用深度神经网络处理NER命名实体识别问题

• Person (PER)
• Organization (ORG)
• Location (LOC)
• Miscellaneous (MISC)

EU  ORG
rejects O
German  MISC
Peter   PER
BRUSSELS    LOC

def test_NER():

  config = Config()
  with tf.Graph().as_default():
    model = NERModel(config)   # 最主要的类

    init = tf.initialize_all_variables()
    saver = tf.train.Saver()

    with tf.Session() as session:
      best_val_loss = float('inf')  # 最好的值时，它的 loss 它的 迭代次数 epoch
      best_val_epoch = 0

      session.run(init)
      for epoch in xrange(config.max_epochs):
        print 'Epoch {}'.format(epoch)
        start = time.time()
        ###
        train_loss, train_acc = model.run_epoch(session, model.X_train,
                                                model.y_train)   # 1.把 train 数据放进迭代里跑，得到 loss 和 accuracy
        val_loss, predictions = model.predict(session, model.X_dev, model.y_dev)   # 2.用这个model去预测 dev 数据，得到loss 和 prediction
        print 'Training loss: {}'.format(train_loss)
        print 'Training acc: {}'.format(train_acc)
        print 'Validation loss: {}'.format(val_loss)
        if val_loss < best_val_loss:            # 用 val 数据的loss去找最小的loss
          best_val_loss = val_loss
          best_val_epoch = epoch
          if not os.path.exists("./weights"):
            os.makedirs("./weights")
        
          saver.save(session, './weights/ner.weights')   # 把最小的 loss 对应的 weights 保存起来
        if epoch - best_val_epoch > config.early_stopping:
          break
        ###
        confusion = calculate_confusion(config, predictions, model.y_dev)  # 3.把 dev 的lable数据放进去，计算prediction的confusion
        print_confusion(confusion, model.num_to_tag)
        print 'Total time: {}'.format(time.time() - start)
      
      saver.restore(session, './weights/ner.weights')   # 再次加载保存过的 weights，用 test 数据做预测，得到预测结果
      print 'Test'
      print '=-=-='
      print 'Writing predictions to q2_test.predicted'
      _, predictions = model.predict(session, model.X_test, model.y_test)
      save_predictions(predictions, "q2_test.predicted")    # 把预测结果保存起来

if __name__ == "__main__":
  test_NER()

# Load the training set
docs = du.load_dataset('data/ner/train')

# Load the dev set (for tuning hyperparameters)
docs = du.load_dataset('data/ner/dev')

# Load the test set (dummy labels only)
docs = du.load_dataset('data/ner/test.masked')

  def add_embedding(self):
    # The embedding lookup is currently only implemented for the CPU
    with tf.device('/cpu:0'):

      embedding = tf.get_variable('Embedding', [len(self.wv), self.config.embed_size])  # assignment 中的 L   
      window = tf.nn.embedding_lookup(embedding, self.input_placeholder)                # 在 L 中直接把window大小的context的word vector搞定
      window = tf.reshape(
        window, [-1, self.config.window_size * self.config.embed_size])

      return window

  def add_model(self, window):
  
    with tf.variable_scope('Layer1', initializer=xavier_weight_init()) as scope:        # 用initializer=xavier去初始化第一层
      W = tf.get_variable(                                                              # 第一层有 W，b1，h
          'W', [self.config.window_size * self.config.embed_size,
                self.config.hidden_size])
      b1 = tf.get_variable('b1', [self.config.hidden_size])
      h = tf.nn.tanh(tf.matmul(window, W) + b1)
      if self.config.l2:                                                                # L2 regularization for W
          tf.add_to_collection('total_loss', 0.5 * self.config.l2 * tf.nn.l2_loss(W))   # 0.5 * self.config.l2 * tf.nn.l2_loss(W)

    with tf.variable_scope('Layer2', initializer=xavier_weight_init()) as scope:
      U = tf.get_variable('U', [self.config.hidden_size, self.config.label_size])
      b2 = tf.get_variable('b2', [self.config.label_size])
      y = tf.matmul(h, U) + b2
      if self.config.l2:
          tf.add_to_collection('total_loss', 0.5 * self.config.l2 * tf.nn.l2_loss(U))
    output = tf.nn.dropout(y, self.dropout_placeholder)                                 # 返回 output，两个variable_scope都带dropout

    return output 

  def add_loss_op(self, y):

    cross_entropy = tf.reduce_mean(                                                     # 1.关键步骤：loss是用cross entropy定义的
        tf.nn.softmax_cross_entropy_with_logits(y, self.labels_placeholder))                # y是模型预测值，计算cross entropy
    tf.add_to_collection('total_loss', cross_entropy)           # Stores value in the collection with the given name.
                                                                # collections are not sets, it is possible to add a value to a collection several times.
    loss = tf.add_n(tf.get_collection('total_loss'))            # Adds all input tensors element-wise. inputs: A list of Tensor with same shape and type

    return loss 

  def add_training_op(self, loss):

    optimizer = tf.train.AdamOptimizer(self.config.lr)
    global_step = tf.Variable(0, name='global_step', trainable=False)
    train_op = optimizer.minimize(loss, global_step=global_step)    # 2.关键步骤：用 AdamOptimizer 使 loss 达到最小，所以更关键的是 loss

    return train_op