k-means clustering: save the number of iterations in a matrix
7 vues (au cours des 30 derniers jours)
Afficher commentaires plus anciens
pavlos
le 31 Jan 2014
Réponse apportée : Phu Lai
le 8 Déc 2016
Hello,
With the following command the number of iterations are displayed in the Command Window.
idx = kmeans(X,30,'display','iter');
iter phase num sum
1 1 365 40.9896
2 1 60 37.0645
3 1 18 36.001
4 1 3 35.7457
5 1 1 35.6472
6 2 39 34.8684
7 2 32 34.1611
8 2 16 33.7392
9 2 11 33.3268
10 2 4 33.2144
11 2 1 33.2046
11 iterations, total sum of distances = 33.2046
Is there a way to save the displayed number of iterations and the total sum of distances in a matrix stored in the Workspace?
Thank you. Best,
Pavlos
0 commentaires
Réponse acceptée
David Sanchez
le 31 Jan 2014
[IDX,C,sumd,D] = kmeans(X,30,'display','iter');
where
sum(sumd)
is the sum of distances The number of iterations is not stored in any variable
0 commentaires
Plus de réponses (1)
Phu Lai
le 8 Déc 2016
Another way around to get the number of iterations is modifying the kmeans.m source code :)
0 commentaires
Voir également
Catégories
En savoir plus sur Cluster Analysis and Anomaly Detection dans Help Center et File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!