Remove HepData raw tables from the repository

.github/workflows/check_newcd.yml

@@ -19,16 +19,17 @@ jobs:
       - name: Run the filters 📦
         shell: bash -l {0}
         run: |
+          download-hepdata
           here=$PWD
           readarray -d '' array < <(find ./nnpdf_data/nnpdf_data/commondata -name "filter.py" -print0)
           for datname in "${array[@]}"; do dirpath=${datname%/*}; cd $dirpath; python filter.py || exit $?; cd $here; done
-      - name: Check for modified files 🛎️
-        uses: tj-actions/verify-changed-files@v20
-        id: verify-changed-files
-      - name: List all changed files and fail if any 🛎️
-        if: steps.verify-changed-files.outputs.files_changed == 'true'
-        env:
-          CHANGED_FILES: ${{ steps.verify-changed-files.outputs.changed_files }}
+      - name: Check for modified commondata files 🛎️
+        shell: bash -l {0}
         run: |
-          echo "Changed files: $CHANGED_FILES"
-          exit 1
+          if [[ -n "$(git diff --name-only)" ]]; then
+            echo "Some of the commondata files have changed ❌"
+            git diff --name-only
+            exit 1
+          else
+            echo "No modified commondata files ✅"
+          fi

nnpdf_data/nnpdf_data/commondata/CDF_Z0_1P96TEV/metadata.yaml

@@ -11,6 +11,7 @@ iNSPIRE:
   url: https://inspirehep.net/literature/856131
 hepdata:
   url: https://www.hepdata.net/record/ins856131
+  version: 1
 implemented_observables:
 - observable_name: ZRAP
   observable:

nnpdf_data/nnpdf_data/commondata/CMS_1JET_13TEV_DIF/metadata.yaml

@@ -10,7 +10,7 @@ iNSPIRE:
   url: "https://inspirehep.net/literature/1972986"
 hepdata:
   url: "https://www.hepdata.net/record/ins1972986"
-  version: 1
+  version: 2
 
 version: 1
 version_comment: "NA"

nnpdf_data/nnpdf_data/commondata/LHCB_WPWM_7TEV_MUON/filter.py

@@ -36,8 +36,8 @@ def load_yaml(table_id: int, version: int = 1) -> dict:
         ditionary containing the table contents
 
     """
-    filename = f"HEPData-ins1373300-v{version}-Table_{table_id}"
-    table = pathlib.Path(f"./rawdata/{filename}.yaml")
+    foldername = f"HEPData-ins1373300-v{version}-yaml"
+    table = pathlib.Path(f"./{foldername}/Table{table_id}.yaml")
 
     return yaml.safe_load(table.read_text())
 
@@ -147,13 +147,15 @@ def get_errors(hepdata: dict, bin_index: list, indx: int = 0) -> dict:
     return {"stat": stat, "sys_corr": sys_corr, "sys_beam": sys_beam, "sys_lumi": sys_lumi}
 
 
-def read_corrmatrix(nb_datapoints: int) -> np.ndarray:
+def read_corrmatrix(nb_datapoints: int, version: int = 1) -> np.ndarray:
     """Read the matrix and returns a symmetrized verions.
 
     Parameters
     ----------
     nb_datapoints: int
         total number of datapoints
+    version: int
+        HepData version
 
     Returns
     -------
@@ -162,7 +164,7 @@ def read_corrmatrix(nb_datapoints: int) -> np.ndarray:
 
     """
     corrmat = pd.read_csv(
-        "./rawdata/corrmat.corr",
+        f"./HEPData-ins1373300-v{version}-yaml/corrmat.corr",
         names=[f'{i}' for i in range(nb_datapoints)],
         delim_whitespace=True,
     )
@@ -366,7 +368,7 @@ def main_filter(boson: str = "Z") -> None:
 
     errors_combined = concatenate_dicts(combined_errors)
     # Compute the Artifical Systematics from CovMat
-    corrmat = read_corrmatrix(nb_datapoints=nbpoints)
+    corrmat = read_corrmatrix(nb_datapoints=nbpoints, version=version)
     covmat = multiply_syst(corrmat, errors_combined["sys_corr"])
     artunc = generate_artificial_unc(ndata=nbpoints, covmat_list=covmat.tolist(), no_of_norm_mat=0)
     errors = format_uncertainties(errors_combined, artunc, bslice)