class MetricCallback[source]

MetricCallback(after_create=None, before_fit=None, before_epoch=None, before_train=None, before_batch=None, after_pred=None, after_loss=None, before_backward=None, before_step=None, after_cancel_step=None, after_step=None, after_cancel_batch=None, after_batch=None, after_cancel_train=None, after_train=None, before_validate=None, after_cancel_validate=None, after_validate=None, after_cancel_epoch=None, after_epoch=None, after_cancel_fit=None, after_fit=None) :: Callback

Computes metric scores using HuggingFace datasets metric

Rouge

class RougeScore[source]

RougeScore(tokenizer, argmax:bool=True, agg:str='mid', measure:str='fmeasure', scores:List[str]=['rouge1', 'rouge2', 'rougeL', 'rougeLsum']) :: Callback

Computes ROUGE score using HF datasets metric. Adds ValueMetric for each score in scores

Seqeval

class Seqeval[source]

Seqeval(label_list, scores:Tuple[str]=('accuracy', 'f1', 'precision', 'recall')) :: MetricCallback

Computes scores using seqeval https://github.com/chakki-works/seqeval

Fin