Struct linfa::composing::platt_scaling::Platt

pub struct Platt<F, O> { /* private fields */ }

Fitted Platt model

This model contains a sigmoid scaling parameters and a second univariate, uncalibrated model. The output of the uncalibrated model is scaled with the following function:

g(x) = 1 / (1 + exp(A * f(x) + B)

The scaling factors A and B are estimated with the Newton’s method, presented in the following paper: https://www.csie.ntu.edu.tw/~cjlin/papers/plattprob.pdf

Implementations

impl<F: Float, O> Platt<F, O>

pub fn params() -> PlattParams<F, O>

Create default parameter set for the Platt scaling algorithm

The default values are:

• maxiter: 100,
• minstep: 1e-10,
• sigma: 1e-12

Trait Implementations

impl<F: Clone, O: Clone> Clone for Platt<F, O>

fn clone(&self) -> Platt<F, O>

Returns a copy of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl<F: Debug, O: Debug> Debug for Platt<F, O>

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl<F: PartialEq, O: PartialEq> PartialEq<Platt<F, O>> for Platt<F, O>

fn eq(&self, other: &Platt<F, O>) -> bool

This method tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

This method tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl<F: Float, D, O> PredictInplace<ArrayBase<D, Dim<[usize; 2]>>, ArrayBase<OwnedRepr<Pr>, Dim<[usize; 1]>>> for Platt<F, O>where
    D: Data<Elem = F>,
    O: PredictInplace<ArrayBase<D, Ix2>, ArrayBase<D, Ix1>>,

fn predict_inplace(&self, data: &ArrayBase<D, Ix2>, targets: &mut Array1<Pr>)

Predict something in place

fn default_target(&self, x: &ArrayBase<D, Ix2>) -> Array1<Pr>

Create targets that predict_inplace works with.

impl<F: Eq, O: Eq> Eq for Platt<F, O>

impl<F, O> StructuralEq for Platt<F, O>

impl<F, O> StructuralPartialEq for Platt<F, O>