doc2vec.py 文件源码

def fit(self, docs, y):
        assert len(docs) == len(y)
        model = self.model
        n_epochs = self.n_epochs
        verbose = self.verbose
        decay = (self.alpha - self.min_alpha) / n_epochs
        X = [TaggedDocument(self.analyzer(doc), [label])
             for doc, label in zip(docs, y)]

        if verbose > 0:
            print("First 3 tagged documents:\n", X[:3])
            print("Training doc2vec model")
        # d2v = Doc2Vec()
        # d2v.build_vocab(X)
        # if self.intersect is not None:
        #     d2v.intersect_word2vec_format(self.intersect)
        model.build_vocab(X)
        for epoch in range(n_epochs):
            if verbose:
                print("Doc2Vec: Epoch {} of {}.".format(epoch + 1, n_epochs))
            model.train(X)
            model.alpha -= decay  # apply global decay
            model.min_alpha = model.alpha  # but no decay inside one epoch

        if verbose > 0:
            print("Finished.")
            print("model:", self.model)

        if self._matching:
            self._matching.fit(docs)
        else:
            # if we dont do matching, its enough to fit a nearest neighbors on
            # all centroids before query time
            dvs = np.asarray([model.docvecs[tag] for tag in y])
            self._neighbors.fit(dvs)

        self._y = y

        return self