prepare 0.4.0, tweaks docs

Author: kno10

CHANGELOG.md

@@ -2,6 +2,10 @@
 
 For changes to the main Rust package, please see <https://github.com/kno10/rust-kmedoids/blob/main/CHANGELOG.md>
 
+## kmedoids 0.4.0 (2022-09-23)
+
+- Add PAMSIL, PAMMEDSIL, FastMSC, FasterMSC (Lars Lenssen)
+
 ## kmedoids 0.3.3 (2022-07-06)
 
 - Improved platform support (prebuilt for manylinux, OSX, Windows) by David Muhr

Cargo.toml

@@ -1,7 +1,7 @@
 [package]
 edition = "2021"
 name = "kmedoids"
-version = "0.3.3"
+version = "0.4.0"
 authors = ["Erich Schubert <erich.schubert@tu-dortmund.de>", "Lars Lenssen <lars.lenssen@tu-dortmund.de>"]
 description = "k-Medoids clustering with the FasterPAM algorithm"
 homepage = "https://github.com/kno10/python-kmedoids"
@@ -14,14 +14,14 @@ name = "kmedoids"
 crate-type = ["cdylib"]
 
 [dependencies]
-rustkmedoids = { version = "0.3.3", package = "kmedoids", git = "https://github.com/kno10/rust-kmedoids" }
+rustkmedoids = { version = "0.4.0", package = "kmedoids", git = "https://github.com/kno10/rust-kmedoids" }
 numpy = "0.16"
 ndarray = "0.15"
 rand = "0.8"
 rayon = "1.5"
 
 [dependencies.pyo3]
-version = "0.16.5"
+version = "0.17.1"
 features = ["extension-module"]
 
 [package.metadata.maturin]

docs/index.rst

@@ -137,7 +137,7 @@ PAM
 
 .. _Alternating:
 
-Alternating k=medoids (k-means style)
+Alternating k-medoids (k-means style)
 =====================================
 
 .. autofunction:: alternating
@@ -221,11 +221,12 @@ an earlier (slower, and now obsolete) version was published as:
      | https://doi.org/10.1007/978-3-030-32047-8_16  
      | Preprint: https://arxiv.org/abs/1810.05691
 
-For further details on the implemented algorithm FasterMSC, see:
+For further details on medoid Silhouette clustering with FasterMSC, see:
 
-	 | Lars Lenssen, Erich Schubert:
-	 | Clustering by Direct Optimization of the Medoid Silhouette
-	 | In: 15th International Conference on Similarity Search and Applications (SISAP 2022).
+     | Lars Lenssen, Erich Schubert:  
+     | **Clustering by Direct Optimization of the Medoid Silhouette**  
+     | In: 15th International Conference on Similarity Search and Applications (SISAP 2022).  
+     | https://doi.org/10.1007/978-3-031-17849-8_15
 
 This is a port of the original Java code from `ELKI <https://elki-project.github.io/>`_ to Rust.
 The `Rust version <https://github.com/kno10/rust-kmedoids>`_ is then wrapped for use with Python.

kmedoids/__init__.py

@@ -36,6 +36,7 @@
 | Lars Lenssen, Erich Schubert:
 | Clustering by Direct Optimization of the Medoid Silhouette
 | In: 15th International Conference on Similarity Search and Applications (SISAP 2022).
+| https://doi.org/10.1007/978-3-031-17849-8_15
 
 | Leonard Kaufman, Peter J. Rousseeuw:
 | Clustering by means of medoids.
@@ -460,6 +461,7 @@ def fastmsc(diss, medoids, max_iter=100, init="random", random_state=None):
 	| Lars Lenssen, Erich Schubert:
 	| Clustering by Direct Optimization of the Medoid Silhouette
 	| In: 15th International Conference on Similarity Search and Applications (SISAP 2022).
+	| https://doi.org/10.1007/978-3-031-17849-8_15
 
 	:param diss: square numpy array of dissimilarities
 	:type diss: ndarray
@@ -503,6 +505,7 @@ def fastermsc(diss, medoids, max_iter=100, init="random", random_state=None):
 	| Lars Lenssen, Erich Schubert:
 	| Clustering by Direct Optimization of the Medoid Silhouette
 	| In: 15th International Conference on Similarity Search and Applications (SISAP 2022).
+	| https://doi.org/10.1007/978-3-031-17849-8_15
 
 	:param diss: square numpy array of dissimilarities
 	:type diss: ndarray
@@ -646,7 +649,7 @@ class SKLearnClusterer(BaseEstimator, ClusterMixin, TransformerMixin):
 
 
 class KMedoids(SKLearnClusterer):
-	"""K-Medoids Clustering using PAM and FasterPAM (sklearn-compatible API).
+	"""K-Medoids Clustering using PAM, FasterPAM, and FasterMSC (sklearn-compatible API).
 
 	References:
 
@@ -665,6 +668,7 @@ class KMedoids(SKLearnClusterer):
 	| Lars Lenssen, Erich Schubert:
 	| Clustering by Direct Optimization of the Medoid Silhouette
 	| In: 15th International Conference on Similarity Search and Applications (SISAP 2022).
+	| https://doi.org/10.1007/978-3-031-17849-8_15
 
 	| Leonard Kaufman, Peter J. Rousseeuw:
 	| Clustering by means of medoids.
@@ -760,7 +764,9 @@ def fit(self, X, y=None):
 		else:
 			raise ValueError(
 				f"method={self.method} is not supported. Supported methods "
-				f"are 'fasterpam', 'fastpam1', 'pam' and 'alternate'."
+				f"are 'fasterpam', 'fastpam1', 'pam', 'alternate', "
+				f"'fastermsc', 'fastmsc', 'pamsil', and 'pammedsil'. "
+				f"Recommended values are 'fasterpam' for classic k-medoids and 'fastermsc' for Silhouette optimization."
 			)
 		self.labels_ = result.labels
 		self.medoid_indices_ = result.medoids