```python
1、配置
class BertConfig(object):
"""Configuration for `BertModel`."""
def __init__(self,
vocab_size,
hidden_size=768,
num_hidden_layers=12,
num_attention_heads=12, # attention的頭數
class BertConfig(object):
"""Configuration for `BertModel`."""
def __init__(self,
vocab_size, # 詞典大小
hidden_size=768, # 隱層神經元個數
num_hidden_layers=12, # Transformer的層數12
num_attention_heads=12, attention的頭數
intermediate_size=3072, # 中間層神經元個數
hidden_act="gelu", # 激活函數
hidden_dropout_prob=0.1, # 隱層dropout的比例
attention_probs_dropout_prob=0.1, # attention裏面dropout比例
max_position_embeddings=512, # sequence最大長度
type_vocab_size=16, # token_type_ids的詞典大小
initializer_range=0.02):
self.vocab_size = vocab_size
self.hidden_size = hidden_size
self.num_hidden_layers = num_hidden_layers
self.num_attention_heads = num_attention_heads
self.hidden_act = hidden_act
self.intermediate_size = intermediate_size
self.hidden_dropout_prob = hidden_dropout_prob
self.attention_probs_dropout_prob = attention_probs_dropout_prob
self.max_position_embeddings = max_position_embeddings
self.type_vocab_size = type_vocab_size
self.initializer_range = initializer_range
模型配置，比較簡單，依次是：詞典大小、隱層神經元個數、transformer的層數、attention的頭數、激活函數、中間層神經元個數、隱層dropout比例、attention裏面dropout比例、sequence最大長度、token_type_ids的詞典大小、truncated_normal_initializer的stdev。
2、word embedding
def embedding_lookup(input_ids,
vocab_size,
embedding_size=128,
initializer_range=0.02,
word_embedding_name="word_embeddings",
use_one_hot_embeddings=False):
if input_ids.shape.ndims == 2:
input_ids = tf.expand_dims(input_ids, axis=[-1])
embedding_table = tf.get_variable(
name=word_embedding_name,
shape=[vocab_size, embedding_size],
initializer=create_initializer(initializer_range))
if use_one_hot_embeddings:
flat_input_ids = tf.reshape(input_ids, [-1])
one_hot_input_ids = tf.one_hot(flat_input_ids, depth=vocab_size)
output = tf.matmul(one_hot_input_ids, embedding_table)
else:
output = tf.nn.embedding_lookup(embedding_table, input_ids)
input_shape = get_shape_list(input_ids)
output = tf.reshape(output,
input_shape[0:-1] + [input_shape[-1] * embedding_size])
return (output, embedding_table)
構造embedding_table，進行word embedding，可選one_hot的方式，返回embedding的結果和embedding_table
3、詞向量的後續處理
def embedding_postprocessor(input_tensor,
use_token_type=False,
token_type_ids=None,
token_type_vocab_size=16,
token_type_embedding_name="token_type_embeddings",
use_position_embeddings=True,
position_embedding_name="position_embeddings",
initializer_range=0.02,
max_position_embeddings=512,
dropout_prob=0.1):
input_shape = get_shape_list(input_tensor, expected_rank=3)
batch_size = input_shape[0]
seq_length = input_shape[1]
width = input_shape[2]
output = input_tensor
if use_token_type:
if token_type_ids is None:
raise ValueError("`token_type_ids` must be specified if"
"`use_token_type` is True.")
token_type_table = tf.get_variable(
name=token_type_embedding_name,
shape=[token_type_vocab_size, width],
initializer=create_initializer(initializer_range))
flat_token_type_ids = tf.reshape(token_type_ids, [-1])
one_hot_ids = tf.one_hot(flat_token_type_ids, depth=token_type_vocab_size)
token_type_embeddings = tf.matmul(one_hot_ids, token_type_table)
token_type_embeddings = tf.reshape(token_type_embeddings,
[batch_size, seq_length, width])
output += token_type_embeddings
if use_position_embeddings:
assert_op = tf.assert_less_equal(seq_length, max_position_embeddings)
with tf.control_dependencies([assert_op]):
full_position_embeddings = tf.get_variable(
name=position_embedding_name,
shape=[max_position_embeddings, width],
initializer=create_initializer(initializer_range))
position_embeddings = tf.slice(full_position_embeddings, [0, 0],
[seq_length, -1])
num_dims = len(output.shape.as_list())
position_broadcast_shape = []
for _ in range(num_dims - 2):
position_broadcast_shape.append(1)
position_broadcast_shape.extend([seq_length, width])
position_embeddings = tf.reshape(position_embeddings,
position_broadcast_shape)
output += position_embeddings
output = layer_norm_and_dropout(output, dropout_prob)
return output
主要是信息添加，可以將word的位置和word對應的token type等信息添加到詞向量裏面，並且layer正則化和dropout之後返回
4、構造attention mask
def create_attention_mask_from_input_mask(from_tensor, to_mask):
from_shape = get_shape_list(from_tensor, expected_rank=[2, 3])
batch_size = from_shape[0]
from_seq_length = from_shape[1]
to_shape = get_shape_list(to_mask, expected_rank=2)
to_seq_length = to_shape[1]
to_mask = tf.cast(
tf.reshape(to_mask, [batch_size, 1, to_seq_length]), tf.float32)
broadcast_ones = tf.ones(
shape=[batch_size, from_seq_length, 1], dtype=tf.float32)
mask = broadcast_ones * to_mask
return mask
將shape爲[batch_size, to_seq_length]的2D mask轉換爲一個shape 爲[batch_size,
from_seq_length, to_seq_length] 的3D mask用於attention當中。
5、attention layer
def attention_layer(from_tensor,
to_tensor,
attention_mask=None,
num_attention_heads=1,
size_per_head=512,
query_act=None,
key_act=None,
value_act=None,
attention_probs_dropout_prob=0.0,
initializer_range=0.02,
do_return_2d_tensor=False,
batch_size=None,
from_seq_length=None,
to_seq_length=None):
def transpose_for_scores(input_tensor, batch_size, num_attention_heads,
seq_length, width):
output_tensor = tf.reshape(
input_tensor, [batch_size, seq_length, num_attention_heads, width])
output_tensor = tf.transpose(output_tensor, [0, 2, 1, 3])
return output_tensor
from_shape = get_shape_list(from_tensor, expected_rank=[2, 3])
to_shape = get_shape_list(to_tensor, expected_rank=[2, 3])
if len(from_shape) != len(to_shape):
raise ValueError(
"The rank of `from_tensor` must match the rank of `to_tensor`.")
if len(from_shape) == 3:
batch_size = from_shape[0]
from_seq_length = from_shape[1]
to_seq_length = to_shape[1]
elif len(from_shape) == 2:
if (batch_size is None or from_seq_length is None or to_seq_length is None):
raise ValueError(
"When passing in rank 2 tensors to attention_layer, the values "
"for `batch_size`, `from_seq_length`, and `to_seq_length` "
"must all be specified.")
# Scalar dimensions referenced here:
# B = batch size (number of sequences)
# F = `from_tensor` sequence length
# T = `to_tensor` sequence length
# N = `num_attention_heads`
# H = `size_per_head`
from_tensor_2d = reshape_to_matrix(from_tensor)
to_tensor_2d = reshape_to_matrix(to_tensor)
# `query_layer` = [B*F, N*H]
query_layer = tf.layers.dense(
from_tensor_2d,
num_attention_heads * size_per_head,
activation=query_act,
name="query",
kernel_initializer=create_initializer(initializer_range))
# `key_layer` = [B*T, N*H]
key_layer = tf.layers.dense(
to_tensor_2d,
num_attention_heads * size_per_head,
activation=key_act,
name="key",
kernel_initializer=create_initializer(initializer_range))
# `value_layer` = [B*T, N*H]
value_layer = tf.layers.dense(
to_tensor_2d,
num_attention_heads * size_per_head,
activation=value_act,
name="value",
kernel_initializer=create_initializer(initializer_range))
# `query_layer` = [B, N, F, H]
query_layer = transpose_for_scores(query_layer, batch_size,
num_attention_heads, from_seq_length,
size_per_head)
# `key_layer` = [B, N, T, H]
key_layer = transpose_for_scores(key_layer, batch_size, num_attention_heads,
to_seq_length, size_per_head)
attention_scores = tf.matmul(query_layer, key_layer, transpose_b=True)
attention_scores = tf.multiply(attention_scores,
1.0 / math.sqrt(float(size_per_head)))
if attention_mask is not None:
# `attention_mask` = [B, 1, F, T]
attention_mask = tf.expand_dims(attention_mask, axis=[1])
adder = (1.0 - tf.cast(attention_mask, tf.float32)) * -10000.0
attention_scores += adder
attention_probs = tf.nn.softmax(attention_scores)
attention_probs = dropout(attention_probs, attention_probs_dropout_prob)
# `value_layer` = [B, T, N, H]
value_layer = tf.reshape(
value_layer,
[batch_size, to_seq_length, num_attention_heads, size_per_head])
# `value_layer` = [B, N, T, H]
value_layer = tf.transpose(value_layer, [0, 2, 1, 3])
# `context_layer` = [B, N, F, H]
context_layer = tf.matmul(attention_probs, value_layer)
# `context_layer` = [B, F, N, H]
context_layer = tf.transpose(context_layer, [0, 2, 1, 3])
if do_return_2d_tensor:
# `context_layer` = [B*F, N*V]
context_layer = tf.reshape(
context_layer,
[batch_size * from_seq_length, num_attention_heads * size_per_head])
else:
# `context_layer` = [B, F, N*V]
context_layer = tf.reshape(
context_layer,
[batch_size, from_seq_length, num_attention_heads * size_per_head])
return context_layer
整個網絡的重頭戲來了！tansformer的主要內容都在這裏面，輸入的from_tensor當作query，to_tensor當作key和value。當self attention的時候from_tensor和to_tensor是同一個值。
（1）函數一開始對輸入的shape進行校驗，獲取batch_size、from_seq_length 、to_seq_length 。輸入如果是3D張量則轉化成2D矩陣(以輸入爲word_embedding爲例[batch_size, seq_lenth, hidden_size] -> [batch_size*seq_lenth, hidden_size])
（2）通過全連接線性投影生成query_layer、key_layer 、value_layer，輸出的第二個維度變成num_attention_heads * size_per_head（整個模型默認hidden_size=num_attention_heads * size_per_head）。然後通過transpose_for_scores轉換成多頭。
（3）根據公式計算attention_probs（attention score）：
Attention Score計算公式
如果attention_mask is not None，對mask的部分加上一個很大的負數，這樣softmax之後相應的概率值接近爲0，再dropout。
（4）最後再將value和attention_probs相乘，返回3D張量或者2D矩陣
總結：
同學們可以將這段代碼與網絡結構圖對照起來看：
Attention Layer
該函數相比其他版本的的transformer很多地方都有簡化，有以下四點：
（1）缺少scale的操作；
（2）沒有Causality mask，個人猜測主要是bert沒有decoder的操作，所以對角矩陣mask是不需要的，從另一方面來說正好體現了雙向transformer的特點；
（3）沒有query mask。跟（2）理由類似，encoder都是self attention，query和key相同所以只需要一次key mask就夠了
（4）沒有query的Residual層和normalize
6、Transformer
def transformer_model(input_tensor,
attention_mask=None,
hidden_size=768,
num_hidden_layers=12,
num_attention_heads=12,
intermediate_size=3072,
intermediate_act_fn=gelu,
hidden_dropout_prob=0.1,
attention_probs_dropout_prob=0.1,
initializer_range=0.02,
do_return_all_layers=False):
if hidden_size % num_attention_heads != 0:
raise ValueError(
"The hidden size (%d) is not a multiple of the number of attention "
"heads (%d)" % (hidden_size, num_attention_heads))
attention_head_size = int(hidden_size / num_attention_heads)
input_shape = get_shape_list(input_tensor, expected_rank=3)
batch_size = input_shape[0]
seq_length = input_shape[1]
input_width = input_shape[2]
if input_width != hidden_size:
raise ValueError("The width of the input tensor (%d) != hidden size (%d)" %
(input_width, hidden_size))
prev_output = reshape_to_matrix(input_tensor)
all_layer_outputs = []
for layer_idx in range(num_hidden_layers):
with tf.variable_scope("layer_%d" % layer_idx):
layer_input = prev_output
with tf.variable_scope("attention"):
attention_heads = []
with tf.variable_scope("self"):
attention_head = attention_layer(
from_tensor=layer_input,
to_tensor=layer_input,
attention_mask=attention_mask,
num_attention_heads=num_attention_heads,
size_per_head=attention_head_size,
attention_probs_dropout_prob=attention_probs_dropout_prob,
initializer_range=initializer_range,
do_return_2d_tensor=True,
batch_size=batch_size,
from_seq_length=seq_length,
to_seq_length=seq_length)
attention_heads.append(attention_head)
attention_output = None
if len(attention_heads) == 1:
attention_output = attention_heads[0]
else:
attention_output = tf.concat(attention_heads, axis=-1)
with tf.variable_scope("output"):
attention_output = tf.layers.dense(
attention_output,
hidden_size,
kernel_initializer=create_initializer(initializer_range))
attention_output = dropout(attention_output, hidden_dropout_prob)
attention_output = layer_norm(attention_output + layer_input)
with tf.variable_scope("intermediate"):
intermediate_output = tf.layers.dense(
attention_output,
intermediate_size,
activation=intermediate_act_fn,
kernel_initializer=create_initializer(initializer_range))
with tf.variable_scope("output"):
layer_output = tf.layers.dense(
intermediate_output,
hidden_size,
kernel_initializer=create_initializer(initializer_range))
layer_output = dropout(layer_output, hidden_dropout_prob)
layer_output = layer_norm(layer_output + attention_output)
prev_output = layer_output
all_layer_outputs.append(layer_output)
if do_return_all_layers:
final_outputs = []
for layer_output in all_layer_outputs:
final_output = reshape_from_matrix(layer_output, input_shape)
final_outputs.append(final_output)
return final_outputs
else:
final_output = reshape_from_matrix(prev_output, input_shape)
return final_output
transformer是對attention的利用，分以下幾步：
（1）計算attention_head_size，attention_head_size = int(hidden_size / num_attention_heads)即將隱層的輸出等分給各個attention頭。然後將input_tensor轉換成2D矩陣；
（2）對input_tensor進行多頭attention操作，再做：線性投影——dropout——layer norm——intermediate線性投影——線性投影——dropout——attention_output的residual——layer norm
其中intermediate線性投影的hidden_size可以自行指定，其他層的線性投影hidden_size需要統一，目的是爲了對齊。
（3）如此循環計算若干次，且保存每一次的輸出，最後返回所有層的輸出或者最後一層的輸出。
總結：
進一步證實該函數transformer只存在encoder，而不存在decoder操作，所以所有層的多頭attention操作都是基於self encoder的。對應論文紅框的部分：
The Transformer - model architecture
7、BertModel
class BertModel(object):
def __init__(self,
config,
is_training,
input_ids,
input_mask=None,
token_type_ids=None,
use_one_hot_embeddings=True,
scope=None):
config = copy.deepcopy(config)
if not is_training:
config.hidden_dropout_prob = 0.0
config.attention_probs_dropout_prob = 0.0
input_shape = get_shape_list(input_ids, expected_rank=2)
batch_size = input_shape[0]
seq_length = input_shape[1]
if input_mask is None:
input_mask = tf.ones(shape=[batch_size, seq_length], dtype=tf.int32)
if token_type_ids is None:
token_type_ids = tf.zeros(shape=[batch_size, seq_length], dtype=tf.int32)
with tf.variable_scope(scope, default_name="bert"):
with tf.variable_scope("embeddings"):
(self.embedding_output, self.embedding_table) = embedding_lookup(
input_ids=input_ids,
vocab_size=config.vocab_size,
embedding_size=config.hidden_size,
initializer_range=config.initializer_range,
word_embedding_name="word_embeddings",
use_one_hot_embeddings=use_one_hot_embeddings)
self.embedding_output = embedding_postprocessor(
input_tensor=self.embedding_output,
use_token_type=True,
token_type_ids=token_type_ids,
token_type_vocab_size=config.type_vocab_size,
token_type_embedding_name="token_type_embeddings",
use_position_embeddings=True,
position_embedding_name="position_embeddings",
initializer_range=config.initializer_range,
max_position_embeddings=config.max_position_embeddings,
dropout_prob=config.hidden_dropout_prob)
with tf.variable_scope("encoder"):
attention_mask = create_attention_mask_from_input_mask(
input_ids, input_mask)
self.all_encoder_layers = transformer_model(
input_tensor=self.embedding_output,
attention_mask=attention_mask,
hidden_size=config.hidden_size,
num_hidden_layers=config.num_hidden_layers,
num_attention_heads=config.num_attention_heads,
intermediate_size=config.intermediate_size,
intermediate_act_fn=get_activation(config.hidden_act),
hidden_dropout_prob=config.hidden_dropout_prob,
attention_probs_dropout_prob=config.attention_probs_dropout_prob,
initializer_range=config.initializer_range,
do_return_all_layers=True)
self.sequence_output = self.all_encoder_layers[-1]
with tf.variable_scope("pooler"):
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))
終於到模型入口了。
(1）設置各種參數，如果input_mask爲None的話，就指定所有input_mask值爲1，即不進行過濾；如果token_type_ids是None的話，就指定所有token_type_ids值爲0；
(2）對輸入的input_ids進行embedding操作，再embedding_postprocessor操作，前面我們說了。主要是加入位置和token_type信息到詞向量裏面；
(3）轉換attention_mask 後，通過調用transformer_model進行encoder操作；
(4）獲取最後一層的輸出sequence_output和pooled_output，pooled_output是取sequence_output的第一個切片然後線性投影獲得（可以用於分類問題）
8、總結：
（1）bert主要流程是先embedding（包括位置和token_type的embedding），然後調用transformer得到輸出結果，其中embedding、embedding_table、所有transformer層輸出、最後transformer層輸出以及pooled_output都可以獲得，用於遷移學習的fine-tune和預測任務；
（2）bert對於transformer的使用僅限於encoder，沒有decoder的過程。這是因爲模型存粹是爲了預訓練服務，而預訓練是通過語言模型，不同於NLP其他特定任務。在做遷移學習時可以自行添加；
（3）正因爲沒有decoder的操作，所以在attention函數裏面也相應地減少了很多不必要的功能。