aac_metrics.classes.fer module¶

class FER( return_all_scores: bool = True, *, echecker: str | BERTFlatClassifier = 'echecker_clotho_audiocaps_base', error_threshold: float = 0.9, device: str | device | None = 'cuda_if_available', batch_size: int | None = 32, reset_state: bool = True, return_probs: bool = False, verbose: int = 0, )[source]¶

Bases: AACMetric[tuple[FERScores, FERScores] | Tensor]

Return Fluency Error Rate (FER) detected by a pre-trained BERT model.

Paper: https://arxiv.org/abs/2110.04684
Original implementation: https://github.com/blmoistawinde/fense

For more information, see fer().

compute() → tuple[FERScores, FERScores] | Tensor[source]¶

extra_repr() → str[source]¶

Return the extra representation of the module.

To print customized extra information, you should re-implement this method in your own modules. Both single-line and multi-line strings are acceptable.

full_state_update: ClassVar[bool | None] = False¶

get_output_names() → tuple[str, ...][source]¶

higher_is_better: ClassVar[bool | None] = False¶

is_differentiable: ClassVar[bool | None] = False¶

max_value: ClassVar[float] = 1.0¶

min_value: ClassVar[float] = -1.0¶

reset() → None[source]¶

training: bool¶

update(

candidates: list[str],

*args,

**kwargs,

) → None[source]¶