INNER CODE UNIT · Python
_get_spans
NorskRegnesentral/skweak · skweak/aggregation.py:360
def _get_spans(self, agg_df: pandas.DataFrame) -> List[Tuple[int,int,str]]:
"""Takes as input a 2D dataframe of shape (nb_entries, nb_labels)
assocating each token to the probability of an output label, and
returns a list of tuples (start, end, label).
"""
return [(start, end, label) for (start, end), label in
utils.token_array_to_spans(agg_df.values, self.out_labels).items()]
@abstractmethod
def _get_probs(self, agg_df: pandas.DataFrame) -> Dict:
"""Takes as input a 2D dataframe of shape (nb_entries, nb_labels)
assocating each token to the probability of an output label, and returns
a dictionary mapping each token to a distribution over labels"""
return utils.token_array_to_probs(agg_df.values, self.out_labels)
def get_observation_df(self, doc: Doc):