query_expander.py 文件源码

def __load_embeding_model(self, file_path, max_vocab_size=100000):
        self.__embed_vectors = dict()
        if not file_path:
            print('Embeddings file not provided')
            return
        if not os.path.exists(file_path):
            print('Embeddings file not found:', file_path)
            return

        print('Loading the embedding model from:', file_path)
        bar = progressbar.ProgressBar(max_value=max_vocab_size)
        with open(file_path, "r") as embed_f:
            for line in embed_f:
                try:
                    tab = line.rstrip().split()
                    word = tab[0].lower()
                    if not word in self.__embed_vectors:
                        vec = numpy.array(tab[1:], dtype=float)
                        self.__embed_vectors[word] = vec
                except ValueError:
                    continue
                bar.update(len(self.__embed_vectors))
                if len(self.__embed_vectors) == max_vocab_size:
                    bar.finish()
                    return