How to Build a Machine Learning Pipeline in Rust

Performance
2026-04-17T12:05:42+00:00
Build a Rust ML pipeline using `linfa` and `ndarray` to load data, train a model, and generate predictions in a single workflow.

Build a machine learning pipeline in Rust by defining data structures, loading data, training a model, and evaluating results using the ndarray and linfa crates.

use ndarray::{Array1, Array2};
use linfa::prelude::*;
use linfa_linalg::svd::SVD;

fn main() {
    let data = Array2::from_shape_vec((4, 3), vec![1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0, 11.0, 12.0]).unwrap();
    let labels = Array1::from_vec(vec![0, 1, 0, 1]);
    let dataset = linfa::Dataset::new(data, labels);
    let model = linfa_logistic::LogisticRegression::params().fit(&dataset).unwrap();
    let predictions = model.predict(&data);
    println!("Predictions: {:?}", predictions);
}

Add ndarray = "0.15" and linfa = "0.8" to your Cargo.toml dependencies.