https://github.com/eigenein/rust-hyperopt

Tree-of-Parzen-estimators hyperparameter optimization
https://github.com/eigenein/rust-hyperopt

hyperopt hyperparameter-optimization hyperparameter-search hyperparameter-tuning kernel-density-estimation machine-learning rust-crate tree-of-parzen-estimator

Last synced: 2 days ago
JSON representation

Tree-of-Parzen-estimators hyperparameter optimization

• Host: GitHub
• URL: https://github.com/eigenein/rust-hyperopt
• Owner: eigenein
• License: mit
• Created: 2024-04-09T20:39:27.000Z (9 months ago)
• Default Branch: main
• Last Pushed: 2024-04-17T12:44:22.000Z (8 months ago)
• Last Synced: 2024-05-02T01:27:19.776Z (8 months ago)
• Topics: hyperopt, hyperparameter-optimization, hyperparameter-search, hyperparameter-tuning, kernel-density-estimation, machine-learning, rust-crate, tree-of-parzen-estimator
• Language: Rust
• Homepage: https://crates.io/crates/hyperopt
• Size: 108 KB
• Stars: 0
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • Contributing: CONTRIBUTING.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# `hyperopt`

Tree-of-Parzen-estimators hyperparameter optimization for Rust

[![Documentation](https://img.shields.io/docsrs/hyperopt?style=for-the-badge)

](https://docs.rs/hyperopt)

[![Check status](https://img.shields.io/github/actions/workflow/status/eigenein/rust-hyperopt/check.yaml?style=for-the-badge)]((https://github.com/eigenein/rust-hyperopt/actions/workflows/check.yaml))

[![Code coverage](https://img.shields.io/codecov/c/github/eigenein/rust-hyperopt?style=for-the-badge)

](https://app.codecov.io/gh/eigenein/rust-hyperopt)

![Maintenance](https://img.shields.io/maintenance/yes/2024?style=for-the-badge)

## Examples

### Continuous

```rust

use std::f64::consts::{FRAC_PI_2, PI};

use approx::assert_abs_diff_eq;

use fastrand::Rng;

use ordered_float::NotNan;

use hyperopt::Optimizer;

use hyperopt::kernel::continuous::Epanechnikov;

use hyperopt::kernel::universal::Uniform;

fn main() {

    let min = NotNan::new(FRAC_PI_2).unwrap();

    let max = NotNan::new(PI + FRAC_PI_2).unwrap();

    let mut optimizer = Optimizer::new(

        min..=max,                       // parameter search limits

        Uniform::with_bounds(min..=max), // our initial guess is just as bad

        Rng::with_seed(42),

    );

    // Run 50 trials for the cosine function and try to find the point `(π, -1)`:

    for _ in 0..50 {

        // Generate new trials using Epanechnikov kernel with `>`

        // as both parameter and density:

        let x = optimizer.new_trial::>>();

        

        // Tell the optimizer the result of evaluation:

        optimizer.feed_back(x, NotNan::new(x.cos()).unwrap());

    }

    let best_trial = optimizer.best_trial().unwrap();

    assert_abs_diff_eq!(best_trial.parameter.into_inner(), PI, epsilon = 0.05);

    assert_abs_diff_eq!(best_trial.metric.into_inner(), -1.0, epsilon = 0.01);

}

```

### Discrete

```rust

use fastrand::Rng;

use ordered_float::OrderedFloat;

use hyperopt::Optimizer;

use hyperopt::kernel::discrete::Binomial;

use hyperopt::kernel::universal::Uniform;

fn main() {

    let mut optimizer = Optimizer::new(

        -100..=100,

        Uniform::with_bounds(-100..=100),

        Rng::with_seed(42),

    );

    for _ in 0..30 {

        // Use the binomial kernel for `i32` as parameter

        // and `OrderedFloat` as density:

        let x = optimizer.new_trial::>>();

        

        // Optimize the parabola: https://www.wolframalpha.com/input?i=x%5E2+-+4x

        optimizer.feed_back(x, x * x - 4 * x);

    }

    let best_trial = optimizer.best_trial().unwrap();

    assert_eq!(best_trial.parameter, 2);

    assert_eq!(best_trial.metric, -4);

}

```

## Features

- `ordered-float` enables support for `OrderedFloat` and `NotNan` types