Function layer_norm_device

pub fn layer_norm_device<'a>(
    x: impl AsRef<Array>,
    weight: impl Into<Option<&'a Array>>,
    bias: impl Into<Option<&'a Array>>,
    eps: f32,
    stream: impl AsRef<Stream>,
) -> Result<Array>

Layer normalization.

The normalization is with respect to the last axis of the input x.

Params

• x: input array
• weight: A multiplicative weight to scale the result by. The weight should be one-dimensional with the same size as the last axis of x. If not given no scaling will occur.
• bias: An additive offset to be added to the result. The bias should be one-dimensional with the same size as the last axis of x. It not given no offset will occur.
• eps: A small additive constant for numerical stability
• stream: stream or device to evaluate on