Struct FewShotLearner 

pub struct FewShotLearner { /* private fields */ }

Few-shot learner for quick adaptation.

Given a small number of examples, adapts quickly using meta-learned priors.

Implementations

impl FewShotLearner

pub fn from_meta(meta: &MetaLearner, task_embedding: Option<&[f32]>) -> Self

Create from meta-learner initialization.

pub fn with_adapt_lr(self, lr: f32) -> Self

Set adaptation learning rate.

pub fn with_adapt_steps(self, steps: usize) -> Self

Set number of adaptation steps.

pub fn add_example(&mut self, features: Vec<f32>, target: f32)

Add an example to the support set.

pub fn adapt(&mut self)

Adapt to the support set.

Performs gradient descent on support set to adapt parameters from meta-learned initialization.

pub fn predict(&self, features: &[f32]) -> f32

Predict for new input.

pub fn get_adapted_params(&self) -> &[f32]

Get final adapted parameters.

pub fn support_size(&self) -> usize

Number of support examples.

Trait Implementations

impl Clone for FewShotLearner

fn clone(&self) -> FewShotLearner

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for FewShotLearner

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for FewShotLearner

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for FewShotLearner

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.