Struct Linear 

pub struct Linear<D: Dtype, IT, OT, const RANK: usize> {
    pub in_features: usize,
    pub out_features: usize,
    pub has_bias: bool,
    /* private fields */
}

Fully-connected linear layer: output = input × Wᵀ [+ b]

Operates on the last dimension of an N-D tensor, passing all leading dimensions through unchanged (e.g. a batch dimension).

Type parameters:

• D — element dtype
• IT — input tensor type (any rank, last dim = in_features)
• OT — output tensor type (same rank, last dim = out_features)
• RANK — tensor rank (1 = unbatched, 2 = batched, etc.)

Tensor bounds are on impls, not the struct, so SymTensor can have its own Layer impl without a coherence conflict.

Fields

in_features: usize

Size of the last input dimension.

out_features: usize

Size of the last output dimension.

has_bias: bool

Whether to add a learned bias.

Implementations

impl<D: Dtype, IT, OT, const RANK: usize> Linear<D, IT, OT, RANK>

pub fn new(in_features: usize, out_features: usize, has_bias: bool) -> Self

Creates a new Linear layer.

Trait Implementations

impl<D: Dtype, IT: Tensor<D, RANK> + EagerTensor, OT: Tensor<D, RANK>, const RANK: usize> Layer<IT> for Linear<D, IT, OT, RANK>

type Output = OT

This layer’s output type.

fn call(&self, _input: IT) -> Self::Output

Applies this layer to input.

impl<D: Dtype, const RANK: usize> Layer<SymTensor> for Linear<D, SymTensor, SymTensor, RANK>

type Output = SymTensor

This layer’s output type.

fn call(&self, input: SymTensor) -> SymTensor

Applies this layer to input.