912 RegioStaR mapping

pycode/memilio-epidata/memilio/epidata/defaultDict.py

@@ -106,7 +106,11 @@
     'region_name': 'County',
     'region_id': 'ID_County',
     'desc': 'Description',
-    'incidence': 'Incidence'
+    'incidence': 'Incidence',
+    'values': 'Values',
+    'description': 'Description',
+    'variable': 'Variable',
+    'district': 'District'
 }
 
 GerEng = {
@@ -155,7 +159,11 @@
     'ags5': EngEng['idCounty'],
     'm_code': EngEng['npiCode'],
     'code': EngEng['npiCode'],
-    'Bundesland_Id': EngEng['idState']
+    'Bundesland_Id': EngEng['idState'],
+    'Werte': EngEng['values'],
+    'Beschreibung': EngEng['description'],
+    'Variable': EngEng['variable'],
+    'gem_20': EngEng['district']
 }
 
 EsEng = {'fecha': EngEng['date'],

pycode/memilio-epidata/memilio/epidata/geoModificationGermany.py

@@ -632,3 +632,63 @@ def merge_df_counties_all(
         df = merge_df_counties(df, key, val, sorting, columns, method)
 
     return df
+
+
+def regiostar_mapping(RegioStaR='7', count_multiple_entries=False):
+    """! Mapping of RegioStaR categories to countyIDs
+
+    @param RegioStar str One of [2,4,17,7(default),5,Gem7,Gem5]
+    @return Dict Mapped Regiostar
+    """
+    if not isinstance(RegioStaR, str):
+        RegioStaR = str(RegioStaR)
+    RegioStaR = 'Regiostar'+RegioStaR
+    # read file
+    xlsx = pd.ExcelFile(
+        'https://www.mcloud.de/downloads/mcloud/536149D1-2902-4975-9F7D-253191C0AD07/RegioStaR-Referenzdateien.xlsx', engine='openpyxl')
+    # save different sheets into different variables
+    codeplan = pd.read_excel(xlsx, sheet_name='Codeplan', header=4)
+    codeplan.rename(dd.GerEng, axis=1, inplace=True)
+    codelist = pd.read_excel(xlsx, sheet_name='ReferenzGebietsstand2020')
+    codelist.rename(dd.GerEng, axis=1, inplace=True)
+    # get start and end point of regiostar values
+    idx_begin = np.where(codeplan[dd.EngEng['variable']] == RegioStaR)[0][0]
+    idx_end = idx_begin+1
+
+    while pd.isnull(codeplan[dd.EngEng['variable']][idx_end]):
+        idx_end += 1
+    # cut codeplan
+    codeplan = codeplan.iloc[idx_begin:idx_end]
+    # save in array
+    regio_values = codeplan[dd.EngEng['values']].values
+    # save them as dict
+    RegioDict = dict()
+    #
+    new_cols = list(codelist.columns[:9])
+    new_cols += [c.lower() for c in codelist.columns[9:]]
+    codelist.columns = new_cols
+
+    if count_multiple_entries:
+        # create zero-filled dataframe with countyIDs as rows and RegioStaRIDs as columns
+        count_df = pd.DataFrame(
+            0, columns=codeplan[dd.EngEng['description']], index=dd.County.keys())
+
+    for v in regio_values:
+        regio_desc = codeplan[codeplan[dd.EngEng['values']]
+                              == v][dd.EngEng['description']].values[0]
+        subframe = codelist.iloc[np.where(codelist[RegioStaR.lower()] == v)]
+        # remove last 3 digits of gem_20 -> now represents CountyID
+        district_ids = subframe[dd.EngEng['district']].values
+        county_ids = [str(d_id)[:-3] for d_id in district_ids]
+        if count_multiple_entries:
+            for c_id in county_ids:
+                count_df.at[int(c_id), regio_desc] += 1
+        unique_county_ids = set(county_ids)
+        # write into dict
+        RegioDict[regio_desc] = unique_county_ids
+
+    if count_multiple_entries:
+        gd.write_dataframe(count_df, os.path.join(
+            dd.defaultDict['out_folder'], 'Germany'), 'multiple_entries_'+RegioStaR.lower(), 'txt')
+
+    return RegioDict