Struct AnyMoePipeline

pub struct AnyMoePipeline { /* private fields */ }

Implementations

impl AnyMoePipeline

pub fn new( target: Arc<Mutex<dyn Pipeline>>, config: AnyMoeConfig, inputs: AnyMoeTrainingInputs, prefix: String, mlp: String, model_ids: Vec<String>, token: TokenSource, revision: Option<String>, layers: Vec<usize>, silent: bool, ) -> Result<AnyMoePipeline, Error>

Trait Implementations

impl Pipeline for AnyMoePipeline

fn forward_inputs( &mut self, inputs: Box<dyn Any>, _return_raw_logits: bool, ) -> Result<ForwardInputsResult, Error>

fn sample_causal_gen<'life0, 'life1, 'life2, 'life3, 'async_trait>( &'life0 self, seqs: &'life1 mut [&'life2 mut Sequence], logits: Vec<Tensor>, prefix_cacher: &'life3 mut PrefixCacheManagerV2, disable_eos_stop: bool, rng: Arc<Mutex<Isaac64Rng>>, ) -> Pin<Box<dyn Future<Output = Result<(), Error>> + Send + 'async_trait>>

where 'life0: 'async_trait, 'life1: 'async_trait, 'life2: 'async_trait, 'life3: 'async_trait, AnyMoePipeline: 'async_trait,

fn category(&self) -> ModelCategory

fn step<'life0, 'life1, 'life2, 'life3, 'life4, 'async_trait>( &'life0 mut self, input_seqs: &'life1 mut [&'life2 mut Sequence], is_prompt: bool, return_raw_logits: bool, prefix_cacher: &'life3 mut PrefixCacheManagerV2, disable_eos_stop: bool, rng: Arc<Mutex<Isaac64Rng>>, backend_metadata: CacheBackendMetadata<'life4>, ) -> Pin<Box<dyn Future<Output = Result<Duration, Error>> + Send + 'async_trait>>

where 'life0: 'async_trait, 'life1: 'async_trait, 'life2: 'async_trait, 'life3: 'async_trait, 'life4: 'async_trait, Self: 'async_trait,

Returns the total of model execution time.