aprender-core 0.31.2

Next-generation machine learning library in pure Rust
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77

#![allow(clippy::disallowed_methods)]
//! Boston Housing example - Linear Regression
//!
//! Demonstrates linear regression using simulated housing data.

use aprender::prelude::*;

fn main() {
    println!("Boston Housing - Linear Regression Example");
    println!("==========================================\n");

    // Simulated housing data
    // Features: [sqft, bedrooms, bathrooms, age]
    // Target: price (in thousands)
    let x = Matrix::from_vec(
        10,
        4,
        vec![
            1500.0, 3.0, 2.0, 10.0, 2000.0, 4.0, 2.5, 5.0, 1200.0, 2.0, 1.0, 30.0, 1800.0, 3.0,
            2.0, 15.0, 2500.0, 5.0, 3.0, 2.0, 1000.0, 2.0, 1.0, 50.0, 2200.0, 4.0, 3.0, 8.0,
            1600.0, 3.0, 2.0, 20.0, 3000.0, 5.0, 4.0, 1.0, 1400.0, 3.0, 1.5, 25.0,
        ],
    )
    .expect("Example data should be valid");

    // Prices (simulated based on features)
    let y = Vector::from_slice(&[
        250.0, 350.0, 180.0, 280.0, 450.0, 150.0, 380.0, 260.0, 520.0, 220.0,
    ]);

    // Train linear regression model
    println!("Training linear regression model...");
    let mut model = LinearRegression::new();
    model.fit(&x, &y).expect("Failed to fit model");

    // Print coefficients
    let coef = model.coefficients();
    println!("\nModel Coefficients:");
    println!("  sqft:      {:.4}", coef[0]);
    println!("  bedrooms:  {:.4}", coef[1]);
    println!("  bathrooms: {:.4}", coef[2]);
    println!("  age:       {:.4}", coef[3]);
    println!("  intercept: {:.4}", model.intercept());

    // Make predictions
    let predictions = model.predict(&x);

    // Print predictions vs actual
    println!("\nPredictions vs Actual:");
    println!("{:>10} {:>10} {:>10}", "Actual", "Predicted", "Error");
    println!("{}", "-".repeat(32));

    for i in 0..y.len() {
        let actual = y.as_slice()[i];
        let predicted = predictions.as_slice()[i];
        let error = actual - predicted;
        println!("{actual:>10.1} {predicted:>10.1} {error:>10.1}");
    }

    // Calculate metrics
    let r2 = model.score(&x, &y);
    let mse_val = mse(&predictions, &y);
    let mae_val = mae(&predictions, &y);

    println!("\nModel Performance:");
    println!("  R² Score: {r2:.4}");
    println!("  MSE:      {mse_val:.4}");
    println!("  MAE:      {mae_val:.4}");

    // Predict on new house
    let new_house =
        Matrix::from_vec(1, 4, vec![1900.0, 4.0, 2.0, 12.0]).expect("Example data should be valid");
    let predicted_price = model.predict(&new_house);
    println!("\nNew House Prediction:");
    println!("  Features: 1900 sqft, 4 bed, 2 bath, 12 years old");
    println!("  Predicted Price: ${:.0}k", predicted_price.as_slice()[0]);
}