Fix

history.py

@@ -2,6 +2,7 @@ import matplotlib.pyplot as plt
 from cassandra.cluster import Cluster
 from datetime import datetime
 import numpy as np
+import math
 
 import loading
 
@@ -66,6 +67,7 @@ def verifyYearValidity(dateMin, dateMax):
 	return True
 
 def plotHistory(station, indicator, dateMin, dateMax):
+	dateMax = dateMax + 1
 	# Accept only indicator with numeric values (not factors)
 	if indicator in numeric_columns:
 		if not verifyYearValidity(dateMin, dateMax):
@@ -109,11 +111,14 @@ def plotHistory(station, indicator, dateMin, dateMax):
 
 		# Configure graduation of plot
 		# Need to find a nice step for graduation (no more than 6 values to be understandable)
-		step_graduation = max((dateMax - dateMin) // 5, 1)
+		step_graduation = max(math.ceil((dateMax - dateMin) / 5), 1)
 		# Convert the graduation to the date format (yyyy-month-01)
 		graduation = ["20" + "0" * (2 - len(str(i))) + str(i) + "-01-01" for i in range(int(str(dateMin)[2:4]), int(str(dateMax)[2:4]), step_graduation)]
-		# Add the last value of graduation for the last day of measures
-		graduation.append("20" + "0" * (2 - len(str(dateMax - 1)[2:4])) + str(dateMax - 1)[2:4] + "-12-31")
+
+		# Pour ne pas que la date de fin se superpose avec une autre date
+		if (dateMax - dateMin) % 2 == 0 or step_graduation == 1:
+			# Add the last value of graduation for the last day of measures
+			graduation.append("20" + "0" * (2 - len(str(dateMax - 1)[2:4])) + str(dateMax - 1)[2:4] + "-12-31")
 		
 		# Plot, with both measures and season mean
 		fig, ax1 = plt.subplots()

kmeans.py

@@ -74,14 +74,14 @@ def applyKmeans(deciles, nb_indicators, indicators_list, startPeriod, endPeriod)
 		stations_name.append(station)
 		for i in range(nb_indicators):
 			t += deciles[station][i][indicators_list[i]]
-			print (t)
 		table.append(t)
 
+	nb_clusters = 3
 	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
 
-	kmeans = KMeans(n_clusters = 3, max_iter = 100).fit(table)
+	kmeans = KMeans(n_clusters = nb_clusters, max_iter = 100).fit(table)
 
 	res = {}
 	i = 0