Struct RNNTLoss 

pub struct RNNTLoss { /* private fields */ }

RNN Transducer (RNNT) loss, as described in Sequence Transduction with Recurrent Neural Networks.

Computes the negative log-likelihood over a 2D lattice of encoder time steps (T) and output labels (U), marginalizing over all valid alignments.

Example

let rnnt = RNNTLossConfig::new().init();

// logits: [B, T, U+1, V] from the joiner network
let loss = rnnt.forward(logits, targets, logit_lengths, target_lengths);

Implementations

impl RNNTLoss

pub fn forward<B>( &self, logits: Tensor<B, 4>, targets: Tensor<B, 2, Int>, logit_lengths: Tensor<B, 1, Int>, target_lengths: Tensor<B, 1, Int>, ) -> Tensor<B, 1>

where B: Backend,

Computes per-sample RNNT loss (no reduction). Returns shape [B].

• logits: [B, T, U+1, V] — joiner output (raw logits or log-probs)
• targets: [B, U] — target label indices (must not contain blank)
• logit_lengths: [B] — actual encoder lengths per sample
• target_lengths: [B] — actual target lengths per sample

pub fn forward_with_reduction<B>( &self, logits: Tensor<B, 4>, targets: Tensor<B, 2, Int>, logit_lengths: Tensor<B, 1, Int>, target_lengths: Tensor<B, 1, Int>, reduction: Reduction, ) -> Tensor<B, 1>

where B: Backend,

Computes RNNT loss with the given reduction. Returns shape [1].

Trait Implementations

impl<B> AutodiffModule<B> for RNNTLoss

where B: AutodiffBackend,

type InnerModule = RNNTLoss

Inner module without auto-differentiation.

fn valid(&self) -> <RNNTLoss as AutodiffModule<B>>::InnerModule

Returns the same module, but on the inner backend without auto-differentiation.

fn from_inner(module: <RNNTLoss as AutodiffModule<B>>::InnerModule) -> RNNTLoss

Wraps an inner module back into an auto-diff module.

impl Clone for RNNTLoss

fn clone(&self) -> RNNTLoss

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for RNNTLoss

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

Formats the value using the given formatter.

impl Display for RNNTLoss

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

Formats the value using the given formatter.

impl<B> Module<B> for RNNTLoss

where B: Backend,

type Record = EmptyRecord

Type to save and load the module.

fn visit<V>(&self, _visitor: &mut V)

where V: ModuleVisitor<B>,

Visit each tensor parameter in the module with a visitor.

fn map<M>(self, _mapper: &mut M) -> RNNTLoss

where M: ModuleMapper<B>,

Map each tensor parameter in the module with a mapper.

fn load_record(self, _record: <RNNTLoss as Module<B>>::Record) -> RNNTLoss

Load the module state from a record.

fn into_record(self) -> <RNNTLoss as Module<B>>::Record

Convert the module into a record containing the state.

fn to_device(self, _: &<B as BackendTypes>::Device) -> RNNTLoss

Move the module and all of its sub-modules to the given device.

fn fork(self, _: &<B as BackendTypes>::Device) -> RNNTLoss

Fork the module and all of its sub-modules to the given device.

fn collect_devices( &self, devices: Vec<<B as BackendTypes>::Device>, ) -> Vec<<B as BackendTypes>::Device>

Return all the devices found in the underneath module tree added to the given vector without duplicates.

fn devices(&self) -> Vec<<B as BackendTypes>::Device>

Return all the devices found in the underneath module tree without duplicates.

fn no_grad(self) -> Self

Each tensor in the module tree will not require grad.

fn train<AB>(self) -> Self::TrainModule

where AB: AutodiffBackend<InnerBackend = B>, Self: HasAutodiffModule<AB>,

Move the module and all of its sub-modules to the autodiff backend.

fn num_params(&self) -> usize

Get the number of parameters the module has, including all of its sub-modules.

fn save_file<FR, PB>( self, file_path: PB, recorder: &FR, ) -> Result<(), RecorderError>

where FR: FileRecorder<B>, PB: Into<PathBuf>,

Save the module to a file using the provided file recorder.

fn load_file<FR, PB>( self, file_path: PB, recorder: &FR, device: &<B as BackendTypes>::Device, ) -> Result<Self, RecorderError>

where FR: FileRecorder<B>, PB: Into<PathBuf>,

Load the module from a file using the provided file recorder.

fn quantize_weights(self, quantizer: &mut Quantizer) -> Self

Quantize the weights of the module.

impl ModuleDisplay for RNNTLoss

fn format(&self, passed_settings: DisplaySettings) -> String

Formats the module with provided display settings.

fn custom_settings(&self) -> Option<DisplaySettings>

Custom display settings for the module.

fn custom_content(&self, _content: Content) -> Option<Content>

Custom attributes for the module.

impl ModuleDisplayDefault for RNNTLoss

fn content(&self, content: Content) -> Option<Content>

Attributes of the module used for display purposes.

fn num_params(&self) -> usize

Gets the number of the parameters of the module.