ptb_word_lm.py 文件源码

def run_epoch(session, model, data, is_train=False, verbose=False):
  """Runs the model on the given data."""
  epoch_size = ((len(data) // model.batch_size) - 1) // model.num_steps
  start_time = time.time()
  costs = 0.0
  iters = 0
  state = session.run(model.initial_state)

  for step, (x, y) in enumerate(reader.ptb_iterator(data, model.batch_size, model.num_steps)):
    if is_train:
      fetches = [model.cost, model.final_state, model.train_op]
    else:
      fetches = [model.cost, model.final_state]
    feed_dict = {}
    feed_dict[model.input_data] = x
    feed_dict[model.targets] = y
    for layer_num, (c, h) in enumerate(model.initial_state):
      feed_dict[c] = state[layer_num].c
      feed_dict[h] = state[layer_num].h

    if is_train:
      cost, state, _ = session.run(fetches, feed_dict)
    else:
      cost, state = session.run(fetches, feed_dict)

    costs += cost
    iters += model.num_steps

    if verbose and step % (epoch_size // 10) == 10:
      print("%.3f perplexity: %.3f speed: %.0f wps" %
            (step * 1.0 / epoch_size, np.exp(costs / iters),
             iters * model.batch_size / (time.time() - start_time)))

  return np.exp(costs / iters)