Struct AttentionConfig 

pub struct AttentionConfig {
    pub dim: usize,
    pub num_heads: usize,
    pub dropout: f32,
    pub scale: Option<f32>,
    pub causal: bool,
}

Configuration for standard attention mechanisms.

Fields

dim: usize

Model dimension (d_model)

num_heads: usize

Number of attention heads

dropout: f32

Dropout probability (0.0 to 1.0)

scale: Option<f32>

Scaling factor (default: 1/sqrt(d_k))

causal: bool

Whether to use causal masking

Implementations

impl AttentionConfig

pub fn builder() -> AttentionConfigBuilder

Creates a new builder for AttentionConfig.

pub fn validate(&self) -> AttentionResult<()>

Validates the configuration.

pub fn head_dim(&self) -> usize

Returns the dimension per head (d_k).

pub fn effective_scale(&self) -> f32

Returns the effective scale factor.

Trait Implementations

impl Clone for AttentionConfig

fn clone(&self) -> AttentionConfig

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for AttentionConfig

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for AttentionConfig

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for AttentionConfig

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.