CharRecallPrecision

class mmeval.metrics.CharRecallPrecision(letter_case: str = 'unchanged', invalid_symbol: str = '[^A-Za-z0-9一-龥]', **kwargs)

Calculate the char level recall & precision.

Parameters

• letter_case (str) –

  There are three options to alter the letter cases

  • unchanged: Do not change prediction texts and labels.

  • upper: Convert prediction texts and labels into uppercase characters.

  • lower: Convert prediction texts and labels into lowercase characters.

  Usually, it only works for English characters. Defaults to ‘unchanged’.

• invalid_symbol (str) – A regular expression to filter out invalid or not cared characters. Defaults to ‘[^A-Za-z0-9u4e00-u9fa5]’.

• **kwargs – Keyword parameters passed to BaseMetric.

Examples

>>> from mmeval import CharRecallPrecision
>>> metric = CharRecallPrecision()
>>> metric(['helL', 'HEL'], ['hello', 'HELLO'])
{'char_recall': 0.6, 'char_precision': 0.8571428571428571}
>>> metric = CharRecallPrecision(letter_case='upper')
>>> metric(['helL', 'HEL'], ['hello', 'HELLO'])
{'char_recall': 0.7, 'char_precision': 1.0}

add(predictions: Sequence[str], groundtruths: Sequence[str]) → None

Process one batch of data and predictions.

Parameters

• predictions (list[str]) – The prediction texts.

• groundtruths (list[str]) – The ground truth texts.

compute_metric(results: Sequence[Tuple[int, int, int]]) → Dict

Compute the metrics from processed results.

Parameters

results (list[tuple]) – The processed results of each batch.

Returns

The computed metrics. The keys are the names of the metrics, and the values are corresponding results.

Return type

Dict