Skip to content
体验新版
项目
组织
正在加载...
登录
切换导航
打开侧边栏
PaddlePaddle
Paddle
提交
b0ad9c90
P
Paddle
项目概览
PaddlePaddle
/
Paddle
12 个月 前同步成功
通知
2292
Star
20931
Fork
5422
代码
文件
提交
分支
Tags
贡献者
分支图
Diff
Issue
1423
列表
看板
标记
里程碑
合并请求
543
Wiki
0
Wiki
分析
仓库
DevOps
项目成员
Pages
P
Paddle
项目概览
项目概览
详情
发布
仓库
仓库
文件
提交
分支
标签
贡献者
分支图
比较
Issue
1,423
Issue
1,423
列表
看板
标记
里程碑
合并请求
543
合并请求
543
Pages
分析
分析
仓库分析
DevOps
Wiki
0
Wiki
成员
成员
收起侧边栏
关闭侧边栏
动态
分支图
创建新Issue
提交
Issue看板
体验新版 GitCode,发现更多精彩内容 >>
提交
b0ad9c90
编写于
6月 28, 2017
作者:
C
caoying03
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
enable intializing memory state for lstmemory_group.
上级
633082ad
变更
2
隐藏空白更改
内联
并排
Showing
2 changed file
with
57 addition
and
42 deletion
+57
-42
python/paddle/trainer_config_helpers/layers.py
python/paddle/trainer_config_helpers/layers.py
+26
-25
python/paddle/trainer_config_helpers/networks.py
python/paddle/trainer_config_helpers/networks.py
+31
-17
未找到文件。
python/paddle/trainer_config_helpers/layers.py
浏览文件 @
b0ad9c90
...
...
@@ -1149,10 +1149,10 @@ def pooling_layer(input,
@
layer_support
(
DROPOUT
)
def
lstmemory
(
input
,
name
=
None
,
size
=
None
,
reverse
=
False
,
act
=
None
,
gate_act
=
None
,
size
=
None
,
state_act
=
None
,
bias_attr
=
None
,
param_attr
=
None
,
...
...
@@ -1194,6 +1194,8 @@ def lstmemory(input,
:param name: The lstmemory layer name.
:type name: basestring
:param size: DEPRECATED. size of the lstm cell
:type size: int
:param input: input layer name.
:type input: LayerOutput
:param reverse: is sequence process reversed or not.
...
...
@@ -1220,15 +1222,15 @@ def lstmemory(input,
assert
state_act
.
support_hppl
assert
act
.
support_hppl
assert
input
.
size
is
not
None
and
input
.
size
%
4
==
0
if
size
is
not
None
:
if
input
.
size
/
4
==
size
:
plog
=
logger
.
warning
else
:
plog
=
logger
.
fatal
plog
(
"NOTE: The lstmemory layer[%s]'s size is set by previous input "
"layer. The lstm size should be equal with input layer size/4. The"
" size which is set explicitly will be ignored."
%
name
)
plog
(
"size of lstmemory layer: %s is automatically set to "
"size of input layer / 4. The parameter size passing to "
"this layer is ignored."
%
(
name
))
Layer
(
name
=
name
,
...
...
@@ -1255,11 +1257,11 @@ def lstmemory(input,
@
wrap_name_default
(
"gru"
)
@
layer_support
(
DROPOUT
)
def
grumemory
(
input
,
size
=
None
,
name
=
None
,
reverse
=
False
,
act
=
None
,
gate_act
=
None
,
size
=
None
,
bias_attr
=
None
,
param_attr
=
None
,
layer_attr
=
None
):
...
...
@@ -1318,6 +1320,8 @@ def grumemory(input,
:type name: None|basestring
:param input: input layer.
:type input: LayerOutput.
:param size: DEPRECATED. size of the gru cell
:type size: int
:param reverse: Whether sequence process is reversed or not.
:type reverse: bool
:param act: activation type, TanhActivation by default. This activation
...
...
@@ -1334,9 +1338,6 @@ def grumemory(input,
:type param_attr: ParameterAttribute|None|False
:param layer_attr: Extra Layer attribute
:type layer_attr: ExtraLayerAttribute|None
:param size: Stub parameter of size, but actually not used. If set this size
will get a warning.
:type size: None
:return: LayerOutput object.
:rtype: LayerOutput
"""
...
...
@@ -1348,9 +1349,9 @@ def grumemory(input,
plog
=
logger
.
warning
else
:
plog
=
logger
.
fatal
plog
(
"
NOTE: the gru memory layer's size is set by previous input layer,
"
"
and should be input size / 3. Set size explicitly will be
"
"
ignored."
)
plog
(
"
size of grumemory layer: %s is automatically set to
"
"
size of input layer / 3. The parameter size passing to this
"
"
layer is ignored."
%
(
name
)
)
Layer
(
name
=
name
,
...
...
@@ -2524,8 +2525,8 @@ def img_cmrnorm_layer(input,
@
wrap_bias_attr_default
()
@
wrap_param_attr_default
(
default_factory
=
lambda
_
:
ParamAttr
(
initial_mean
=
1.0
,
initial_std
=
0.
))
@
wrap_param_attr_default
(
default_factory
=
lambda
_
:
ParamAttr
(
initial_mean
=
1.0
,
initial_std
=
0.
))
@
wrap_act_default
(
act
=
ReluActivation
())
@
wrap_name_default
(
"batch_norm"
)
@
layer_support
(
DROPOUT
)
...
...
@@ -3013,25 +3014,25 @@ def lstm_step_layer(input,
bias_attr
=
None
,
layer_attr
=
None
):
"""
LSTM Step Layer.
It used in recurrent_group. The lstm equations are shown
as follow
.
LSTM Step Layer.
This function is used only in recurrent_group.
The lstm equations are shown as follows
.
.. math::
i_t & =
\\
sigma(W_{x
i}x_{t} + W_{hi}h_{t-1} + W_{c
i}c_{t-1} + b_i)
i_t & =
\\
sigma(W_{x
_i}x_{t} + W_{h_i}h_{t-1} + W_{c_
i}c_{t-1} + b_i)
f_t & =
\\
sigma(W_{x
f}x_{t} + W_{hf}h_{t-1} + W_{c
f}c_{t-1} + b_f)
f_t & =
\\
sigma(W_{x
_f}x_{t} + W_{h_f}h_{t-1} + W_{c_
f}c_{t-1} + b_f)
c_t & = f_tc_{t-1} + i_t tanh (W_{x
c}x_t+W_{h
c}h_{t-1} + b_c)
c_t & = f_tc_{t-1} + i_t tanh (W_{x
_c}x_t+W_{h_
c}h_{t-1} + b_c)
o_t & =
\\
sigma(W_{x
o}x_{t} + W_{ho}h_{t-1} + W_{c
o}c_t + b_o)
o_t & =
\\
sigma(W_{x
_o}x_{t} + W_{h_o}h_{t-1} + W_{c_
o}c_t + b_o)
h_t & = o_t tanh(c_t)
The input of lstm step is :math:`Wx_t + Wh_{t-1}`, and user should use
:code:`mixed_layer` and :code:`full_matrix_projection` to calculate these
input vector.
input vector
s
.
The state of lstm step is :math:`c_{t-1}`. And lstm step layer will do
...
...
@@ -3042,14 +3043,14 @@ def lstm_step_layer(input,
...
This layer
contain
s two outputs. Default output is :math:`h_t`. The other
output is :math:`o_t`, wh
ich
name is 'state' and can use
This layer
ha
s two outputs. Default output is :math:`h_t`. The other
output is :math:`o_t`, wh
ose
name is 'state' and can use
:code:`get_output_layer` to extract this output.
:param name: Layer's name.
:type name: basestring
:param size: Layer's size. NOTE: lstm layer's size, should be equal
as
:code:`input.size/4`, and should be equal
as
:param size: Layer's size. NOTE: lstm layer's size, should be equal
to
:code:`input.size/4`, and should be equal
to
:code:`state.size`.
:type size: int
:param input: input layer. :math:`Wx_t + Wh_{t-1}`
...
...
python/paddle/trainer_config_helpers/networks.py
浏览文件 @
b0ad9c90
...
...
@@ -614,6 +614,7 @@ def simple_lstm(input,
@
wrap_name_default
(
'lstm_unit'
)
def
lstmemory_unit
(
input
,
memory_boot
=
None
,
name
=
None
,
size
=
None
,
param_attr
=
None
,
...
...
@@ -626,9 +627,9 @@ def lstmemory_unit(input,
lstm_layer_attr
=
None
,
get_output_layer_attr
=
None
):
"""
Define calculations that a LSTM unit performs
in
a single time step.
This function itself is not a recurrent layer, so
that
it can not be
directly
applied to sequence input
. This function is always used in
Define calculations that a LSTM unit performs
during
a single time step.
This function itself is not a recurrent layer, so it can not be
directly
used to process sequence inputs
. This function is always used in
recurrent_group (see layers.py for more details) to implement attention
mechanism.
...
...
@@ -638,13 +639,13 @@ def lstmemory_unit(input,
.. math::
i_t & =
\\
sigma(W_{x
i}x_{t} + W_{hi}h_{t-1} + W_{c
i}c_{t-1} + b_i)
i_t & =
\\
sigma(W_{x
_i}x_{t} + W_{h_i}h_{t-1} + W_{c_
i}c_{t-1} + b_i)
f_t & =
\\
sigma(W_{x
f}x_{t} + W_{hf}h_{t-1} + W_{c
f}c_{t-1} + b_f)
f_t & =
\\
sigma(W_{x
_f}x_{t} + W_{h_f}h_{t-1} + W_{c_
f}c_{t-1} + b_f)
c_t & = f_tc_{t-1} + i_t tanh (W_{x
c}x_t+W_{h
c}h_{t-1} + b_c)
c_t & = f_tc_{t-1} + i_t tanh (W_{x
_c}x_t+W_{h_
c}h_{t-1} + b_c)
o_t & =
\\
sigma(W_{x
o}x_{t} + W_{ho}h_{t-1} + W_{c
o}c_t + b_o)
o_t & =
\\
sigma(W_{x
_o}x_{t} + W_{h_o}h_{t-1} + W_{c_
o}c_t + b_o)
h_t & = o_t tanh(c_t)
...
...
@@ -661,6 +662,8 @@ def lstmemory_unit(input,
:param input: input layer name.
:type input: LayerOutput
:param memory_boot: the initialization state of the LSTM cell.
:type memory_boot: LayerOutput | None
:param name: lstmemory unit name.
:type name: basestring
:param size: lstmemory unit size.
...
...
@@ -692,7 +695,8 @@ def lstmemory_unit(input,
assert
input
.
size
%
4
==
0
size
=
input
.
size
/
4
out_mem
=
memory
(
name
=
name
,
size
=
size
)
state_mem
=
memory
(
name
=
"%s_state"
%
name
,
size
=
size
)
state_mem
=
memory
(
name
=
"%s_state"
%
name
,
size
=
size
,
boot_layer
=
memory_boot
)
with
mixed_layer
(
name
=
"%s_input_recurrent"
%
name
,
...
...
@@ -726,6 +730,7 @@ def lstmemory_unit(input,
def
lstmemory_group
(
input
,
size
=
None
,
name
=
None
,
memory_boot
=
None
,
reverse
=
False
,
param_attr
=
None
,
act
=
None
,
...
...
@@ -737,7 +742,7 @@ def lstmemory_group(input,
lstm_layer_attr
=
None
,
get_output_layer_attr
=
None
):
"""
lstm_group is a recurrent
layer
group version of Long Short Term Memory. It
lstm_group is a recurrent
_
group version of Long Short Term Memory. It
does exactly the same calculation as the lstmemory layer (see lstmemory in
layers.py for the maths) does. A promising benefit is that LSTM memory
cell states, or hidden states in every time step are accessible to the
...
...
@@ -748,8 +753,8 @@ def lstmemory_group(input,
NOTE: In PaddlePaddle's implementation, the following input-to-hidden
multiplications:
:math:`W_{x
i}x_{t}` , :math:`W_{x
f}x_{t}`,
:math:`W_{x
c}x_t`, :math:`W_{x
o}x_{t}` are not done in lstmemory_unit to
:math:`W_{x
_i}x_{t}` , :math:`W_{x_
f}x_{t}`,
:math:`W_{x
_c}x_t`, :math:`W_{x_
o}x_{t}` are not done in lstmemory_unit to
speed up the calculations. Consequently, an additional mixed_layer with
full_matrix_projection must be included before lstmemory_unit is called.
...
...
@@ -765,8 +770,10 @@ def lstmemory_group(input,
:param input: input layer name.
:type input: LayerOutput
:param name:
lstmemory group name
.
:param name:
name of the lstmemory group
.
:type name: basestring
:param memory_boot: the initialization state of LSTM cell.
:type memory_boot: LayerOutput | None
:param size: lstmemory group size.
:type size: int
:param reverse: is lstm reversed
...
...
@@ -798,6 +805,7 @@ def lstmemory_group(input,
def
__lstm_step__
(
ipt
):
return
lstmemory_unit
(
input
=
ipt
,
memory_boot
=
memory_boot
,
name
=
name
,
size
=
size
,
mixed_bias_attr
=
mixed_bias_attr
,
...
...
@@ -819,6 +827,7 @@ def lstmemory_group(input,
@
wrap_name_default
(
'gru_unit'
)
def
gru_unit
(
input
,
memory_boot
=
None
,
size
=
None
,
name
=
None
,
gru_bias_attr
=
None
,
...
...
@@ -829,8 +838,8 @@ def gru_unit(input,
naive
=
False
):
"""
Define calculations that a gated recurrent unit performs in a single time
step. This function itself is not a recurrent layer, so
that
it can not be
directly
applied to sequence input. This function is almost
always used in
step. This function itself is not a recurrent layer, so it can not be
directly
used to process sequence inputs. This function is
always used in
the recurrent_group (see layers.py for more details) to implement attention
mechanism.
...
...
@@ -838,6 +847,8 @@ def gru_unit(input,
:param input: input layer name.
:type input: LayerOutput
:param memory_boot: the initialization state of the LSTM cell.
:type memory_boot: LayerOutput | None
:param name: name of the gru group.
:type name: basestring
:param size: hidden size of the gru.
...
...
@@ -856,7 +867,7 @@ def gru_unit(input,
if
size
is
None
:
size
=
input
.
size
/
3
out_mem
=
memory
(
name
=
name
,
size
=
size
)
out_mem
=
memory
(
name
=
name
,
size
=
size
,
boot_layer
=
memory_boot
)
if
naive
:
__step__
=
gru_step_naive_layer
...
...
@@ -878,6 +889,7 @@ def gru_unit(input,
@
wrap_name_default
(
'gru_group'
)
def
gru_group
(
input
,
memory_boot
=
None
,
size
=
None
,
name
=
None
,
reverse
=
False
,
...
...
@@ -888,7 +900,7 @@ def gru_group(input,
gru_layer_attr
=
None
,
naive
=
False
):
"""
gru_group is a recurrent
layer
group version of Gated Recurrent Unit. It
gru_group is a recurrent
_
group version of Gated Recurrent Unit. It
does exactly the same calculation as the grumemory layer does. A promising
benefit is that gru hidden states are accessible to the user. This is
especially useful in attention model. If you do not need to access
...
...
@@ -908,6 +920,8 @@ def gru_group(input,
:param input: input layer name.
:type input: LayerOutput
:param memory_boot: the initialization state of the LSTM cell.
:type memory_boot: LayerOutput | None
:param name: name of the gru group.
:type name: basestring
:param size: hidden size of the gru.
...
...
@@ -929,6 +943,7 @@ def gru_group(input,
def
__gru_step__
(
ipt
):
return
gru_unit
(
input
=
ipt
,
memory_boot
=
memory_boot
,
name
=
name
,
size
=
size
,
gru_bias_attr
=
gru_bias_attr
,
...
...
@@ -1083,7 +1098,6 @@ def simple_gru2(input,
return
grumemory
(
name
=
name
,
size
=
size
,
input
=
m
,
reverse
=
reverse
,
bias_attr
=
gru_bias_attr
,
...
...
编辑
预览
Markdown
is supported
0%
请重试
或
添加新附件
.
添加附件
取消
You are about to add
0
people
to the discussion. Proceed with caution.
先完成此消息的编辑!
取消
想要评论请
注册
或
登录