[人工智能]Bert模型创建过程解读

def create_model(bert_config, is_training, input_ids, input_mask, segment_ids,
                 labels, num_labels, use_one_hot_embeddings):
    """Creates a classification model."""
    model = modeling.BertModel(
        config=bert_config,
        is_training=is_training,
        input_ids=input_ids,
        input_mask=input_mask,
        token_type_ids=segment_ids,
        use_one_hot_embeddings=use_one_hot_embeddings)

    # In the demo, we are doing a simple classification task on the entire
    # segment.
    #
    # If you want to use the token-level output, use model.get_sequence_output()
    # instead.
    output_layer = model.get_pooled_output()  # 句子首词（CLS）对应的隐藏层输出编码
	# 要拿到所有词（字）的编码，要用model.get_sequence_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))   # 创建output_layer对应的权重，相当于创建全连接层（偏置项后面加）

    output_bias = tf.get_variable(
        "output_bias", [num_labels], initializer=tf.zeros_initializer())  # 即偏置项
	
	
    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)  # 和上一行代码这两步运算其实就是全连接层
        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)
        # 到此为止，因该是生层了一行，[batch_size, per_loss]
        loss = tf.reduce_mean(per_example_loss)  # 应该是 mean loss of a batch
	
        return (loss, per_example_loss, logits, probabilities)

如果想在bert后面自己加层，可以在上面的模型创建模块内操作。