Struct nncombinator::device::DeviceCpu

pub struct DeviceCpu<U>where
    U: UnitValue<U>,{ /* private fields */ }

Implementation of Device to be computed by CPU

Implementations

impl<U> DeviceCpu<U>where
    U: UnitValue<U>,

pub fn new() -> Result<DeviceCpu<U>, DeviceError>

note: For the sake of implementation uniformity, DeviceCpu::new is defined as if it may return a DeviceError of type Result, but this error is never actually returned.

Trait Implementations

impl<U, const N: usize> Activation<U, Arr<U, N>, Arr<U, N>, DeviceCpu<U>> for Identity<U, DeviceCpu<U>>where
    U: UnitValue<U>,

fn apply(
    &self,
    _: &DeviceCpu<U>,
    input: &Arr<U, N>
) -> Result<Arr<U, N>, EvaluateError>

Apply the activation function

fn derive(
    &self,
    _: &DeviceCpu<U>,
    _: &Arr<U, N>,
    loss: &Arr<U, N>,
    _: &Arr<U, N>
) -> Result<Arr<U, N>, TrainingError>

Apply derivatives of the activation function

fn is_canonical_link<L: LossFunction<U>>(&self, l: &L) -> bool

Returns whether or not the canonical linkage function can be used.

impl<U, const N: usize> Activation<U, Arr<U, N>, Arr<U, N>, DeviceCpu<U>> for ReLu<U, DeviceCpu<U>>where
    U: UnitValue<U>,

fn apply(
    &self,
    device: &DeviceCpu<U>,
    input: &Arr<U, N>
) -> Result<Arr<U, N>, EvaluateError>

Apply the activation function

fn derive(
    &self,
    device: &DeviceCpu<U>,
    o: &Arr<U, N>,
    loss: &Arr<U, N>,
    u: &Arr<U, N>
) -> Result<Arr<U, N>, TrainingError>

Apply derivatives of the activation function

fn is_canonical_link<L: LossFunction<U>>(&self, _: &L) -> bool

Returns whether or not the canonical linkage function can be used.

impl<U, const N: usize> Activation<U, Arr<U, N>, Arr<U, N>, DeviceCpu<U>> for Sigmoid<U, DeviceCpu<U>>where
    U: UnitValue<U>,

fn apply(
    &self,
    device: &DeviceCpu<U>,
    input: &Arr<U, N>
) -> Result<Arr<U, N>, EvaluateError>

Apply the activation function

fn derive(
    &self,
    device: &DeviceCpu<U>,
    o: &Arr<U, N>,
    loss: &Arr<U, N>,
    u: &Arr<U, N>
) -> Result<Arr<U, N>, TrainingError>

Apply derivatives of the activation function

fn is_canonical_link<L: LossFunction<U>>(&self, l: &L) -> bool

Returns whether or not the canonical linkage function can be used.

impl<U, const N: usize> Activation<U, Arr<U, N>, Arr<U, N>, DeviceCpu<U>> for SoftMax<U, DeviceCpu<U>>where
    U: UnitValue<U>,

fn apply(
    &self,
    device: &DeviceCpu<U>,
    input: &Arr<U, N>
) -> Result<Arr<U, N>, EvaluateError>

Apply the activation function

fn derive(
    &self,
    device: &DeviceCpu<U>,
    o: &Arr<U, N>,
    loss: &Arr<U, N>,
    u: &Arr<U, N>
) -> Result<Arr<U, N>, TrainingError>

Apply derivatives of the activation function

fn is_canonical_link<L: LossFunction<U>>(&self, l: &L) -> bool

Returns whether or not the canonical linkage function can be used.

impl<U, const N: usize> Activation<U, Arr<U, N>, Arr<U, N>, DeviceCpu<U>> for Swish<U, DeviceCpu<U>>where
    U: UnitValue<U>,

fn apply(
    &self,
    device: &DeviceCpu<U>,
    input: &Arr<U, N>
) -> Result<Arr<U, N>, EvaluateError>

Apply the activation function

fn derive(
    &self,
    device: &DeviceCpu<U>,
    o: &Arr<U, N>,
    loss: &Arr<U, N>,
    u: &Arr<U, N>
) -> Result<Arr<U, N>, TrainingError>

Apply derivatives of the activation function

fn is_canonical_link<L: LossFunction<U>>(&self, _: &L) -> bool

Returns whether or not the canonical linkage function can be used.

impl<U, const N: usize> Activation<U, Arr<U, N>, Arr<U, N>, DeviceCpu<U>> for Tanh<U, DeviceCpu<U>>where
    U: UnitValue<U>,

fn apply(
    &self,
    device: &DeviceCpu<U>,
    input: &Arr<U, N>
) -> Result<Arr<U, N>, EvaluateError>

Apply the activation function

fn derive(
    &self,
    device: &DeviceCpu<U>,
    o: &Arr<U, N>,
    loss: &Arr<U, N>,
    u: &Arr<U, N>
) -> Result<Arr<U, N>, TrainingError>

Apply derivatives of the activation function

fn is_canonical_link<L: LossFunction<U>>(&self, _: &L) -> bool

Returns whether or not the canonical linkage function can be used.

impl<U, I, const N: usize> Activation<U, I, Arr<U, N>, DeviceCpu<U>> for Identity<U, DeviceCpu<U>>where
    U: UnitValue<U>,
    I: IndexedParallelIterator<Item = U> + IntoParallelIterator + Clone,

fn apply(&self, _: &DeviceCpu<U>, input: &I) -> Result<Arr<U, N>, EvaluateError>

Apply the activation function

fn derive(
    &self,
    _: &DeviceCpu<U>,
    _: &I,
    loss: &I,
    _: &I
) -> Result<Arr<U, N>, TrainingError>

Apply derivatives of the activation function

fn is_canonical_link<L: LossFunction<U>>(&self, l: &L) -> bool

Returns whether or not the canonical linkage function can be used.

impl<U, I, const N: usize> Activation<U, I, Arr<U, N>, DeviceCpu<U>> for ReLu<U, DeviceCpu<U>>where
    U: UnitValue<U>,
    I: IndexedParallelIterator<Item = U> + Clone,
    <I as IntoParallelIterator>::Iter: IndexedParallelIterator<Item = U>,

fn apply(&self, _: &DeviceCpu<U>, input: &I) -> Result<Arr<U, N>, EvaluateError>

Apply the activation function

fn derive(
    &self,
    _: &DeviceCpu<U>,
    _: &I,
    loss: &I,
    u: &I
) -> Result<Arr<U, N>, TrainingError>

Apply derivatives of the activation function

fn is_canonical_link<L: LossFunction<U>>(&self, _: &L) -> bool

Returns whether or not the canonical linkage function can be used.

impl<U, I, const N: usize> Activation<U, I, Arr<U, N>, DeviceCpu<U>> for Sigmoid<U, DeviceCpu<U>>where
    U: UnitValue<U>,
    I: IndexedParallelIterator<Item = U> + Clone,
    <I as IntoParallelIterator>::Iter: IndexedParallelIterator<Item = U>,

fn apply(&self, _: &DeviceCpu<U>, input: &I) -> Result<Arr<U, N>, EvaluateError>

Apply the activation function

fn derive(
    &self,
    _: &DeviceCpu<U>,
    o: &I,
    loss: &I,
    _: &I
) -> Result<Arr<U, N>, TrainingError>

Apply derivatives of the activation function

fn is_canonical_link<L: LossFunction<U>>(&self, l: &L) -> bool

Returns whether or not the canonical linkage function can be used.

impl<U, I, const N: usize> Activation<U, I, Arr<U, N>, DeviceCpu<U>> for SoftMax<U, DeviceCpu<U>>where
    U: UnitValue<U>,
    I: IndexedParallelIterator<Item = U> + Clone,
    <I as IntoParallelIterator>::Iter: IndexedParallelIterator<Item = U>,

fn apply(&self, _: &DeviceCpu<U>, input: &I) -> Result<Arr<U, N>, EvaluateError>

Apply the activation function

fn derive(
    &self,
    _: &DeviceCpu<U>,
    o: &I,
    loss: &I,
    _: &I
) -> Result<Arr<U, N>, TrainingError>

Apply derivatives of the activation function

fn is_canonical_link<L: LossFunction<U>>(&self, l: &L) -> bool

Returns whether or not the canonical linkage function can be used.

impl<U, I, const N: usize> Activation<U, I, Arr<U, N>, DeviceCpu<U>> for Swish<U, DeviceCpu<U>>where
    U: UnitValue<U>,
    I: IndexedParallelIterator<Item = U> + Clone,
    <I as IntoParallelIterator>::Iter: IndexedParallelIterator<Item = U>,

fn apply(&self, _: &DeviceCpu<U>, input: &I) -> Result<Arr<U, N>, EvaluateError>

Apply the activation function

fn derive(
    &self,
    _: &DeviceCpu<U>,
    o: &I,
    loss: &I,
    u: &I
) -> Result<Arr<U, N>, TrainingError>

Apply derivatives of the activation function

fn is_canonical_link<L: LossFunction<U>>(&self, _: &L) -> bool

Returns whether or not the canonical linkage function can be used.

impl<U, I, const N: usize> Activation<U, I, Arr<U, N>, DeviceCpu<U>> for Tanh<U, DeviceCpu<U>>where
    U: UnitValue<U>,
    I: IndexedParallelIterator<Item = U> + Clone,
    <I as IntoParallelIterator>::Iter: IndexedParallelIterator<Item = U>,

fn apply(&self, _: &DeviceCpu<U>, input: &I) -> Result<Arr<U, N>, EvaluateError>

Apply the activation function

fn derive(
    &self,
    _: &DeviceCpu<U>,
    o: &I,
    loss: &I,
    _: &I
) -> Result<Arr<U, N>, TrainingError>

Apply derivatives of the activation function

fn is_canonical_link<L: LossFunction<U>>(&self, _: &L) -> bool

Returns whether or not the canonical linkage function can be used.

impl<U, const N: usize> BatchActivation<U, Arr<U, N>, Arr<U, N>, DeviceCpu<U>> for Identity<U, DeviceCpu<U>>where
    U: UnitValue<U>,
    Vec<Arr<U, N>>: FromParallelIterator<Arr<U, N>>,

fn batch_apply(
    &self,
    _: &DeviceCpu<U>,
    input: &VecArr<U, Arr<U, N>>
) -> Result<VecArr<U, Arr<U, N>>, TrainingError>

Apply the activation function

fn batch_derive(
    &self,
    _: &DeviceCpu<U>,
    _: &VecArr<U, Arr<U, N>>,
    loss: &VecArr<U, Arr<U, N>>,
    _: &VecArr<U, Arr<U, N>>
) -> Result<VecArr<U, Arr<U, N>>, TrainingError>

Apply derivatives of the activation function

impl<U, const N: usize> BatchActivation<U, Arr<U, N>, Arr<U, N>, DeviceCpu<U>> for ReLu<U, DeviceCpu<U>>where
    U: UnitValue<U>,
    Vec<Arr<U, N>>: FromParallelIterator<Arr<U, N>>,

fn batch_apply(
    &self,
    device: &DeviceCpu<U>,
    input: &VecArr<U, Arr<U, N>>
) -> Result<VecArr<U, Arr<U, N>>, TrainingError>

Apply the activation function

fn batch_derive(
    &self,
    device: &DeviceCpu<U>,
    o: &VecArr<U, Arr<U, N>>,
    loss: &VecArr<U, Arr<U, N>>,
    u: &VecArr<U, Arr<U, N>>
) -> Result<VecArr<U, Arr<U, N>>, TrainingError>

Apply derivatives of the activation function

impl<U, const N: usize> BatchActivation<U, Arr<U, N>, Arr<U, N>, DeviceCpu<U>> for Sigmoid<U, DeviceCpu<U>>where
    U: UnitValue<U>,
    Vec<Arr<U, N>>: FromParallelIterator<Arr<U, N>>,

fn batch_apply(
    &self,
    device: &DeviceCpu<U>,
    input: &VecArr<U, Arr<U, N>>
) -> Result<VecArr<U, Arr<U, N>>, TrainingError>

Apply the activation function

fn batch_derive(
    &self,
    device: &DeviceCpu<U>,
    o: &VecArr<U, Arr<U, N>>,
    loss: &VecArr<U, Arr<U, N>>,
    u: &VecArr<U, Arr<U, N>>
) -> Result<VecArr<U, Arr<U, N>>, TrainingError>

Apply derivatives of the activation function

impl<U, const N: usize> BatchActivation<U, Arr<U, N>, Arr<U, N>, DeviceCpu<U>> for SoftMax<U, DeviceCpu<U>>where
    U: UnitValue<U>,
    Vec<Arr<U, N>>: FromParallelIterator<Arr<U, N>>,

fn batch_apply(
    &self,
    device: &DeviceCpu<U>,
    input: &VecArr<U, Arr<U, N>>
) -> Result<VecArr<U, Arr<U, N>>, TrainingError>

Apply the activation function

fn batch_derive(
    &self,
    device: &DeviceCpu<U>,
    o: &VecArr<U, Arr<U, N>>,
    loss: &VecArr<U, Arr<U, N>>,
    u: &VecArr<U, Arr<U, N>>
) -> Result<VecArr<U, Arr<U, N>>, TrainingError>

Apply derivatives of the activation function

impl<U, const N: usize> BatchActivation<U, Arr<U, N>, Arr<U, N>, DeviceCpu<U>> for Swish<U, DeviceCpu<U>>where
    U: UnitValue<U>,
    Vec<Arr<U, N>>: FromParallelIterator<Arr<U, N>>,

fn batch_apply(
    &self,
    device: &DeviceCpu<U>,
    input: &VecArr<U, Arr<U, N>>
) -> Result<VecArr<U, Arr<U, N>>, TrainingError>

Apply the activation function

fn batch_derive(
    &self,
    device: &DeviceCpu<U>,
    o: &VecArr<U, Arr<U, N>>,
    loss: &VecArr<U, Arr<U, N>>,
    u: &VecArr<U, Arr<U, N>>
) -> Result<VecArr<U, Arr<U, N>>, TrainingError>

Apply derivatives of the activation function

impl<U, const N: usize> BatchActivation<U, Arr<U, N>, Arr<U, N>, DeviceCpu<U>> for Tanh<U, DeviceCpu<U>>where
    U: UnitValue<U>,
    Vec<Arr<U, N>>: FromParallelIterator<Arr<U, N>>,

fn batch_apply(
    &self,
    device: &DeviceCpu<U>,
    input: &VecArr<U, Arr<U, N>>
) -> Result<VecArr<U, Arr<U, N>>, TrainingError>

Apply the activation function

fn batch_derive(
    &self,
    device: &DeviceCpu<U>,
    o: &VecArr<U, Arr<U, N>>,
    loss: &VecArr<U, Arr<U, N>>,
    u: &VecArr<U, Arr<U, N>>
) -> Result<VecArr<U, Arr<U, N>>, TrainingError>

Apply derivatives of the activation function

impl<U> BatchLossFunction<U, DeviceCpu<U>> for CrossEntropy<U>where
    U: Clone + Copy + UnitValue<U>,

fn batch_linear_derive<const N: usize>(
    &self,
    _: &D,
    expected: &VecArr<U, Arr<U, N>>,
    actual: &VecArr<U, Arr<U, N>>
) -> Result<VecArr<U, Arr<U, N>>, TrainingError>

Differentiation of loss functions

impl<U> BatchLossFunction<U, DeviceCpu<U>> for CrossEntropyMulticlass<U>where
    U: Clone + Copy + UnitValue<U>,

fn batch_linear_derive<const N: usize>(
    &self,
    _: &D,
    expected: &VecArr<U, Arr<U, N>>,
    actual: &VecArr<U, Arr<U, N>>
) -> Result<VecArr<U, Arr<U, N>>, TrainingError>

Differentiation of loss functions

impl<U> BatchLossFunction<U, DeviceCpu<U>> for Mse<U>where
    U: Clone + Copy + UnitValue<U>,

fn batch_linear_derive<const N: usize>(
    &self,
    _: &D,
    expected: &VecArr<U, Arr<U, N>>,
    actual: &VecArr<U, Arr<U, N>>
) -> Result<VecArr<U, Arr<U, N>>, TrainingError>

Differentiation of loss functions

impl<U> Clone for DeviceCpu<U>where
    U: UnitValue<U>,

fn clone(&self) -> Self

Returns a copy of the value.

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

Performs copy-assignment from source.

impl<U> Device<U> for DeviceCpu<U>where
    U: UnitValue<U>,

fn loss_linear<L, const N: usize>(
    &self,
    expected: &Arr<U, N>,
    actual: &Arr<U, N>,
    lossf: &L
) -> Arr<U, N>where
    L: LossFunction<U>,

Calculation of Losses

fn loss_linear_by_canonical_link<const N: usize>(
    &self,
    expected: &Arr<U, N>,
    actual: &Arr<U, N>
) -> Arr<U, N>

Calculation of Losses by canonical link

fn loss_linear_total<L: LossFunction<U>, const N: usize>(
    &self,
    exptected: &Arr<U, N>,
    actual: &Arr<U, N>,
    lossf: &L
) -> U

Calculation of total Losses

fn loss_linear_batch_by_canonical_link<const N: usize>(
    &self,
    expected: &VecArr<U, Arr<U, N>>,
    actual: &VecArr<U, Arr<U, N>>
) -> Result<VecArr<U, Arr<U, N>>, TrainingError>where
    f64: From<U>,

Calculation of loss during batch execution by canonical link

fn batch_linear_reduce<const N: usize>(
    &self,
    loss: &VecArr<U, Arr<U, N>>
) -> Result<Arr<U, N>, TrainingError>

convolutional calculation

impl<U, const NI: usize, const NO: usize> DeviceLinear<U, Arr2<U, NI, NO>, NI, NO> for DeviceCpu<U>where
    U: UnitValue<U>,

fn forward_linear(
    &self,
    bias: &Arr<U, NO>,
    units: &Arr2<U, NI, NO>,
    input: &Arr<U, NI>
) -> Result<Arr<U, NO>, EvaluateError>

Forward propagation calculation

fn backward_linear(
    &self,
    units: &Arr2<U, NI, NO>,
    input: &Arr<U, NO>
) -> Result<Arr<U, NI>, TrainingError>

Error back propagation calculation

fn backward_weight_gradient(
    &self,
    o: &Arr<U, NI>,
    loss: &Arr<U, NO>
) -> Result<Arr2<U, NI, NO>, TrainingError>

Calculate the gradient of the weights

fn batch_backward_linear(
    &self,
    units: &Arr2<U, NI, NO>,
    input: &VecArr<U, Arr<U, NO>>
) -> Result<VecArr<U, Arr<U, NI>>, TrainingError>

Error back propagation in batch

fn batch_forward_linear(
    &self,
    bias: &Arr<U, NO>,
    units: &Arr2<U, NI, NO>,
    input: &VecArr<U, Arr<U, NI>>
) -> Result<VecArr<U, Arr<U, NO>>, TrainingError>

Forward propagation calculation in batch

fn batch_backward_weight_gradient(
    &self,
    o: &VecArr<U, Arr<U, NI>>,
    loss: &VecArr<U, Arr<U, NO>>
) -> Result<Arr2<U, NI, NO>, TrainingError>

Calculate the gradient of the weights in batch