Files
2026-07-13 12:40:42 +08:00

599 lines
16 KiB
Python

# Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
from functools import partial
import numpy as np
import paddle
pos_enc_param_names = (
"src_pos_enc_table",
"trg_pos_enc_table",
)
batch_size = 2
def position_encoding_init(n_position, d_pos_vec):
"""
Generate the initial values for the sinusoid position encoding table.
"""
position_enc = np.array(
[
(
[
pos / np.power(10000, 2 * (j // 2) / d_pos_vec)
for j in range(d_pos_vec)
]
if pos != 0
else np.zeros(d_pos_vec)
)
for pos in range(n_position)
]
)
position_enc[1:, 0::2] = np.sin(position_enc[1:, 0::2]) # dim 2i
position_enc[1:, 1::2] = np.cos(position_enc[1:, 1::2]) # dim 2i+1
return position_enc.astype("float32")
def multi_head_attention(
queries,
keys,
values,
attn_bias,
d_key,
d_value,
d_model,
n_head=1,
dropout_rate=0.0,
):
"""
Multi-Head Attention. Note that attn_bias is added to the logit before
computing softmax activation to mask certain selected positions so that
they will not considered in attention weights.
"""
if not (len(queries.shape) == len(keys.shape) == len(values.shape) == 3):
raise ValueError(
"Inputs: queries, keys and values should all be 3-D tensors."
)
def __compute_qkv(queries, keys, values, n_head, d_key, d_value):
"""
Add linear projection to queries, keys, and values.
"""
queries_flatten = paddle.nn.Flatten(start_axis=2)(queries)
q = paddle.nn.Linear(
in_features=queries_flatten.shape[-1],
out_features=d_key * n_head,
weight_attr=paddle.nn.initializer.XavierNormal(
fan_in=d_model * d_key, fan_out=n_head * d_key
),
bias_attr=False,
)(queries_flatten)
keys_flatten = paddle.nn.Flatten(start_axis=2)(keys)
k = paddle.nn.Linear(
in_features=keys_flatten.shape[-1],
out_features=d_key * n_head,
weight_attr=paddle.nn.initializer.XavierNormal(
fan_in=d_model * d_key, fan_out=n_head * d_key
),
bias_attr=False,
)(keys_flatten)
values_flatten = paddle.nn.Flatten(start_axis=2)(values)
v = paddle.nn.Linear(
in_features=values_flatten.shape[-1],
out_features=d_value * n_head,
weight_attr=paddle.nn.initializer.XavierNormal(
fan_in=d_model * d_value, fan_out=n_head * d_value
),
bias_attr=False,
)(values_flatten)
return q, k, v
def __split_heads(x, n_head):
"""
Reshape the last dimension of input tensor x so that it becomes two
dimensions and then transpose. Specifically, input a tensor with shape
[bs, max_sequence_length, n_head * hidden_dim] then output a tensor
with shape [bs, n_head, max_sequence_length, hidden_dim].
"""
if n_head == 1:
return x
hidden_size = x.shape[-1]
# FIXME(guosheng): Decouple the program desc with batch_size.
reshaped = paddle.reshape(
x=x, shape=[batch_size, -1, n_head, hidden_size // n_head]
)
# permute the dimensions into:
# [batch_size, n_head, max_sequence_len, hidden_size_per_head]
return paddle.transpose(x=reshaped, perm=[0, 2, 1, 3])
def __combine_heads(x):
"""
Transpose and then reshape the last two dimensions of input tensor x
so that it becomes one dimension, which is reverse to __split_heads.
"""
if len(x.shape) == 3:
return x
if len(x.shape) != 4:
raise ValueError("Input(x) should be a 4-D Tensor.")
trans_x = paddle.transpose(x, perm=[0, 2, 1, 3])
# FIXME(guosheng): Decouple the program desc with batch_size.
return paddle.reshape(
x=trans_x,
shape=list(
map(int, [batch_size, -1, trans_x.shape[2] * trans_x.shape[3]])
),
)
def scaled_dot_product_attention(q, k, v, attn_bias, d_model, dropout_rate):
"""
Scaled Dot-Product Attention
"""
# FIXME(guosheng): Optimize the shape in reshape_op or softmax_op.
# The current implementation of softmax_op only supports 2D tensor,
# consequently it cannot be directly used here.
# If to use the reshape_op, Besides, the shape of product inferred in
# compile-time is not the actual shape in run-time. It can't be used
# to set the attribute of reshape_op.
# So, here define the softmax for temporary solution.
def __softmax(x, eps=1e-9):
exp_out = paddle.exp(x=x)
sum_out = paddle.sum(exp_out, axis=-1, keepdim=True)
return paddle.divide(x=exp_out, y=sum_out)
scaled_q = paddle.scale(x=q, scale=d_model**-0.5)
product = paddle.matmul(x=scaled_q, y=k, transpose_y=True)
weights = __softmax(paddle.add(x=product, y=attn_bias))
if dropout_rate:
weights = paddle.nn.functional.dropout(weights, p=dropout_rate)
out = paddle.matmul(weights, v)
return out
q, k, v = __compute_qkv(queries, keys, values, n_head, d_key, d_value)
q = __split_heads(q, n_head)
k = __split_heads(k, n_head)
v = __split_heads(v, n_head)
ctx_multiheads = scaled_dot_product_attention(
q, k, v, attn_bias, d_model, dropout_rate
)
out = __combine_heads(ctx_multiheads)
# Project back to the model size.
out_flatten = paddle.nn.Flatten(start_axis=2)(out)
proj_out = paddle.nn.Linear(
in_features=out_flatten.shape[-1],
out_features=d_model,
weight_attr=paddle.nn.initializer.XavierNormal(),
bias_attr=False,
)(out_flatten)
return proj_out
def positionwise_feed_forward(x, d_inner_hid, d_hid):
"""
Position-wise Feed-Forward Networks.
This module consists of two linear transformations with a ReLU activation
in between, which is applied to each position separately and identically.
"""
x_flatten = paddle.nn.Flatten(start_axis=2)(x)
hidden_l = paddle.nn.Linear(
in_features=x_flatten.shape[-1],
out_features=d_inner_hid,
weight_attr=paddle.nn.initializer.Uniform(
low=-(d_hid**-0.5), high=(d_hid**-0.5)
),
)(x_flatten)
hidden = paddle.nn.ReLU()(hidden_l)
hidden_flatten = paddle.nn.Flatten(start_axis=2)(hidden)
out = paddle.nn.Linear(
in_features=hidden_flatten.shape[-1],
out_features=d_hid,
weight_attr=paddle.nn.initializer.Uniform(
low=-(d_inner_hid**-0.5), high=(d_inner_hid**-0.5)
),
)(hidden_flatten)
return out
def pre_post_process_layer(prev_out, out, process_cmd, dropout=0.0):
"""
Add residual connection, layer normalization and dropout to the out tensor
optionally according to the value of process_cmd.
This will be used before or after multi-head attention and position-wise
feed-forward networks.
"""
for cmd in process_cmd:
if cmd == "a": # add residual connection
out = out + prev_out if prev_out else out
elif cmd == "n": # add layer normalization
out = paddle.nn.LayerNorm(
out.shape[-1:],
weight_attr=paddle.nn.initializer.Constant(1.0),
bias_attr=paddle.nn.initializer.Constant(0.0),
)(out)
elif cmd == "d": # add dropout
if dropout:
out = paddle.nn.functional.dropout(out, p=dropout)
return out
pre_process_layer = partial(pre_post_process_layer, None)
post_process_layer = pre_post_process_layer
def prepare_encoder(
src_word,
src_pos,
src_vocab_size,
src_emb_dim,
src_pad_idx,
src_max_len,
dropout=0.0,
pos_pad_idx=0,
pos_enc_param_name=None,
):
"""Add word embeddings and position encodings.
The output tensor has a shape of:
[batch_size, max_src_length_in_batch, d_model].
This module is used at the bottom of the encoder stacks.
"""
src_word_emb = paddle.nn.Embedding(
num_embeddings=src_vocab_size,
embedding_dim=src_emb_dim,
padding_idx=src_pad_idx,
weight_attr=paddle.nn.initializer.Normal(0.0, 1.0),
)(src_word)
src_pos_enc = paddle.nn.Embedding(
num_embeddings=src_max_len,
embedding_dim=src_emb_dim,
padding_idx=pos_pad_idx,
weight_attr=paddle.ParamAttr(name=pos_enc_param_name, trainable=False),
)(src_pos)
src_pos_enc.stop_gradient = True
enc_input = src_word_emb + src_pos_enc
# FIXME(guosheng): Decouple the program desc with batch_size.
enc_input = paddle.reshape(x=enc_input, shape=[batch_size, -1, src_emb_dim])
return (
paddle.nn.functional.dropout(enc_input, p=dropout)
if dropout
else enc_input
)
prepare_encoder = partial(
prepare_encoder, pos_enc_param_name=pos_enc_param_names[0]
)
prepare_decoder = partial(
prepare_encoder, pos_enc_param_name=pos_enc_param_names[1]
)
def encoder_layer(
enc_input,
attn_bias,
n_head,
d_key,
d_value,
d_model,
d_inner_hid,
dropout_rate=0.0,
):
"""The encoder layers that can be stacked to form a deep encoder.
This module consists of a multi-head (self) attention followed by
position-wise feed-forward networks and both the two components companied
with the post_process_layer to add residual connection, layer normalization
and dropout.
"""
attn_output = multi_head_attention(
enc_input,
enc_input,
enc_input,
attn_bias,
d_key,
d_value,
d_model,
n_head,
dropout_rate,
)
attn_output = post_process_layer(
enc_input, attn_output, "dan", dropout_rate
)
ffd_output = positionwise_feed_forward(attn_output, d_inner_hid, d_model)
return post_process_layer(attn_output, ffd_output, "dan", dropout_rate)
def encoder(
enc_input,
attn_bias,
n_layer,
n_head,
d_key,
d_value,
d_model,
d_inner_hid,
dropout_rate=0.0,
):
"""
The encoder is composed of a stack of identical layers returned by calling
encoder_layer.
"""
for i in range(n_layer):
enc_output = encoder_layer(
enc_input,
attn_bias,
n_head,
d_key,
d_value,
d_model,
d_inner_hid,
dropout_rate,
)
enc_input = enc_output
return enc_output
def decoder_layer(
dec_input,
enc_output,
slf_attn_bias,
dec_enc_attn_bias,
n_head,
d_key,
d_value,
d_model,
d_inner_hid,
dropout_rate=0.0,
):
"""The layer to be stacked in decoder part.
The structure of this module is similar to that in the encoder part except
a multi-head attention is added to implement encoder-decoder attention.
"""
slf_attn_output = multi_head_attention(
dec_input,
dec_input,
dec_input,
slf_attn_bias,
d_key,
d_value,
d_model,
n_head,
dropout_rate,
)
slf_attn_output = post_process_layer(
dec_input,
slf_attn_output,
"dan", # residual connection + dropout + layer normalization
dropout_rate,
)
enc_attn_output = multi_head_attention(
slf_attn_output,
enc_output,
enc_output,
dec_enc_attn_bias,
d_key,
d_value,
d_model,
n_head,
dropout_rate,
)
enc_attn_output = post_process_layer(
slf_attn_output,
enc_attn_output,
"dan", # residual connection + dropout + layer normalization
dropout_rate,
)
ffd_output = positionwise_feed_forward(
enc_attn_output,
d_inner_hid,
d_model,
)
dec_output = post_process_layer(
enc_attn_output,
ffd_output,
"dan", # residual connection + dropout + layer normalization
dropout_rate,
)
return dec_output
def decoder(
dec_input,
enc_output,
dec_slf_attn_bias,
dec_enc_attn_bias,
n_layer,
n_head,
d_key,
d_value,
d_model,
d_inner_hid,
dropout_rate=0.0,
):
"""
The decoder is composed of a stack of identical decoder_layer layers.
"""
for i in range(n_layer):
dec_output = decoder_layer(
dec_input,
enc_output,
dec_slf_attn_bias,
dec_enc_attn_bias,
n_head,
d_key,
d_value,
d_model,
d_inner_hid,
dropout_rate,
)
dec_input = dec_output
return dec_output
def build_inputs(max_length, n_head):
names = [
'src_word',
'src_pos',
'trg_word',
'trg_pos',
'src_slf_attn_bias',
'trg_slf_attn_bias',
'trg_src_attn_bias',
'gold',
'weights',
]
shapes = [
[batch_size * max_length, 1],
[batch_size * max_length, 1],
[batch_size * max_length, 1],
[batch_size * max_length, 1],
[batch_size, n_head, max_length, max_length],
[batch_size, n_head, max_length, max_length],
[batch_size, n_head, max_length, max_length],
[batch_size * max_length, 1],
[batch_size * max_length, 1],
]
dtypes = [
'int64',
'int64',
'int64',
'int64',
'float32',
'float32',
'float32',
'int64',
'float32',
]
all_inputs = []
for name, shape, dtype in zip(names, shapes, dtypes):
data_input = paddle.static.data(
name=name,
shape=shape,
dtype=dtype,
)
data_input.desc.set_need_check_feed(False)
all_inputs.append(data_input)
return all_inputs
def transformer(
src_vocab_size,
trg_vocab_size,
max_length,
n_layer,
n_head,
d_key,
d_value,
d_model,
d_inner_hid,
dropout_rate,
src_pad_idx,
trg_pad_idx,
pos_pad_idx,
):
(
src_word,
src_pos,
trg_word,
trg_pos,
src_slf_attn_bias,
trg_slf_attn_bias,
trg_src_attn_bias,
gold,
weights,
) = build_inputs(max_length, n_head)
enc_input = prepare_encoder(
src_word,
src_pos,
src_vocab_size,
d_model,
src_pad_idx,
max_length,
dropout_rate,
)
enc_output = encoder(
enc_input,
src_slf_attn_bias,
n_layer,
n_head,
d_key,
d_value,
d_model,
d_inner_hid,
dropout_rate,
)
dec_input = prepare_decoder(
trg_word,
trg_pos,
trg_vocab_size,
d_model,
trg_pad_idx,
max_length,
dropout_rate,
)
dec_output = decoder(
dec_input,
enc_output,
trg_slf_attn_bias,
trg_src_attn_bias,
n_layer,
n_head,
d_key,
d_value,
d_model,
d_inner_hid,
dropout_rate,
)
# TODO(guosheng): Share the weight matrix between the embedding layers and
# the pre-softmax linear transformation.
dec_output_flatten = paddle.nn.Flatten(start_axis=2)(dec_output)
predict = paddle.reshape(
x=paddle.nn.Linear(
in_features=dec_output_flatten.shape[-1],
out_features=trg_vocab_size,
weight_attr=paddle.nn.initializer.XavierNormal(),
bias_attr=False,
)(dec_output_flatten),
shape=[-1, trg_vocab_size],
)
predict = paddle.nn.functional.softmax(predict)
cost = paddle.nn.functional.cross_entropy(
input=predict, label=gold, reduction='none', use_softmax=False
)
weighted_cost = cost * weights
return paddle.sum(weighted_cost)