Linear.md

Linear

Applies a linear transformation to the incoming data:

$$y=XW^T+b$$

Attributes/Parameters

in_features: int

$K$ dim of weight.

out_features: int

$N$ dim of weight

bias_term: bool(default: True)

Mark that whether there is bias term. Provide convenience for graph optimization.

Inputs

X: tensor(T1)

Input feature of linear transformation.

Shape: $(*,K)$

W(constant): tensor(T1)

Transformation weight.

Shape: $(N,K)$.

B(constant, optional): tensor(T2)

Transformation bias.

Shape: $(N)$.

Outputs

Y: tensor(T2)

Output feature of linear transformation.

Shape: $(*,N)$.

Type Constraints

T1: float32, float16, int8

T2: float32, float16, int8, int32

enable accumulate with int32 when using int8 linear