Struct TransformerDecoderLayer

pub struct TransformerDecoderLayer<F: Float + Debug + Send + Sync> { /* private fields */ }

Transformer decoder layer

Implements a single layer of the transformer decoder as described in “Attention Is All You Need” by Vaswani et al. It consists of masked multi-head self-attention, multi-head cross-attention over encoder output, and a position-wise feed-forward network, with residual connections and layer normalization.

Implementations

impl<F: Float + Debug + ScalarOperand + Send + Sync + 'static> TransformerDecoderLayer<F>

pub fn new<R: Rng>( d_model: usize, n_heads: usize, d_ff: usize, dropout: f64, epsilon: f64, rng: &mut R, ) -> Result<Self>

Create a new transformer decoder layer

Arguments

• d_model - Model embedding dimension
• n_heads - Number of attention heads
• d_ff - Feed-forward network hidden dimension
• dropout - Dropout rate (0 means no dropout)
• epsilon - Small constant for layer normalization
• rng - Random number generator for weight initialization

Returns

• A new transformer decoder layer

pub fn forward_with_encoder( &self, input: &Array<F, IxDyn>, encoder_output: &Array<F, IxDyn>, ) -> Result<Array<F, IxDyn>>

Forward pass with encoder output

Arguments

• input - Input tensor [batch, tgt_len, d_model]
• encoder_output - Encoder output tensor [batch, src_len, d_model]

Returns

• Output tensor [batch, tgt_len, d_model]

Trait Implementations

impl<F: Float + Debug + ScalarOperand + Send + Sync + 'static> Layer<F> for TransformerDecoderLayer<F>

fn as_any(&self) -> &dyn Any

Get the layer as a dyn Any for downcasting

fn as_any_mut(&mut self) -> &mut dyn Any

Get the layer as a mutable dyn Any for downcasting

fn forward(&self, input: &Array<F, IxDyn>) -> Result<Array<F, IxDyn>>

Forward pass of the layer

fn backward( &self, input: &Array<F, IxDyn>, _grad_output: &Array<F, IxDyn>, ) -> Result<Array<F, IxDyn>>

Backward pass of the layer to compute gradients

fn update(&mut self, learning_rate: F) -> Result<()>

Update the layer parameters with the given gradients

fn params(&self) -> Vec<Array<F, IxDyn>>

Get the parameters of the layer

fn gradients(&self) -> Vec<Array<F, IxDyn>>

Get the gradients of the layer parameters

fn set_gradients(&mut self, _gradients: &[Array<F, IxDyn>]) -> Result<()>

Set the gradients of the layer parameters

fn set_params(&mut self, _params: &[Array<F, IxDyn>]) -> Result<()>

Set the parameters of the layer

fn set_training(&mut self, _training: bool)

Set the layer to training mode (true) or evaluation mode (false)

fn is_training(&self) -> bool

Get the current training mode

fn layer_type(&self) -> &str

Get the type of the layer (e.g., “Dense”, “Conv2D”)

fn parameter_count(&self) -> usize

Get the number of trainable parameters in this layer

fn layer_description(&self) -> String

Get a detailed description of this layer