7fa458e2c4c5df5a9d2cc4e66b2472cd9f3377a8,python/baseline/model.py,Tagger,predict_text,#Tagger#Any#,207

Before Change

        if featurizer is None:
            mxlen = kwargs.get("mxlen", self.mxlen if hasattr(self, "mxlen") else len(tokens))
            maxw = kwargs.get("maxw", self.maxw if hasattr(self, "maxw") else max([len(token) for token in tokens]))
            zero_alloc = kwargs.get("zero_alloc", np.zeros)
            featurizer = WordCharLength(self, mxlen, maxw, zero_alloc)

        // This might be inefficient if the label space is large

After Change

        // This might be inefficient if the label space is large

        label_vocab = revlut(self.get_labels())
        batch_dict = dict()
        for k, vectorizer in vectorizers.items():
            value, length = vectorizer.run(tokens, self.embeddings[k].vocab)
            batch_dict[k] = value
            if length is not None:
                batch_dict["{}_lengths".format(k)] = length

        indices = self.predict(batch_dict)[0]
        output = []
        for j in len(tokens):
            output.append((tokens[j], label_vocab[indices[j].item()]))

Instances