Struct friedrich::gaussian_process::GaussianProcessBuilder

pub struct GaussianProcessBuilder<KernelType: Kernel, PriorType: Prior> { /* fields omitted */ }

Builder to set the parameters of a gaussian process.

This class is meant to be produced by the builder method of the gaussian process and can be used to select the various parameters of the gaussian process :

// training data
let training_inputs = vec![vec![0.8], vec![1.2], vec![3.8], vec![4.2]];
let training_outputs = vec![3.0, 4.0, -2.0, -2.0];

// model parameters
let input_dimension = 1;
let output_noise = 0.1;
let exponential_kernel = Exponential::default();
let linear_prior = LinearPrior::default(input_dimension);

// defining and training a model
let gp = GaussianProcess::builder(training_inputs, training_outputs).set_noise(output_noise)
                                                                    .set_kernel(exponential_kernel)
                                                                    .fit_kernel()
                                                                    .set_prior(linear_prior)
                                                                    .fit_prior()
                                                                    .train();

Methods

impl<KernelType: Kernel, PriorType: Prior> GaussianProcessBuilder<KernelType, PriorType>

pub fn new<T: Input>(training_inputs: T, training_outputs: T::InVector) -> Self

builds a new gaussian process with default parameters

the defaults are :

• constant prior (0 unless fitted)
• a gaussian kernel
• a noise of 1% of the output standard deviation (might be re-fitted in the absence of user provided value)
• does not fit parameters

pub fn set_prior<NewPriorType: Prior>(
    self,
    prior: NewPriorType
) -> GaussianProcessBuilder<KernelType, NewPriorType>

Sets a new prior. See the documentation on priors for more informations.

pub fn set_noise(self, noise: f64) -> Self

Sets the noise parameter. It correspond to the standard deviation of the noise in the outputs of the training set.

pub fn set_kernel<NewKernelType: Kernel>(
    self,
    kernel: NewKernelType
) -> GaussianProcessBuilder<NewKernelType, PriorType>

Changes the kernel of the gaussian process. See the documentations on Kernels for more informations.

pub fn set_fit_parameters(
    self,
    max_iter: usize,
    convergence_fraction: f64
) -> Self

Modifies the stopping criteria of the gradient descent used to fit the noise and kernel parameters.

The optimizer runs for a maximum of max_iter iterations and stops prematurely if all gradients are below convergence_fraction time their associated parameter.

pub fn fit_kernel(self) -> Self

Asks for the parameters of the kernel to be fitted on the training data. The fitting will be done when the train method is called.

pub fn fit_prior(self) -> Self

Asks for the prior to be fitted on the training data. The fitting will be done when the train method is called.

pub fn fit_noise(self) -> Self

Asks for the noise to be fitted on the training data. The fitting will be done when the train method is called.

pub fn train(self) -> GaussianProcess<KernelType, PriorType>

Trains the gaussian process. Fits the parameters if requested.