[人工智能]【Bert】（七）句子关系判断--源码解析（bert后处理模型+损失函数）

论文：https://arxiv.org/pdf/1810.04805.pdf

官方代码：GitHub - google-research/bert: TensorFlow code and pre-trained models for BERT

bert后处理模型

在run_classifier.py中的create_model函数中，“bert后处理模型”代码为：

  output_layer = model.get_pooled_output()

  hidden_size = output_layer.shape[-1].value

  output_weights = tf.get_variable(
      "output_weights", [num_labels, hidden_size],
      initializer=tf.truncated_normal_initializer(stddev=0.02))

  output_bias = tf.get_variable(
      "output_bias", [num_labels], initializer=tf.zeros_initializer())

  with tf.variable_scope("loss"):
    if is_training:
      # I.e., 0.1 dropout
      output_layer = tf.nn.dropout(output_layer, keep_prob=0.9)

    logits = tf.matmul(output_layer, output_weights, transpose_b=True)
    logits = tf.nn.bias_add(logits, output_bias)
    probabilities = tf.nn.softmax(logits, axis=-1)
    log_probs = tf.nn.log_softmax(logits, axis=-1)

bert后处理模型的本意是根据不同的任务将结果输出为想要的结果。

本篇博客是针对两句话语义是否相同来做的判断，属于分类任务，接下来就可以往分类任务的角度来构建输出结果。

output_layer = model.get_pooled_output()

其实对应的就是

      with tf.variable_scope("pooler"):
        # We "pool" the model by simply taking the hidden state corresponding
        # to the first token. We assume that this has been pre-trained
        first_token_tensor = tf.squeeze(self.sequence_output[:, 0:1, :], axis=1)
        self.pooled_output = tf.layers.dense(
            first_token_tensor,
            config.hidden_size,
            activation=tf.tanh,
            kernel_initializer=create_initializer(config.initializer_range))

取出了bert基础模型的输出，将其[CLS]这个标签位的内容取出来，然后全连接输出[batchsize, 768]形状的张量。

  hidden_size = output_layer.shape[-1].value

  output_weights = tf.get_variable(
      "output_weights", [num_labels, hidden_size],
      initializer=tf.truncated_normal_initializer(stddev=0.02))

  output_bias = tf.get_variable(
      "output_bias", [num_labels], initializer=tf.zeros_initializer())

  with tf.variable_scope("loss"):
    if is_training:
      # I.e., 0.1 dropout
      output_layer = tf.nn.dropout(output_layer, keep_prob=0.9)

    logits = tf.matmul(output_layer, output_weights, transpose_b=True)
    logits = tf.nn.bias_add(logits, output_bias)

上述代码其实就是在完成如下操作?

?

损失函数

    probabilities = tf.nn.softmax(logits, axis=-1)
    log_probs = tf.nn.log_softmax(logits, axis=-1)

    one_hot_labels = tf.one_hot(labels, depth=num_labels, dtype=tf.float32)

    per_example_loss = -tf.reduce_sum(one_hot_labels * log_probs, axis=-1)
    loss = tf.reduce_mean(per_example_loss)

损失其实就没什么好讲的，就是简单的交叉熵损失函数，具体使用的公式如下:

训练与测试

本任务的训练与测试的总体环节相同，仅仅是训练多了损失需要更新参数，测试不需要损失，直接输出概率最大的那个类即可。