README.md

Discontinued. The author switched to Tensorflow for personal projects.

Autograd

Rust automatic differentiation library to compute gradient values. It is mainly for nonlinear optimization and machine learning. It is alpha stage yet.

Example

#![feature(thread_local)]
#![feature(std_misc)]
#![feature(alloc)]

#[macro_use]
extern crate autograd;

use autograd::Context;

fn main() {
    // Initialize Autograd context with type f32 and capacity 100.
    let context = new_autograd_context!(f32, 100);

    // Initializes input variables.
    let x1 = context.new_variable(1.5);
    let x2 = context.new_variable(2.0);

    // Computes a math expression.
    let y = (x1 * x2) + x1 + 5.0;
    println!("y   == {}", y.value);

    // Computes gradient with respect to y.
    context.differentiate(y);
    println!("dx1 == {}", context.get_derivative(x1));
    println!("dx2 == {}", context.get_derivative(x2));
}

/* Output
y   == 9.5
dx1 == 3
dx2 == 1.5
/*