Crate numra_fit 

Numra Curve Fitting

Nonlinear curve fitting, polynomial fitting, and evaluation.

This crate provides tools for fitting parametric models to data, built on top of numra-optim’s Levenberg-Marquardt solver and numra-linalg’s SVD-based least squares.

Key functions

• curve_fit() - Fit any y = f(x, params) model to data
• curve_fit_weighted() - Weighted nonlinear least squares
• polyfit() - Fit polynomial of given degree
• polyval() - Evaluate polynomial at given points

Example

use numra_fit::{curve_fit, polyfit, polyval};

// Fit an exponential decay: y = a * exp(-b * x)
let x = vec![0.0, 1.0, 2.0, 3.0, 4.0, 5.0];
let y = vec![5.0, 3.03, 1.84, 1.11, 0.67, 0.41];

let result = curve_fit(
    |x: f64, p: &[f64]| p[0] * (-p[1] * x).exp(),
    &x, &y, &[4.0, 0.3], None,
).unwrap();

println!("a = {:.3}, b = {:.3}", result.params[0], result.params[1]);
println!("R^2 = {:.6}", result.r_squared);

Author: Moussa Leblouba Date: 9 February 2026 Modified: 2 May 2026

Re-exports

pub use curve_fit::curve_fit;

pub use curve_fit::curve_fit_weighted;

pub use curve_fit::curve_fit_with_jacobian;

pub use error::FitError;

pub use polynomial::polyfit;

pub use polynomial::polyval;

pub use types::FitOptions;

pub use types::FitResult;

Modules

curve_fit

Nonlinear curve fitting via Levenberg-Marquardt.

error

Error types for curve fitting.

polynomial

Polynomial fitting and evaluation.

types

Types for curve fitting results and options.