INNER CODE UNIT · Python

fit_and_aggregate

NorskRegnesentral/skweak · skweak/aggregation.py:101

    def fit_and_aggregate(self, docs: Iterable[Doc]) -> Iterable[Doc]:
        """Starts by fitting the parameters of the aggregator (if necessary)
        then applies the resulting model to aggregate the outputs of the
        labelling functions."""

        docs = list(docs)
        self.fit(docs)
        return list(self.pipe(docs))
    
        
    @abstractmethod
    def get_observation_df(self, doc: Doc):
        """Returns a dataframe containing the observed predictions of each labelling
        sources for the document. The dataframe has one row per unique span for span 
        labelling, and one row per token for sequence labelling."""

        raise NotImplementedError("must implement get_observation_df")    

View source record →

📰 Research Paper
Loading…
⏳ Fetching content…