Code shared from the Rust Playground

playground.rs

use std::collections::HashMap;

/// Calculates Shannon Entropy for a slice of data.
/// H = -sum(p(x) * log2(p(x)))
fn calculate_entropy(data: &[&str]) -> f64 {
    let len = data.len() as f64;
    if len == 0.0 {
        return 0.0;
    }

    // 1. Calculate frequencies of each unique element
    let mut counts = HashMap::new();
    for &item in data {
        *counts.entry(item).or_insert(0) += 1;
    }

    // 2. Apply the Shannon Entropy formula
    counts.values().fold(0.0, |acc, &count| {
        let p_x = count as f64 / len;
        // Entropy is usually measured in "bits" using log base 2
        acc - (p_x * p_x.log2())
    })
}

fn main() {
    // Example 1: High certainty (Low entropy)
    let certain_data = vec!["apple", "apple", "apple"];
    
    // Example 2: High uncertainty (High entropy)
    let uncertain_data = vec!["apple", "banana", "cherry"];

    println!("Entropy (Certain): {:.4}", calculate_entropy(&certain_data));
    println!("Entropy (Uncertain): {:.4}", calculate_entropy(&uncertain_data));
}