Update news section to include set of changes

Author: bytesnake

algorithms/linfa-clustering/examples/kmeans.rs

@@ -7,6 +7,8 @@ use ndarray_npy::write_npy;
 use ndarray_rand::rand::SeedableRng;
 use rand_isaac::Isaac64Rng;
 
+use linfa_nn::distance::LInfDist;
+
 // A routine K-means task: build a synthetic dataset, fit the algorithm on it
 // and save both training data and predictions to disk.
 fn main() {
@@ -20,7 +22,7 @@ fn main() {
 
     // Configure our training algorithm
     let n_clusters = expected_centroids.len_of(Axis(0));
-    let model = KMeans::params_with_rng(n_clusters, rng)
+    let model = KMeans::params_with(n_clusters, rng, LInfDist)
         .max_n_iterations(200)
         .tolerance(1e-5)
         .fit(&dataset)

docs/website/content/news/release_050.md

@@ -3,29 +3,47 @@ title = "Release 0.5.0"
 date = "2021-10-20"
 +++
 
-Release 0.5.0
-
 Linfa's 0.5.0 release adds initial support for the OPTICS algorithm, multinomials logistic regression, and the family of nearest neighbor algorithms. Furthermore, we have improved documentation and introduced hyperparameter checking to all algorithms.
 
+<!-- more -->
+
+## New algorithms
+
+[OPTICS](https://en.wikipedia.org/wiki/OPTICS_algorithm) is an algorithm for finding density-based clusters. It can produce reachability-plots, hierarchical structure of clusters. Analysing data without prior assumption of any distribution is a common use-case. The algorithm is added to `linfa-clustering` and an example can be find at [linfa-clustering/examples/optics.rs](https://github.com/rust-ml/linfa/blob/master/algorithms/linfa-clustering/examples/optics.rs).
+
+Extending logistic regression to the multinomial distribution generalizes it to [multiclass problems](https://en.wikipedia.org/wiki/Multinomial_logistic_regression). This release adds support for multinomial logistic regression to `linfa-logistic`, you can experiment with the example at [linfa-logistic/examples/winequality_multi.rs](https://github.com/rust-ml/linfa/blob/master/algorithms/linfa-logistic/examples/winequality_multi.rs).
+
+Nearest neighbor search finds the set of neighborhood points to a given sample. It appears in numerous fields of applications as a distance metric provider. (e.g. clustering) This release adds a family of nearest neighbor algorithms, namely [Ball tree](https://en.wikipedia.org/wiki/Ball_tree), [K-d tree](https://en.wikipedia.org/wiki/K-d_tree) and naive linear search. You can find an example in the next section.
+
+## Improvements
+
+ * use least-square solver from `ndarray-linalg` in `linfa-linear`
+ * make clustering algorithms generic over distance metrics
+ * bump `ndarray` to 0.15
+ * introduce `ParamGuard` trait for explicit and implicit parameter checking (read more in the [CONTRIBUTE.md](https://github.com/rust-ml/linfa/blob/master/CONTRIBUTE.md#parameters-and-checking))
+ * improve documentation in various places
+
 ## Nearest Neighbors
 
 You can now choose from a growing list of NN implementations. The family provides efficient distance metrics to KMeans, DBSCAN etc. The example shows how to use KDTree nearest neighbor to find all the points in a set of observations that are within a certain range of a candidate point.
+
+You can query nearest points explicitly:
+
 ```rust
 // create a KDTree index consisting of all the points in the observations, using Euclidean distance
-let kdtree = CommonNearestNeighbour::KdTree.from_batch(observations, L2Dist).unwrap();
+let kdtree = CommonNearestNeighbour::KdTree.from_batch(observations, L2Dist)?;
 let candidate = observations.row(2);
-let points = kdtree.within_range(candidate.view(), range).unwrap();
+let points = kdtree.within_range(candidate.view(), range)?;
 ```
 
-## Multinomial logistic regression
+Or use one of the distance metrics implicitly, here demonstrated for KMeans:
 
-Logistic regression models problems with two possible discrete outcomes. Extending this to the multinomial distribution yields the Multinomial Logistic Regression which can model datasets with an arbitrary number of outcomes. 
-
-We will try to model the winequality of 1440 samples, you can find the full example [here](https://github.com/rust-ml/linfa/blob/master/algorithms/linfa-logistic/examples/winequality_multi.rs)
 ```rust
-    // fit a Logistic regression model with 150 max iterations
-   let model = MultiLogisticRegression::default()
-       .max_iterations(50)
-       .fit(&train)
-       .unwrap();
+use linfa_nn::distance::LInfDist;
+
+let model = KMeans::params_with(3, rng, LInfDist)
+    .max_n_iterations(200)
+    .tolerance(1e-5)
+    .fit(&dataset)?;
 ```
+