AttentionPostprocessor¶

class mmocr.models.textrecog.AttentionPostprocessor(dictionary, max_seq_len=40, ignore_chars=['padding'], **kwargs)[source]¶

PostProcessor for seq2seq.

Parameters

dictionary (Union[mmocr.models.common.dictionary.dictionary.Dictionary, Dict]) –
max_seq_len (int) –
ignore_chars (Sequence[str]) –

Return type

None

get_single_prediction(probs, data_sample=None)[source]¶

Convert the output probabilities of a single image to index and score.

Parameters

probs (torch.Tensor) – Character probabilities with shape \((T, C)\).
data_sample (TextRecogDataSample, optional) – Datasample of an image. Defaults to None.

Returns

index and score.

Return type

tuple(list[int], list[float])