematics():
+    """
+    returns the kinematics in the form of a list of dictionaries.
+    """
+    kin = []
+
+    tables = [5, 3]
+
+    for table in tables:
+        hepdata_table = f"rawdata/HEPData-ins928289-v1-Table_{table}.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 = {
+                'abs_eta': {'min': None, 'mid': (0.5 * (M['low'] + M['high'])), 'max': None},
+                'm_W2': {'min': None, 'mid': 6463.838404, 'max': None},
+                'sqrts': {'min': None, 'mid': 7000.0, 'max': None},
+            }
+            kin.append(kin_value)
+
+    return kin
+
+
+def get_systematics_dataframe():
+    """
+    returns the absolute systematic uncertainties in the form of a pandas dataframe.
+    """
+    sys_rawdata_path = "rawdata/ATLAS-36PB_WPWM.csv"
+
+    abs_unc_df_arr = []
+
+    data_central = get_data_values()
+
+    df = pd.read_csv(sys_rawdata_path)
+
+    # convert (MULT) percentage unc to absolute unc
+    abs_unc_df = (df.T[2:] * data_central).T / 100
+    abs_unc_df_arr.append(abs_unc_df)
+
+    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
+
+
+if __name__ == "__main__":
+    get_data_values()
+    get_kinematics()
+    get_systematics()