rust-ml · relf · Dec 23, 2025 · Dec 22, 2025 · Dec 22, 2025 · Dec 23, 2025
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -1,9 +1,23 @@
+Version 0.8.1 - 2025-12-22
+==========================
+New Algorithms
+--------------
+ * Least angle regression in a new `linfa-lars` crate by [@aswin-1111]
+ * Random Forest in `linfa-ensemble` by [@ladezai]
+ * AdaBoost in `linfa-ensemble` by [@rathideep22]
+
+Changes
+-------
+ * add bootstrap-with-indices utilities for `Dataset`
+ * fix ndarray version mismatch
+ * fix missing exponential in `linfa-ica`
+
 Version 0.8.0 - 2025-09-30
 ==========================
 New Algorithms
 --------------
- * Bernouilli naive bayes in [linfa-bayes] by [@wildart]
- * Bootstrap aggregation (aka Bagging) in new ensemble methods crate [linfa-ensemble] by [@jk1015]
+ * Bernouilli naive bayes in `linfa-bayes` by [@wildart]
+ * Bootstrap aggregation (aka Bagging) in new ensemble methods crate `linfa-ensemble` by [@jk1015]
 
 Changes
 -------

diff --git a/Cargo.toml b/Cargo.toml
@@ -1,6 +1,6 @@
 [package]
 name = "linfa"
-version = "0.8.0"
+version = "0.8.1"
 authors = [
     "Luca Palmieri <rust@lpalmieri.com>",
     "Lorenz Schmidt <bytesnake@mailbox.org>",

diff --git a/README.md b/README.md
@@ -28,25 +28,26 @@ Where does `linfa` stand right now? [Are we learning yet?](http://www.arewelearn
 `linfa` currently provides sub-packages with the following algorithms: 
 
 
-| Name | Purpose | Status | Category |  Notes | 
-| :--- | :--- | :---| :--- | :---| 
-| [bayes](algorithms/linfa-bayes/) | Naive Bayes | Tested | Supervised learning | Contains Bernouilli, Gaussian and Multinomial Naive Bayes |
-| [clustering](algorithms/linfa-clustering/) | Data clustering | Tested / Benchmarked  | Unsupervised learning | Clustering of unlabeled data; contains K-Means, Gaussian-Mixture-Model, DBSCAN and OPTICS | 
-| [ensemble](algorithms/linfa-ensemble/) | Ensemble methods | Tested | Supervised learning | Contains bagging |
-| [elasticnet](algorithms/linfa-elasticnet/) | Elastic Net | Tested | Supervised learning | Linear regression with elastic net constraints |
-| [ftrl](algorithms/linfa-ftrl/) | Follow The Regularized Leader - proximal | Tested  / Benchmarked | Partial fit | Contains L1 and L2 regularization. Possible incremental update | 
-| [hierarchical](algorithms/linfa-hierarchical/) | Agglomerative hierarchical clustering | Tested | Unsupervised learning | Cluster and build hierarchy of clusters |
-| [ica](algorithms/linfa-ica/) | Independent component analysis | Tested | Unsupervised learning | Contains FastICA implementation |
-| [kernel](algorithms/linfa-kernel/) | Kernel methods for data transformation  | Tested  | Pre-processing | Maps feature vector into higher-dimensional space | 
-| [linear](algorithms/linfa-linear/) | Linear regression | Tested  | Partial fit | Contains Ordinary Least Squares (OLS), Generalized Linear Models (GLM) | 
-| [logistic](algorithms/linfa-logistic/) | Logistic regression | Tested  | Partial fit | Builds two-class logistic regression models |
-| [nn](algorithms/linfa-nn/) | Nearest Neighbours & Distances | Tested / Benchmarked | Pre-processing | Spatial index structures and distance functions |
-| [pls](algorithms/linfa-pls/) | Partial Least Squares | Tested | Supervised learning | Contains PLS estimators for dimensionality reduction and regression |
-| [preprocessing](algorithms/linfa-preprocessing/) | Normalization & Vectorization| Tested / Benchmarked | Pre-processing | Contains data normalization/whitening and count vectorization/tf-idf |
-| [reduction](algorithms/linfa-reduction/) | Dimensionality reduction | Tested | Pre-processing | Diffusion mapping, Principal Component Analysis (PCA), Random projections |
-| [svm](algorithms/linfa-svm/) | Support Vector Machines | Tested  | Supervised learning | Classification or regression analysis of labeled datasets | 
-| [trees](algorithms/linfa-trees/) | Decision trees | Tested / Benchmarked  | Supervised learning | Linear decision trees |
-| [tsne](algorithms/linfa-tsne/) | Dimensionality reduction | Tested | Unsupervised learning | Contains exact solution and Barnes-Hut approximation t-SNE |
+| Name                                             | Purpose                                  | Status                | Category              | Notes                                                                                     |
+| :----------------------------------------------- | :--------------------------------------- | :-------------------- | :-------------------- | :---------------------------------------------------------------------------------------- |
+| [bayes](algorithms/linfa-bayes/)                 | Naive Bayes                              | Tested                | Supervised learning   | Contains Bernouilli, Gaussian and Multinomial Naive Bayes                                 |
+| [clustering](algorithms/linfa-clustering/)       | Data clustering                          | Tested / Benchmarked  | Unsupervised learning | Clustering of unlabeled data; contains K-Means, Gaussian-Mixture-Model, DBSCAN and OPTICS |
+| [ensemble](algorithms/linfa-ensemble/)           | Ensemble methods                         | Tested                | Supervised learning   | Contains bagging, random forest and AdaBoost                                              |
+| [elasticnet](algorithms/linfa-elasticnet/)       | Elastic Net                              | Tested                | Supervised learning   | Linear regression with elastic net constraints                                            |
+| [ftrl](algorithms/linfa-ftrl/)                   | Follow The Regularized Leader - proximal | Tested  / Benchmarked | Partial fit           | Contains L1 and L2 regularization. Possible incremental update                            |
+| [hierarchical](algorithms/linfa-hierarchical/)   | Agglomerative hierarchical clustering    | Tested                | Unsupervised learning | Cluster and build hierarchy of clusters                                                   |
+| [ica](algorithms/linfa-ica/)                     | Independent component analysis           | Tested                | Unsupervised learning | Contains FastICA implementation                                                           |
+| [kernel](algorithms/linfa-kernel/)               | Kernel methods for data transformation   | Tested                | Pre-processing        | Maps feature vector into higher-dimensional space                                         |
+| [lars](algorithms/linfa-lars/)                   | Linear regression                        | Tested                | Supervised learning   | Contains Least Angle Regression (LARS)                                                    |
+| [linear](algorithms/linfa-linear/)               | Linear regression                        | Tested                | Supervised learning   | Contains Ordinary Least Squares (OLS), Generalized Linear Models (GLM)                    |
+| [logistic](algorithms/linfa-logistic/)           | Logistic regression                      | Tested                | Partial fit           | Builds two-class logistic regression models                                               |
+| [nn](algorithms/linfa-nn/)                       | Nearest Neighbours & Distances           | Tested / Benchmarked  | Pre-processing        | Spatial index structures and distance functions                                           |
+| [pls](algorithms/linfa-pls/)                     | Partial Least Squares                    | Tested                | Supervised learning   | Contains PLS estimators for dimensionality reduction and regression                       |
+| [preprocessing](algorithms/linfa-preprocessing/) | Normalization & Vectorization            | Tested / Benchmarked  | Pre-processing        | Contains data normalization/whitening and count vectorization/tf-idf                      |
+| [reduction](algorithms/linfa-reduction/)         | Dimensionality reduction                 | Tested                | Pre-processing        | Diffusion mapping, Principal Component Analysis (PCA), Random projections                 |
+| [svm](algorithms/linfa-svm/)                     | Support Vector Machines                  | Tested                | Supervised learning   | Classification or regression analysis of labeled datasets                                 |
+| [trees](algorithms/linfa-trees/)                 | Decision trees                           | Tested / Benchmarked  | Supervised learning   | Linear decision trees                                                                     |
+| [tsne](algorithms/linfa-tsne/)                   | Dimensionality reduction                 | Tested                | Unsupervised learning | Contains exact solution and Barnes-Hut approximation t-SNE                                |
 
 We believe that only a significant community effort can nurture, build, and sustain a machine learning ecosystem in Rust - there is no other way forward.
 
@@ -56,11 +57,11 @@ If this strikes a chord with you, please take a look at the [roadmap](https://gi
 
 Some algorithm crates need to use an external library for linear algebra routines. By default, we use a pure-Rust implementation. However, you can also choose an external BLAS/LAPACK backend library instead, by enabling the `blas` feature and a feature corresponding to your BLAS backend. Currently you can choose between the following BLAS/LAPACK backends: `openblas`, `netblas` or `intel-mkl`.
 
-|Backend  | Linux | Windows | macOS |
-|:--------|:-----:|:-------:|:-----:|
-|OpenBLAS |✔️      |-        |-      |
-|Netlib   |✔️      |-        |-      |
-|Intel MKL|✔️      |✔️        |✔️      |
+| Backend   | Linux | Windows | macOS |
+| :-------- | :---: | :-----: | :---: |
+| OpenBLAS  |   ✔️   |    -    |   -   |
+| Netlib    |   ✔️   |    -    |   -   |
+| Intel MKL |   ✔️   |    ✔️    |   ✔️   |
 
 Each BLAS backend has two features available. The feature allows you to choose between linking the BLAS library in your system or statically building the library. For example, the features for the `intel-mkl` backend are `intel-mkl-static` and `intel-mkl-system`.
 

diff --git a/algorithms/linfa-bayes/Cargo.toml b/algorithms/linfa-bayes/Cargo.toml
@@ -25,7 +25,7 @@ ndarray = { version = "0.16", features = ["approx"] }
 ndarray-stats = "0.6"
 thiserror = "2.0"
 
-linfa = { version = "0.8.0", path = "../.." }
+linfa = { version = "0.8.1", path = "../.." }
 
 [dev-dependencies]
 approx = "0.5"

diff --git a/algorithms/linfa-clustering/Cargo.toml b/algorithms/linfa-clustering/Cargo.toml
@@ -44,7 +44,7 @@ num-traits = "0.2"
 rand_xoshiro = "0.6"
 space = "0.19"
 thiserror = "2.0"
-linfa = { version = "0.8.0", path = "../.." }
+linfa = { version = "0.8.1", path = "../.." }
 linfa-nn = { version = "0.8.0", path = "../linfa-nn" }
 noisy_float = "0.2.0"
 
@@ -57,7 +57,7 @@ criterion = "0.5"
 serde_json = "1"
 approx = "0.5"
 lax = "=0.17.0"
-linfa = { version = "0.8.0", path = "../..", features = ["benchmarks"] }
+linfa = { version = "0.8.1", path = "../..", features = ["benchmarks"] }
 
 [[bench]]
 name = "k_means"

diff --git a/algorithms/linfa-elasticnet/Cargo.toml b/algorithms/linfa-elasticnet/Cargo.toml
@@ -37,7 +37,7 @@ num-traits = "0.2"
 approx = "0.5"
 thiserror = "2.0"
 
-linfa = { version = "0.8.0", path = "../.." }
+linfa = { version = "0.8.1", path = "../.." }
 
 [dev-dependencies]
 linfa-datasets = { version = "0.8.0", path = "../../datasets", features = [

diff --git a/algorithms/linfa-ensemble/Cargo.toml b/algorithms/linfa-ensemble/Cargo.toml
@@ -31,7 +31,7 @@ ndarray = { version = "0.16", features = ["rayon", "approx"] }
 ndarray-rand = "0.15"
 rand = "0.8.5"
 
-linfa = { version = "0.8.0", path = "../.." }
+linfa = { version = "0.8.1", path = "../.." }
 linfa-trees = { version = "0.8.0", path = "../linfa-trees" }
 
 [dev-dependencies]

diff --git a/algorithms/linfa-ftrl/Cargo.toml b/algorithms/linfa-ftrl/Cargo.toml
@@ -32,15 +32,15 @@ thiserror = "2.0"
 rand = "0.8.5"
 rand_xoshiro = "0.6.0"
 
-linfa = { version = "0.8.0", path = "../.." }
+linfa = { version = "0.8.1", path = "../.." }
 
 [dev-dependencies]
 criterion = "0.5"
 approx = "0.5"
 linfa-datasets = { version = "0.8.0", path = "../../datasets", features = [
     "winequality",
 ] }
-linfa = { version = "0.8.0", path = "../..", features = ["benchmarks"] }
+linfa = { version = "0.8.1", path = "../..", features = ["benchmarks"] }
 
 [[bench]]
 name = "ftrl"

diff --git a/algorithms/linfa-hierarchical/Cargo.toml b/algorithms/linfa-hierarchical/Cargo.toml
@@ -24,7 +24,7 @@ ndarray = { version = "0.16" }
 kodama = "0.2"
 thiserror = "2"
 
-linfa = { version = "0.8.0", path = "../.." }
+linfa = { version = "0.8.1", path = "../.." }
 linfa-kernel = { version = "0.8.0", path = "../linfa-kernel" }
 
 [dev-dependencies]

diff --git a/algorithms/linfa-ica/Cargo.toml b/algorithms/linfa-ica/Cargo.toml
@@ -34,13 +34,13 @@ num-traits = "0.2"
 rand_xoshiro = "0.6"
 thiserror = "2.0"
 
-linfa = { version = "0.8.0", path = "../.." }
+linfa = { version = "0.8.1", path = "../.." }
 
 [dev-dependencies]
 ndarray-npy = { version = "0.9", default-features = false }
 paste = "1.0"
 criterion = "0.5"
-linfa = { version = "0.8.0", path = "../..", features = ["benchmarks"] }
+linfa = { version = "0.8.1", path = "../..", features = ["benchmarks"] }
 
 [[bench]]
 name = "fast_ica"

diff --git a/algorithms/linfa-kernel/Cargo.toml b/algorithms/linfa-kernel/Cargo.toml
@@ -28,5 +28,5 @@ ndarray = "0.16"
 num-traits = "0.2"
 sprs = { version = "=0.11.2", default-features = false }
 
-linfa = { version = "0.8.0", path = "../.." }
+linfa = { version = "0.8.1", path = "../.." }
 linfa-nn = { version = "0.8.0", path = "../linfa-nn" }
diff --git a/algorithms/linfa-lars/Cargo.toml b/algorithms/linfa-lars/Cargo.toml
@@ -25,7 +25,7 @@ default-features = false
 features = ["std", "derive"]
 
 [dependencies]
-linfa = { version = "0.8.0", path = "../.." }
+linfa = { version = "0.8.1", path = "../.." }
 linfa-linalg = { version = "0.2", default-features = false }
 linfa-preprocessing = { version = "0.8.0" , path = "../linfa-preprocessing"}
 ndarray = { version = "0.16", features = ["approx"] }
@@ -39,4 +39,4 @@ linfa-datasets = { version = "0.8.0", path = "../../datasets", features = [
 ] }
 approx = "0.5"
 ndarray-rand = "0.15"
-rand_xoshiro = "0.6"
+rand_xoshiro = "0.6"
diff --git a/algorithms/linfa-linear/Cargo.toml b/algorithms/linfa-linear/Cargo.toml
@@ -37,7 +37,7 @@ argmin = { version = "0.11.0", default-features = false }
 argmin-math = { version = "0.5", features = ["ndarray_v0_16-nolinalg"] }
 thiserror = "2.0"
 
-linfa = { version = "0.8.0", path = "../.." }
+linfa = { version = "0.8.1", path = "../.." }
 
 [dev-dependencies]
 linfa-datasets = { version = "0.8.0", path = "../../datasets", features = [
@@ -46,7 +46,7 @@ linfa-datasets = { version = "0.8.0", path = "../../datasets", features = [
 approx = "0.5"
 criterion = "0.5"
 statrs = "0.18"
-linfa = { version = "0.8.0", path = "../..", features = ["benchmarks"] }
+linfa = { version = "0.8.1", path = "../..", features = ["benchmarks"] }
 
 [[bench]]
 name = "ols_bench"

diff --git a/algorithms/linfa-logistic/Cargo.toml b/algorithms/linfa-logistic/Cargo.toml
@@ -30,7 +30,7 @@ argmin = { version = "0.11.0", default-features = false }
 argmin-math = { version = "0.5", features = ["ndarray_v0_16-nolinalg"] }
 thiserror = "2.0"
 
-linfa = { version = "0.8.0", path = "../.." }
+linfa = { version = "0.8.1", path = "../.." }
 
 [dev-dependencies]
 approx = "0.5"

diff --git a/algorithms/linfa-nn/Cargo.toml b/algorithms/linfa-nn/Cargo.toml
@@ -33,14 +33,14 @@ thiserror = "2.0"
 
 kdtree = "0.7.0"
 
-linfa = { version = "0.8.0", path = "../.." }
+linfa = { version = "0.8.1", path = "../.." }
 
 [dev-dependencies]
 approx = "0.5"
 criterion = "0.5"
 rand_xoshiro = "0.6"
 ndarray-rand = "0.15"
-linfa = { version = "0.8.0", path = "../..", features = ["benchmarks"] }
+linfa = { version = "0.8.1", path = "../..", features = ["benchmarks"] }
 
 [[bench]]
 name = "nn"

diff --git a/algorithms/linfa-pls/Cargo.toml b/algorithms/linfa-pls/Cargo.toml
@@ -33,10 +33,10 @@ ndarray-rand = "0.15"
 num-traits = "0.2"
 paste = "1.0"
 thiserror = "2.0"
-linfa = { version = "0.8.0", path = "../.." }
+linfa = { version = "0.8.1", path = "../.." }
 
 [dev-dependencies]
-linfa = { version = "0.8.0", path = "../..", features = ["benchmarks"] }
+linfa = { version = "0.8.1", path = "../..", features = ["benchmarks"] }
 linfa-datasets = { version = "0.8.0", path = "../../datasets", features = [
     "linnerud",
 ] }

diff --git a/algorithms/linfa-preprocessing/Cargo.toml b/algorithms/linfa-preprocessing/Cargo.toml
@@ -18,7 +18,7 @@ blas = ["ndarray-linalg", "linfa/ndarray-linalg"]
 serde = ["serde_crate", "ndarray/serde", "serde_regex"]
 
 [dependencies]
-linfa = { version = "0.8.0", path = "../.." }
+linfa = { version = "0.8.1", path = "../.." }
 ndarray = { version = "0.16", features = ["approx"] }
 ndarray-linalg = { version = "0.17", optional = true }
 linfa-linalg = { version = "0.2", default-features = false }
@@ -52,7 +52,7 @@ iai = "0.1"
 curl = "0.4.35"
 flate2 = "1.0.20"
 tar = "0.4.33"
-linfa = { version = "0.8.0", path = "../..", features = ["benchmarks"] }
+linfa = { version = "0.8.1", path = "../..", features = ["benchmarks"] }
 criterion = "0.5"
 statrs = "0.18"
 

diff --git a/algorithms/linfa-reduction/Cargo.toml b/algorithms/linfa-reduction/Cargo.toml
@@ -42,7 +42,7 @@ num-traits = "0.2"
 thiserror = "2.0"
 rand = { version = "0.8", features = ["small_rng"] }
 
-linfa = { version = "0.8.0", path = "../.." }
+linfa = { version = "0.8.1", path = "../.." }
 linfa-kernel = { version = "0.8.0", path = "../linfa-kernel" }
 sprs = { version = "=0.11.2" }
 rand_xoshiro = "0.6.0"

diff --git a/algorithms/linfa-svm/Cargo.toml b/algorithms/linfa-svm/Cargo.toml
@@ -29,7 +29,7 @@ ndarray-rand = "0.15"
 num-traits = "0.2"
 thiserror = "2.0"
 
-linfa = { version = "0.8.0", path = "../.." }
+linfa = { version = "0.8.1", path = "../.." }
 linfa-kernel = { version = "0.8.0", path = "../linfa-kernel" }
 
 [dev-dependencies]

diff --git a/algorithms/linfa-trees/Cargo.toml b/algorithms/linfa-trees/Cargo.toml
@@ -27,7 +27,7 @@ features = ["std", "derive"]
 ndarray = { version = "0.16", features = ["rayon", "approx"] }
 ndarray-rand = "0.15"
 
-linfa = { version = "0.8.0", path = "../.." }
+linfa = { version = "0.8.1", path = "../.." }
 
 [dev-dependencies]
 rand = { version = "0.8", features = ["small_rng"] }
@@ -36,7 +36,7 @@ approx = "0.5"
 linfa-datasets = { version = "0.8.0", path = "../../datasets/", features = [
     "iris",
 ] }
-linfa = { version = "0.8.0", path = "../..", features = ["benchmarks"] }
+linfa = { version = "0.8.1", path = "../..", features = ["benchmarks"] }
 
 [[bench]]
 name = "decision_tree"

diff --git a/algorithms/linfa-tsne/Cargo.toml b/algorithms/linfa-tsne/Cargo.toml
@@ -20,7 +20,7 @@ ndarray-rand = "0.15"
 bhtsne = "0.4.0"
 pdqselect = "=0.1.1"
 
-linfa = { version = "0.8.0", path = "../.." }
+linfa = { version = "0.8.1", path = "../.." }
 
 [dev-dependencies]
 rand = "0.8"

diff --git a/datasets/Cargo.toml b/datasets/Cargo.toml
@@ -8,7 +8,7 @@ license = "MIT OR Apache-2.0"
 repository = "https://github.com/rust-ml/linfa"
 
 [dependencies]
-linfa = { version = "0.8.0", path = ".." }
+linfa = { version = "0.8.1", path = ".." }
 ndarray = { version = "0.16" }
 ndarray-csv = "=0.5.3"
 csv = "1.1"

diff --git a/docs/website/content/news/release_081.md b/docs/website/content/news/release_081.md
@@ -0,0 +1,35 @@
++++
+title = "Release 0.8.1"
+date = "2025-12-23"
++++
+
+`linfa 0.8.1` adds random forest, and AdaBoost boosting algorithms in `linfa-ensemble` and least angle regression in a new `linfa-lars` crate.
+
+<!-- more -->
+
+## Improvements and fixes
+
+* `linfa-ica`: fix missing exponential by @lmmx in https://github.com/rust-ml/linfa/pull/426
+* `linfa`:
+  * add bootstrap-with-indices utilities for `Dataset` 
+  * fix ndarray 0.16/0.17 versions mismatch (0.17 not supported) 
+
+## New algorithms
+
+### Least Angle regression
+
+Least Angle Regression (LARS) is an algorithm used in regression for high dimensional data.
+
+See [sklearn.linear_model](https://scikit-learn.org/stable/modules/linear_model.html#least-angle-regression)
+
+### Ensemble methods
+
+Ensemble methods combine the predictions of several base estimators built with a given learning algorithm in order to improve generalizability / robustness over a single estimator.
+
+#### Random Forest
+
+See [sklearn.ensemble forest](https://scikit-learn.org/stable/modules/ensemble.html#forest)
+
+#### AdaBoost 
+
+See [sklearn.ensemble AdaBoost](https://scikit-learn.org/stable/modules/ensemble.html#adaboost)