Python data_utils 模块，prepare_wmt_data() 实例源码

def load_data(self):
    # TODO: make configurable
    self.data_dir = "/data/WMT15/"

    print("Preparing WMT data in %s" % self.data_dir)
    en_train, fr_train, en_dev, fr_dev, _, _ = data_utils.prepare_wmt_data(
        self.data_dir, self.en_vocab_size, self.fr_vocab_size)

    # Read data into buckets and compute their sizes.
    print ("Reading development and training data (limit: %d)."
           % self.max_train_data_size)
    self.dev_set = self.read_data(en_dev, fr_dev)
    self.train_set = self.read_data(en_train, fr_train, self.max_train_data_size)
    train_bucket_sizes = [len(self.train_set[b]) for b in xrange(len(self._buckets))]
    train_total_size = float(sum(train_bucket_sizes))

    # A bucket scale is a list of increasing numbers from 0 to 1 that we'll use
    # to select a bucket. Length of [scale[i], scale[i+1]] is proportional to
    # the size if i-th training bucket, as used later.
    self.train_buckets_scale = [sum(train_bucket_sizes[:i + 1]) / train_total_size
                           for i in xrange(len(train_bucket_sizes))]