@@ -125,10 +125,10 @@ Choose the optimal number of clusters
125125 X, _ = fetch_openml(' mnist_784' , version = 1 , return_X_y = True , as_frame = False )
126126 X = X[:10000 ]
127127 diss = euclidean_distances(X)
128- bk = kmedoids.bestk (diss, 100 )
129- print (" Optimal number of clusters according to the Medoid Silhouette:" , bk .bestk)
130- print (" Medoid Silhouette over range of k:" , bk .losses)
131- print (" Range of k:" , bk .rangek)
128+ dm = kmedoids.dynmsc (diss, 100 )
129+ print (" Optimal number of clusters according to the Medoid Silhouette:" , dm .bestk)
130+ print (" Medoid Silhouette over range of k:" , dm .losses)
131+ print (" Range of k:" , dm .rangek)
132132
133133 Memory Requirements
134134-------------------
@@ -151,10 +151,9 @@ Implemented Algorithms
151151* :ref: `FasterMSC<fastermsc> ` (Lenssen and Schubert, 2022)
152152* :ref: `FastMSC<fastmsc> ` (Lenssen and Schubert, 2022)
153153* :ref: `DynMSC<dynmsc> ` (Lenssen and Schubert, 2023)
154- * :ref: `Bestk<bestk> ` (Lenssen and Schubert, 2023)
155154* :ref: `PAMSIL<pamsil> ` (Van der Laan and Pollard, 2003)
156155* :ref: `PAMMEDSIL<pammedsil> ` (Van der Laan and Pollard, 2003)
157- * :ref: `Medoid Silhouette <medoid_silhouette> ` (Van der Laan and Pollard, 2003)
156+ * :ref: `MedoidSilhouette <medoid_silhouette> ` (Van der Laan and Pollard, 2003)
158157
159158Note that the k-means style "alternating" algorithm yields rather poor result quality
160159(see Schubert and Rousseeuw 2021 for an example and explanation).
@@ -215,13 +214,6 @@ DynMSC
215214
216215.. autofunction :: dynmsc
217216
218- .. _Bestk :
219-
220- DynMSC
221- =========
222-
223- .. autofunction :: bestk
224-
225217.. _PAMSIL :
226218
227219PAMSIL
0 commit comments