clean code + add map to kmeans

kmeans.py

@@ -3,15 +3,17 @@ from datetime import datetime
 
 from sklearn.cluster import KMeans
 import numpy as np
+import folium
 
 import loading as l
 import history as h
 
+colours = ['blue', 'red', 'green', 'orange', 'pink', 'white', 'purple', 'gray']
 
 def getDatasForPeriod(startPeriod, endPeriod, indicators):
     datas = []
     for i in range(int(startPeriod[0:4]), int(endPeriod[0:4]) + 1):
-		datas += session.execute(f"SELECT year, month, day, station, {indicators} FROM {l.table_name_date} where year = {i}")
+        datas += session.execute(f"SELECT year, month, day, station, lat, lon, {indicators} FROM {l.table_name_date} where year = {i}")
 
     return datas
 
@@ -35,12 +37,12 @@ def getDecileForAllStations(startPeriod, endPeriod, table, nb_indicators, indica
             if t[3] not in l.keys():
                 l[t[3]] = []
                 for i in range(nb_indicators):
-					if t[4 + i] != None:
-						l[t[3]].append({indicators_list[i] : [float(t[4 + i])]})
+                    if t[6 + i] != None:
+                        l[t[3]].append({indicators_list[i] : [float(t[6 + i])]})
             else:
                 for i in range(nb_indicators):
-					if t[4 + i] != None:
-						l[t[3]][i][indicators_list[i]].append(float(t[4 + i]))
+                    if t[6 + i] != None:
+                        l[t[3]][i][indicators_list[i]].append(float(t[6 + i]))
 
     # Sort all lists of values
     for station in l.keys():
@@ -77,7 +79,7 @@ def applyKmeans(deciles, nb_indicators, indicators_list, startPeriod, endPeriod)
             t += deciles[station][i][indicators_list[i]]
         table.append(t)
 
-	nb_clusters = 3
+    nb_clusters = 4
     if len(stations_name) < nb_clusters:
         print(f"Le nombre de villes ayant des données est trop inférieur ({len(stations_name)}) pour appliquer les kmeans pour la période du {startPeriod} au {endPeriod}")
         return None
@@ -124,14 +126,26 @@ def kmeans(startPeriod, endPeriod, indicators_list):
 
     table = getDatasForPeriod(startPeriod, endPeriod, indicators)
     table = list(table)
-
+    # Get coordinates
+    coord = dict()
+    for t in table:
+        coord[t[3]]=(t[4], t[5])
     # Get the map with all deciles for all stations and indicators
     table_deciles = getDecileForAllStations(startPeriod, endPeriod, table, nb_indicators, indicators_list_numeric)
 
     station_with_center = applyKmeans(table_deciles, nb_indicators, indicators_list_numeric, startPeriod, endPeriod)
     if station_with_center != None:
-		print(f"Voici les villes et le cluster auxquelles elles appartiennent:")
-		print(f"{station_with_center}")
+        file_name = f"{startPeriod} to {endPeriod}.html"
+        # Create map
+        m = folium.Map(location=[64.2815, 27.6753])
+        # Add Marker for each station
+        for key, value in station_with_center.items():
+            folium.Marker([coord[key][0], coord[key][1]], popup=f"<b>{key}</b>", icon=folium.Icon(color=colours[value])).add_to(m)
+        # Save map
+        m.save(file_name)
+        print(f"La carte a été enregistrée à {file_name}")
+    else:
+        print(f"Aucune clusterisation déterminée")
 
 if __name__ == '__main__':
     cluster = Cluster()

loading.py

@@ -307,7 +307,7 @@ def createTableQueryPartitionningByDate(table):
             peak_wind_drct    	decimal,
             peak_wind_time    	decimal,
             metar             	varchar,
-			PRIMARY KEY((year, month), day, station, hour, minute)
+            PRIMARY KEY((year), month, day, station, hour, minute)
         )"""
     return query
 
@@ -358,7 +358,7 @@ def insertQueryData(row, table):
     """
     return query
 
-# Not to execute when importing file
+    # Not to execute when importing file
 if __name__ == '__main__':
     cluster = Cluster()