def load_embeddings()

in question_generation_model.py [0:0]

• 21 lines of code
• 9 McCabe index (conditional complexity)

    def load_embeddings(self):
        """
        Load Glove vectors
        :return:
        """
        glove_dir = os.getcwd()  # '/Volumes/Data/data/glove.6B'
        embeddings_index = {}  # empty dictionary
        f = open(os.path.join(glove_dir, self.datasets.embedding_file), encoding="utf-8")
        for line in f:
            values = line.split()
            word = values[0]
            coefs = np.asarray(values[1:], dtype='float32')
            embeddings_index[word] = coefs
        f.close()

        self.logger.info("Embedding_index: %s" % str(len(embeddings_index)))

        # Get 200-dim dense vector for each of the words in our vocabulary
        embedding_matrix = np.zeros((self.vocab_size, self.embedding_dim))
        for word, i in self.word_to_idx.items():
            # if i < max_words:
            embedding_vector = embeddings_index.get(word)

            if word == '<START>' or word == '<END>':
                embedding_vector = np.random.rand(self.embedding_dim)

            if embedding_vector is not None and len(embedding_vector) > 0:
                embedding_matrix[i] = embedding_vector
            else:
                # Words not found in the embedding index will be all zeros
                embedding_matrix[i] = np.zeros(self.embedding_dim)

        return embedding_matrix