Fix KeyError('wyckoff_spglib') in fetch_process_wbm_dataset.py (#34)

* fix Generating Aflow labels raised exception=KeyError('wyckoff_spglib') (closes #32)

* move out_dir of site-stats.json.gz to data/(wbm|mp) (closes #33)

data/wbm/fetch_process_wbm_dataset.py

@@ -595,11 +595,11 @@ def fix_bad_struct_index_mismatch(material_id: str) -> str:
 
     wyckoff_col = "wyckoff_spglib"
     if wyckoff_col not in df_wbm:
-        df_wbm[wyckoff_col] = None
+        df_summary[wyckoff_col] = None
 
     for idx, struct in tqdm(df_wbm.initial_structure.items(), total=len(df_wbm)):
         if not pd.isna(df_summary.loc[idx, wyckoff_col]):
-            continue
+            continue  # Aflow label already computed
         try:
             struct = Structure.from_dict(struct)
             df_summary.loc[idx, wyckoff_col] = get_aflow_label_from_spglib(struct)

pyproject.toml

@@ -96,6 +96,7 @@ select = [
   "PYI", # flakes8-pyi
   "Q",   # flake8-quotes
   "RET", # flake8-return
+  "RSE", # flake8-raise
   "RUF", # Ruff-specific rules
   "SIM", # flake8-simplify
   "TID", # tidy imports

scripts/compute_struct_fingerprints.py

@@ -36,7 +36,7 @@
 slurm_array_task_id = int(os.getenv("SLURM_ARRAY_TASK_ID", 0))
 slurm_array_task_count = 100
 
-out_dir = f"{ROOT}/data/{data_name}/structure-fingerprints"
+out_dir = f"{ROOT}/data/{data_name}"
 os.makedirs(out_dir, exist_ok=True)
 
 slurm_vars = slurm_submit(