Struct FnAcceleratedModel 

pub struct FnAcceleratedModel<F> { /* private fields */ }

Adapts a device-aware prediction closure into an accelerated model.

Ordinary Predict calls use ExecutionDevice::Cpu; bind another device with DeviceModel when passing the model to an explainer.

Implementations

impl<F> FnAcceleratedModel<F>

pub fn new(predict_fn: F) -> Self

pub fn with_n_features(self, n_features: usize) -> Self

pub fn with_n_outputs(self, n_outputs: usize) -> Self

Trait Implementations

impl<F> AcceleratedPredict for FnAcceleratedModel<F>

where F: Fn(ArrayView2<'_, f64>, ExecutionDevice) -> Result<Array2<f64>>,

fn predict_on( &self, x: ArrayView2<'_, f64>, device: ExecutionDevice, ) -> Result<Array2<f64>>

fn predict_owned_on( &self, x: Array2<f64>, device: ExecutionDevice, ) -> Result<Array2<f64>>

Owned-batch device prediction fast path.

impl<F> Predict for FnAcceleratedModel<F>

where F: Fn(ArrayView2<'_, f64>, ExecutionDevice) -> Result<Array2<f64>>,

fn predict(&self, x: ArrayView2<'_, f64>) -> Result<Array2<f64>>

Predict outputs for a batch of samples.

fn n_features(&self) -> Option<usize>

Returns the number of input features expected by the model.

fn n_outputs(&self) -> Option<usize>

Returns the number of outputs produced by the model.

fn predict_owned(&self, x: Array2<f64>) -> Result<Array2<f64>>

Predicts from an owned batch.