Struct nncombinator::activation::SoftMax

pub struct SoftMax<U, D>where
    U: UnitValue<U>,
    D: Device<U>,{ /* private fields */ }

SoftMax Implementation

Implementations

impl<U, D> SoftMax<U, D>where
    U: UnitValue<U>,
    D: Device<U>,

pub fn new(_: &D) -> SoftMax<U, D>

Create an instance of SoftMax

Trait Implementations

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>, DeviceGpu<U>> for SoftMax<U, DeviceGpu<U>>where
    U: UnitValue<U> + DataTypeInfo,
    DeviceGpu<U>: Device<U>,
    CudaPtr<U>: TryFrom<U, Error = CudaError>,
    SoftMaxForward<U>: Kernel<Args = ActivationForwardArgs<U>>,
    SoftMaxBackward<U>: Kernel<Args = ActivationBackwardArgs<U>>,

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

Apply the activation function

fn derive(
    &self,
    _: &DeviceGpu<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, 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, 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>, DeviceGpu<U>> for SoftMax<U, DeviceGpu<U>>where
    U: UnitValue<U> + DataTypeInfo,
    DeviceGpu<U>: Device<U>,
    CudaPtr<U>: TryFrom<U, Error = CudaError>,
    SoftMaxForward<U>: Kernel<Args = ActivationForwardArgs<U>>,
    SoftMaxBackward<U>: Kernel<Args = ActivationBackwardArgs<U>>,

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

Apply the activation function

fn batch_derive(
    &self,
    _: &DeviceGpu<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