o yi @s\ddlmZmZmZddlZddlmZmZddlmZm Z ddl m Z Gddde Z dS))AnyListUnionN)Tensortensor) _cer_compute _cer_update)MetriccseZdZUdZdZeed<dZeed<dZeed<e ed<e ed<de ffd d Z d e e ee fd e e ee fd dfddZd e fddZZS) CharErrorRateabCharacter Error Rate (`CER`_) is a metric of the performance of an automatic speech recognition (ASR) system. This value indicates the percentage of characters that were incorrectly predicted. The lower the value, the better the performance of the ASR system with a CharErrorRate of 0 being a perfect score. Character error rate can then be computed as: .. math:: CharErrorRate = \frac{S + D + I}{N} = \frac{S + D + I}{S + D + C} where: - :math:`S` is the number of substitutions, - :math:`D` is the number of deletions, - :math:`I` is the number of insertions, - :math:`C` is the number of correct characters, - :math:`N` is the number of characters in the reference (N=S+D+C). Compute CharErrorRate score of transcribed segments against references. As input to ``forward`` and ``update`` the metric accepts the following input: - ``preds`` (:class:`~str`): Transcription(s) to score as a string or list of strings - ``target`` (:class:`~str`): Reference(s) for each speech input as a string or list of strings As output of ``forward`` and ``compute`` the metric returns the following output: - ``cer`` (:class:`~torch.Tensor`): A tensor with the Character Error Rate score Args: kwargs: Additional keyword arguments, see :ref:`Metric kwargs` for more info. Examples: >>> preds = ["this is the prediction", "there is an other sample"] >>> target = ["this is the reference", "there is another one"] >>> cer = CharErrorRate() >>> cer(preds, target) tensor(0.3415) Fis_differentiablehigher_is_betterfull_state_updateerrorstotalkwargsc sJtjdi||jdtdtjddd|jdtdtjddddS)Nrr)dtypesum)dist_reduce_fxr)super__init__ add_statertorchfloat)selfr __class__rI/home/ubuntu/.local/lib/python3.10/site-packages/torchmetrics/text/cer.pyrGszCharErrorRate.__init__predstargetreturnNcCs.t||\}}|j|7_|j|7_dS)z*Update state with predictions and targets.N)rrr)rrrrrrrrupdateOszCharErrorRate.updatecCst|j|jS)z#Calculate the character error rate.)rrr)rrrrcomputeUszCharErrorRate.compute)__name__ __module__ __qualname____doc__r bool__annotations__r r rrrrstrrr!r" __classcell__rrrrr s  '  .r ) typingrrrrrr torchmetrics.functional.text.cerrrtorchmetrics.metricr r rrrrs