update medoids

new_medoids = []
        for i in range(0, self.k):
            new_medoid = self.medoids[i]
            old_medoids_cost = self.medoids_cost[i]
            for j in range(len(clusters[i])):
                
                #Cost of the current data points to be compared with the current optimal cost
                cur_medoids_cost = 0
                for dpoint_index in range(len(clusters[i])):
                    cur_medoids_cost += euclideanDistance(clusters[i][j], clusters[i][dpoint_index])
                
                #If current cost is less than current optimal cost,
                #make the current data point new medoid of the cluster
                if cur_medoids_cost < old_medoids_cost:
                    new_medoid = clusters[i][j]
                    old_medoids_cost = cur_medoids_cost
            
            #Now we have the optimal medoid of the current cluster
            new_medoids.append(new_medoid)