TranslationTask¶
- class flash.text.seq2seq.translation.model.TranslationTask(backbone='t5-small', tokenizer_kwargs=None, max_source_length=128, max_target_length=128, padding='max_length', loss_fn=None, optimizer='Adam', lr_scheduler=None, metrics=None, learning_rate=None, num_beams=4, n_gram=4, smooth=True, enable_ort=False)[source]¶
The
TranslationTask
is aTask
for Seq2Seq text translation. For more details, see Translation.You can change the backbone to any translation model from HuggingFace/transformers using the
backbone
argument.- Parameters
max_source_length¶ (
int
) – The maximum length to pad / truncate input sequences to.max_target_length¶ (
int
) – The maximum length to pad / truncate target sequences to.padding¶ (
Union
[str
,bool
]) – The type of padding to apply. One of: “longest” orTrue
, “max_length”, “do_not_pad” orFalse
.loss_fn¶ (
Optional
[TypeVar
(LOSS_FN_TYPE
,Callable
,Mapping
,Sequence
,None
)]) – Loss function for training.optimizer¶ (
TypeVar
(OPTIMIZER_TYPE
,str
,Callable
,Tuple
[str
,Dict
[str
,Any
]],None
)) – Optimizer to use for training.lr_scheduler¶ (
Optional
[TypeVar
(LR_SCHEDULER_TYPE
,str
,Callable
,Tuple
[str
,Dict
[str
,Any
]],Tuple
[str
,Dict
[str
,Any
],Dict
[str
,Any
]],None
)]) – The LR scheduler to use during training.metrics¶ (
Optional
[TypeVar
(METRICS_TYPE
,Metric
,Mapping
,Sequence
,None
)]) – Metrics to compute for training and evaluation. Defauls to calculating the BLEU metric. Changing this argument currently has no effect.learning_rate¶ (
Optional
[float
]) – Learning rate to use for training, defaults to 1e-5num_beams¶ (
Optional
[int
]) – Number of beams to use in validation when generating predictions. Defaults to 4n_gram¶ (
bool
) – Maximum n_grams to use in metric calculation. Defaults to 4smooth¶ (
bool
) – Apply smoothing in BLEU calculation. Defaults to Trueenable_ort¶ (
bool
) – Enable Torch ONNX Runtime Optimization: https://onnxruntime.ai/docs/#onnx-runtime-for-training
- classmethod available_finetuning_strategies(cls)¶
Returns a list containing the keys of the available Finetuning Strategies.
- classmethod available_lr_schedulers(cls)¶
Returns a list containing the keys of the available LR schedulers.
- classmethod available_optimizers(cls)¶
Returns a list containing the keys of the available Optimizers.