[Feature] Add Least Angle Regression (#421)

* Implement Least Angle Regression

* cargo fmt

* cargo clippy

* cargo clippy 2

* improve documentation

* remove linnerud dependency

* add example

* cargo clippy in doc test

* Apply suggestions from code review

Co-authored-by: Rémi Lafage <[email protected]>

* close bracket

* remove unwanted stride handling

* correct doc example

---------

Co-authored-by: Rémi Lafage <[email protected]>

Author: aswin-1111

algorithms/linfa-lars/Cargo.toml

@@ -0,0 +1,42 @@
+[package]
+name = "linfa-lars"
+version = "0.8.0"
+authors = ["Aswin <[email protected]>"]
+description = "Least Angle Regression methods"
+edition = "2024"
+license = "MIT OR Apache-2.0"
+
+repository = "https://github.com/rust-ml/linfa"
+readme = "README.md"
+
+keywords = ["machine-learning", "linfa", "ai", "ml", "linear"]
+categories = ["algorithms", "mathematics", "science"]
+
+[features]
+default = []
+serde = ["serde_crate", "ndarray/serde", "linfa/serde"]
+blas = ["ndarray-linalg", "linfa/ndarray-linalg"]
+
+[dependencies.serde_crate]
+package = "serde"
+optional = true
+version = "1.0"
+default-features = false
+features = ["std", "derive"]
+
+[dependencies]
+linfa = { version = "0.8.0", path = "../.." }
+linfa-linalg = { version = "0.2", default-features = false }
+linfa-preprocessing = { version = "0.8.0" , path = "../linfa-preprocessing"}
+ndarray = { version = "0.16", features = ["approx"] }
+ndarray-linalg = { version = "0.17", optional = true }
+ndarray-stats = "0.6"
+thiserror = "2.0"
+
+[dev-dependencies]
+linfa-datasets = { version = "0.8.0", path = "../../datasets", features = [
+    "diabetes",
+] }
+approx = "0.5"
+ndarray-rand = "0.15"
+rand_xoshiro = "0.6"

algorithms/linfa-lars/README.md

@@ -0,0 +1,32 @@
+# Least Angle Regression
+
+`linfa-lars` aims to provide pure Rust implementation of least-angle regression algorithm. 
+
+## The Big Picture
+
+`linfa-lars` is a crate in the [`linfa`](https://crates.io/crates/linfa) ecosystem, an effort to create a toolkit for classical Machine Learning implemented in pure Rust, akin to Python's `scikit-learn`.
+
+## Current state
+
+The `linfa-lars` crate currently provides an implementation of the Least-angle regression (LARS) algorithm
+
+See also:
+ * [Wikipedia on Least Angle Regression](https://en.wikipedia.org/wiki/Least-angle_regression)
+
+
+## BLAS/Lapack backend
+
+See [this section](../../README.md#blaslapack-backend) to enable an external BLAS/LAPACK backend.
+
+## Examples
+
+There is an usage example in the `examples/` directory. To run, use:
+
+```bash
+$ cargo run --example lars
+```
+
+## License
+Dual-licensed to be compatible with the Rust project.
+
+Licensed under the Apache License, Version 2.0 <http://www.apache.org/licenses/LICENSE-2.0> or the MIT license <http://opensource.org/licenses/MIT>, at your option. This file may not be copied, modified, or distributed except according to those terms.

algorithms/linfa-lars/examples/lars.rs

@@ -0,0 +1,20 @@
+use linfa::prelude::*;
+use linfa_lars::Lars;
+
+fn main() {
+    // load Diabetes dataset
+    let (train, valid) = linfa_datasets::diabetes().split_with_ratio(0.90);
+
+    let model = Lars::params()
+        .fit_intercept(true)
+        .verbose(true)
+        .fit(&train)
+        .unwrap();
+
+    println!("hyperplane:  {}", model.hyperplane());
+    println!("intercept:  {}", model.intercept());
+
+    // validate
+    let y_est = model.predict(&valid);
+    println!("predicted variance: {}", valid.r2(&y_est).unwrap());
+}