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
66ea7184
P
Paddle
项目概览
PaddlePaddle
/
Paddle
大约 1 年 前同步成功
通知
2298
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
66ea7184
编写于
12月 26, 2018
作者:
H
haowang101779990
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
en api improve format Dec 27
test=develop
上级
988bc2b5
变更
9
显示空白变更内容
内联
并排
Showing
9 changed file
with
379 addition
and
291 deletion
+379
-291
python/paddle/fluid/data_feeder.py
python/paddle/fluid/data_feeder.py
+1
-2
python/paddle/fluid/framework.py
python/paddle/fluid/framework.py
+2
-2
python/paddle/fluid/layers/control_flow.py
python/paddle/fluid/layers/control_flow.py
+5
-4
python/paddle/fluid/layers/detection.py
python/paddle/fluid/layers/detection.py
+62
-58
python/paddle/fluid/layers/io.py
python/paddle/fluid/layers/io.py
+5
-6
python/paddle/fluid/layers/nn.py
python/paddle/fluid/layers/nn.py
+266
-201
python/paddle/fluid/layers/tensor.py
python/paddle/fluid/layers/tensor.py
+8
-3
python/paddle/fluid/metrics.py
python/paddle/fluid/metrics.py
+14
-8
python/paddle/fluid/transpiler/distribute_transpiler.py
python/paddle/fluid/transpiler/distribute_transpiler.py
+16
-7
未找到文件。
python/paddle/fluid/data_feeder.py
浏览文件 @
66ea7184
...
@@ -272,8 +272,7 @@ class DataFeeder(object):
...
@@ -272,8 +272,7 @@ class DataFeeder(object):
dict: the result of conversion.
dict: the result of conversion.
Raises:
Raises:
ValueError: If drop_last is False and the data batch which cannot
ValueError: If drop_last is False and the data batch which cannot fit for devices.
fit for devices.
"""
"""
def
__reader_creator__
():
def
__reader_creator__
():
...
...
python/paddle/fluid/framework.py
浏览文件 @
66ea7184
...
@@ -1646,8 +1646,8 @@ class Program(object):
...
@@ -1646,8 +1646,8 @@ class Program(object):
parameters, e.g., :code:`trainable`, :code:`optimize_attr`, need
parameters, e.g., :code:`trainable`, :code:`optimize_attr`, need
to print.
to print.
Returns
Returns
:
(str)
: The debug string.
str
: The debug string.
Raises:
Raises:
ValueError: If any of required fields is not set and throw_on_error is
ValueError: If any of required fields is not set and throw_on_error is
...
...
python/paddle/fluid/layers/control_flow.py
浏览文件 @
66ea7184
...
@@ -1452,6 +1452,7 @@ class DynamicRNN(object):
...
@@ -1452,6 +1452,7 @@ class DynamicRNN(object):
def
step_input
(
self
,
x
):
def
step_input
(
self
,
x
):
"""
"""
Mark a sequence as a dynamic RNN input.
Mark a sequence as a dynamic RNN input.
Args:
Args:
x(Variable): The input sequence.
x(Variable): The input sequence.
...
@@ -1505,6 +1506,7 @@ class DynamicRNN(object):
...
@@ -1505,6 +1506,7 @@ class DynamicRNN(object):
"""
"""
Mark a variable as a RNN input. The input will not be scattered into
Mark a variable as a RNN input. The input will not be scattered into
time steps.
time steps.
Args:
Args:
x(Variable): The input variable.
x(Variable): The input variable.
...
@@ -1629,13 +1631,11 @@ class DynamicRNN(object):
...
@@ -1629,13 +1631,11 @@ class DynamicRNN(object):
Args:
Args:
init(Variable|None): The initialized variable.
init(Variable|None): The initialized variable.
shape(list|tuple): The memory shape. NOTE the shape does not contain
shape(list|tuple): The memory shape. NOTE the shape does not contain batch_size.
batch_size.
value(float): the initalized value.
value(float): the initalized value.
need_reorder(bool): True if the initialized memory depends on the
need_reorder(bool): True if the initialized memory depends on the input sample.
input sample.
dtype(str|numpy.dtype): The data type of the initialized memory.
dtype(str|numpy.dtype): The data type of the initialized memory.
...
@@ -1714,6 +1714,7 @@ class DynamicRNN(object):
...
@@ -1714,6 +1714,7 @@ class DynamicRNN(object):
"""
"""
Update the memory from ex_mem to new_mem. NOTE that the shape and data
Update the memory from ex_mem to new_mem. NOTE that the shape and data
type of :code:`ex_mem` and :code:`new_mem` must be same.
type of :code:`ex_mem` and :code:`new_mem` must be same.
Args:
Args:
ex_mem(Variable): the memory variable.
ex_mem(Variable): the memory variable.
new_mem(Variable): the plain variable generated in RNN block.
new_mem(Variable): the plain variable generated in RNN block.
...
...
python/paddle/fluid/layers/detection.py
浏览文件 @
66ea7184
...
@@ -65,7 +65,7 @@ def rpn_target_assign(bbox_pred,
...
@@ -65,7 +65,7 @@ def rpn_target_assign(bbox_pred,
rpn_negative_overlap
=
0.3
,
rpn_negative_overlap
=
0.3
,
use_random
=
True
):
use_random
=
True
):
"""
"""
**
Target Assign Layer for region proposal network (RPN) in Faster-RCNN detection.
**
**
Target Assign Layer for region proposal network (RPN) in Faster-RCNN detection.
**
This layer can be, for given the Intersection-over-Union (IoU) overlap
This layer can be, for given the Intersection-over-Union (IoU) overlap
between anchors and ground truth boxes, to assign classification and
between anchors and ground truth boxes, to assign classification and
...
@@ -148,6 +148,7 @@ def rpn_target_assign(bbox_pred,
...
@@ -148,6 +148,7 @@ def rpn_target_assign(bbox_pred,
cls_logits=cls_logits,
cls_logits=cls_logits,
anchor_box=anchor_box,
anchor_box=anchor_box,
gt_boxes=gt_boxes)
gt_boxes=gt_boxes)
"""
"""
helper
=
LayerHelper
(
'rpn_target_assign'
,
**
locals
())
helper
=
LayerHelper
(
'rpn_target_assign'
,
**
locals
())
...
@@ -1525,20 +1526,23 @@ def anchor_generator(input,
...
@@ -1525,20 +1526,23 @@ def anchor_generator(input,
anchors, e.g. [0.5, 1.0, 2.0].
anchors, e.g. [0.5, 1.0, 2.0].
variance(list|tuple): The variances to be used in box regression deltas.
variance(list|tuple): The variances to be used in box regression deltas.
Default:[0.1, 0.1, 0.2, 0.2].
Default:[0.1, 0.1, 0.2, 0.2].
stride(list|turple): The anchors stride across width and height,
stride(list|turple): The anchors stride across width and height,e.g. [16.0, 16.0]
e.g. [16.0, 16.0]
offset(float): Prior boxes center offset. Default: 0.5
offset(float): Prior boxes center offset. Default: 0.5
name(str): Name of the prior box op. Default: None.
name(str): Name of the prior box op. Default: None.
Returns:
Returns:
Anchors(Variable): The output anchors with a layout of [H, W, num_anchors, 4].
Anchors(Variable),Variances(Variable):
H is the height of input, W is the width of input,
num_anchors is the box count of each position.
two variables:
- Anchors(Variable): The output anchors with a layout of [H, W, num_anchors, 4].
\
H is the height of input, W is the width of input,
\
num_anchors is the box count of each position.
\
Each anchor is in (xmin, ymin, xmax, ymax) format an unnormalized.
Each anchor is in (xmin, ymin, xmax, ymax) format an unnormalized.
Variances(Variable): The expanded variances of anchors
- Variances(Variable): The expanded variances of anchors
\
with a layout of [H, W, num_priors, 4].
with a layout of [H, W, num_priors, 4].
\
H is the height of input, W is the width of input
H is the height of input, W is the width of input
\
num_anchors is the box count of each position.
num_anchors is the box count of each position.
\
Each variance is in (xcenter, ycenter, w, h) format.
Each variance is in (xcenter, ycenter, w, h) format.
...
@@ -1748,7 +1752,7 @@ def generate_proposals(scores,
...
@@ -1748,7 +1752,7 @@ def generate_proposals(scores,
eta
=
1.0
,
eta
=
1.0
,
name
=
None
):
name
=
None
):
"""
"""
**
Generate proposal Faster-RCNN
**
**
Generate proposal Faster-RCNN
**
This operation proposes RoIs according to each box with their probability to be a foreground object and
This operation proposes RoIs according to each box with their probability to be a foreground object and
the box can be calculated by anchors. Bbox_deltais and scores to be an object are the output of RPN. Final proposals
the box can be calculated by anchors. Bbox_deltais and scores to be an object are the output of RPN. Final proposals
...
@@ -1762,7 +1766,6 @@ def generate_proposals(scores,
...
@@ -1762,7 +1766,6 @@ def generate_proposals(scores,
4. Remove predicted boxes with small area.
4. Remove predicted boxes with small area.
5. Apply NMS to get final proposals as output.
5. Apply NMS to get final proposals as output.
Args:
Args:
scores(Variable): A 4-D Tensor with shape [N, A, H, W] represents the probability for each box to be an object.
scores(Variable): A 4-D Tensor with shape [N, A, H, W] represents the probability for each box to be an object.
N is batch size, A is number of anchors, H and W are height and width of the feature map.
N is batch size, A is number of anchors, H and W are height and width of the feature map.
...
@@ -1777,6 +1780,7 @@ def generate_proposals(scores,
...
@@ -1777,6 +1780,7 @@ def generate_proposals(scores,
nms_thresh(float): Threshold in NMS, 0.5 by default.
nms_thresh(float): Threshold in NMS, 0.5 by default.
min_size(float): Remove predicted boxes with either height or width < min_size. 0.1 by default.
min_size(float): Remove predicted boxes with either height or width < min_size. 0.1 by default.
eta(float): Apply in adaptive NMS, if adaptive threshold > 0.5, adaptive_threshold = adaptive_threshold * eta in each iteration.
eta(float): Apply in adaptive NMS, if adaptive threshold > 0.5, adaptive_threshold = adaptive_threshold * eta in each iteration.
"""
"""
helper
=
LayerHelper
(
'generate_proposals'
,
**
locals
())
helper
=
LayerHelper
(
'generate_proposals'
,
**
locals
())
...
...
python/paddle/fluid/layers/io.py
浏览文件 @
66ea7184
...
@@ -949,12 +949,11 @@ def shuffle(reader, buffer_size):
...
@@ -949,12 +949,11 @@ def shuffle(reader, buffer_size):
is determined by argument buf_size.
is determined by argument buf_size.
Args:
Args:
param reader: the original reader whose output will be shuffled.
reader(callable): the original reader whose output will be shuffled.
type reader: callable
buf_size(int): shuffle buffer size.
param buf_size: shuffle buffer size.
type buf_size: int
Returns:
return: the new reader whose output is shuffled.
callable: the new reader whose output is shuffled.
rtype: callable
"""
"""
return
__create_unshared_decorated_reader__
(
return
__create_unshared_decorated_reader__
(
'create_shuffle_reader'
,
reader
,
{
'buffer_size'
:
int
(
buffer_size
)})
'create_shuffle_reader'
,
reader
,
{
'buffer_size'
:
int
(
buffer_size
)})
...
...
python/paddle/fluid/layers/nn.py
浏览文件 @
66ea7184
...
@@ -233,7 +233,7 @@ def fc(input,
...
@@ -233,7 +233,7 @@ def fc(input,
dimensions will be flatten to form the first dimension of the final matrix (height of
dimensions will be flatten to form the first dimension of the final matrix (height of
the matrix), and the rest `rank(X) - num_flatten_dims` dimensions are flattened to
the matrix), and the rest `rank(X) - num_flatten_dims` dimensions are flattened to
form the second dimension of the final matrix (width of the matrix). For example, suppose
form the second dimension of the final matrix (width of the matrix). For example, suppose
`X` is a
6
-dimensional tensor with a shape [2, 3, 4, 5, 6], and `num_flatten_dims` = 3.
`X` is a
5
-dimensional tensor with a shape [2, 3, 4, 5, 6], and `num_flatten_dims` = 3.
Then, the flattened matrix will have a shape [2 x 3 x 4, 5 x 6] = [24, 30].
Then, the flattened matrix will have a shape [2 x 3 x 4, 5 x 6] = [24, 30].
param_attr (ParamAttr|list of ParamAttr, default None): The parameter attribute for learnable
param_attr (ParamAttr|list of ParamAttr, default None): The parameter attribute for learnable
parameters/weights of this layer.
parameters/weights of this layer.
...
@@ -505,31 +505,33 @@ def lstm(input,
...
@@ -505,31 +505,33 @@ def lstm(input,
In the forward pass the output ht and cell output ct for a given iteration can be computed from the recurrent input ht-1,
In the forward pass the output ht and cell output ct for a given iteration can be computed from the recurrent input ht-1,
the cell input ct-1 and the previous layer input xt given matrices W, R and biases bW, bR from the following equations:
the cell input ct-1 and the previous layer input xt given matrices W, R and biases bW, bR from the following equations:
$$ i_t =
\\
sigma(W_{ix}x_{t} + W_{ih}h_{t-1} + bx_i + bh_i) $$
.. math::
i_t &= \sigma(W_{ix}x_{t} + W_{ih}h_{t-1} + bx_i + bh_i)
$$ f_t =
\\
sigma(W_{fx}x_{t} + W_{fh}h_{t-1} + bx_f + bh_f) $$
f_t &= \sigma(W_{fx}x_{t} + W_{fh}h_{t-1} + bx_f + bh_f)
$$ o_t =
\\
sigma(W_{ox}x_{t} + W_{oh}h_{t-1} + bx_o + bh_o) $$
o_t &= \sigma(W_{ox}x_{t} + W_{oh}h_{t-1} + bx_o + bh_o)
$$
\\
tilde{c_t} = tanh(W_{cx}x_t + W_{ch}h_{t-1} + bx_c + bh_c) $$
\\
tilde{c_t} &= tanh(W_{cx}x_t + W_{ch}h_{t-1} + bx_c + bh_c)
$$ c_t = f_t
\\
odot c_{t-1} + i_t
\\
odot
\\
tilde{c_t} $$
c_t &= f_t \odot c_{t-1} + i_t \odot
\\
tilde{c_t}
$$ h_t = o_t
\\
odot tanh(c_t) $$
h_t &= o_t \odot tanh(c_t)
-
W
terms denote weight matrices (e.g. $W_{ix}$ is the matrix
-
$W$
terms denote weight matrices (e.g. $W_{ix}$ is the matrix
of weights from the input gate to the input)
of weights from the input gate to the input)
- The b terms denote bias vectors ($bx_i$ and $bh_i$ are the input gate bias vector).
- The b terms denote bias vectors ($bx_i$ and $bh_i$ are the input gate bias vector).
- sigmoid is the logistic sigmoid function.
- sigmoid is the logistic sigmoid function.
- $i, f, o$ and $c$ are the input gate, forget gate, output gate,
- $i, f, o$ and $c$ are the input gate, forget gate, output gate,
and cell activation vectors, respectively, all of which have the same size as
and cell activation vectors, respectively, all of which have the same size as
the cell output activation vector $h$.
the cell output activation vector $h$.
- The
$\odot$
is the element-wise product of the vectors.
- The
:math:`\odot`
is the element-wise product of the vectors.
- `tanh` is the activation functions.
-
:math:
`tanh` is the activation functions.
-
$
\t
ilde{c_t}$
is also called candidate hidden state,
-
:math:`
\\
tilde{c_t}`
is also called candidate hidden state,
which is computed based on the current input and the previous hidden state.
which is computed based on the current input and the previous hidden state.
Where sigmoid is the sigmoid operator:
sigmoid(x) = 1 / (1 + e^-x), * represents a point-wise multiplication,
Where sigmoid is the sigmoid operator:
:math:`sigmoid(x) = 1 / (1 + e^{-x})` , * represents a point-wise multiplication,
X represensts a matrix multiplication
X represensts a matrix multiplication
...
@@ -556,13 +558,17 @@ def lstm(input,
...
@@ -556,13 +558,17 @@ def lstm(input,
Returns:
Returns:
rnn_out(Tensor): result of LSTM hidden, shape is (seq_len x batch_size x hidden_size)
rnn_out(Tensor),last_h(Tensor),last_c(Tensor):
Three tensors, rnn_out, last_h, last_c:
- rnn_out is result of LSTM hidden, shape is (seq_len x batch_size x hidden_size)
\
if is_bidirec set to True, shape will be ( seq_len x batch_sze x hidden_size*2)
if is_bidirec set to True, shape will be ( seq_len x batch_sze x hidden_size*2)
last_h(Tensor): the hidden state of the last step of LSTM
- last_h is the hidden state of the last step of LSTM
\
shape is ( num_layers x batch_size x hidden_size )
shape is ( num_layers x batch_size x hidden_size )
\
if is_bidirec set to True, shape will be ( num_layers*2 x batch_size x hidden_size)
if is_bidirec set to True, shape will be ( num_layers*2 x batch_size x hidden_size)
last_c(Tensor): the cell state of the last step of LSTM
- last_c(Tensor): the cell state of the last step of LSTM
\
shape is ( num_layers x batch_size x hidden_size )
shape is ( num_layers x batch_size x hidden_size )
\
if is_bidirec set to True, shape will be ( num_layers*2 x batch_size x hidden_size)
if is_bidirec set to True, shape will be ( num_layers*2 x batch_size x hidden_size)
...
@@ -1220,6 +1226,8 @@ def dropout(x,
...
@@ -1220,6 +1226,8 @@ def dropout(x,
probability) the outputs of some units to zero, while others are remain
probability) the outputs of some units to zero, while others are remain
unchanged.
unchanged.
dropout op can be removed from the program to make the program more efficient.
Args:
Args:
x (Variable): The input tensor variable.
x (Variable): The input tensor variable.
dropout_prob (float): Probability of setting units to zero.
dropout_prob (float): Probability of setting units to zero.
...
@@ -1230,20 +1238,22 @@ def dropout(x,
...
@@ -1230,20 +1238,22 @@ def dropout(x,
units will be dropped. DO NOT use a fixed seed in training.
units will be dropped. DO NOT use a fixed seed in training.
name (str|None): A name for this layer(optional). If set None, the layer
name (str|None): A name for this layer(optional). If set None, the layer
will be named automatically.
will be named automatically.
dropout_implementation(string): ['downgrade_in_infer'(defauld)|'upscale_in_train']
dropout_implementation(string): ['downgrade_in_infer'(default)|'upscale_in_train']
1. downgrade_in_infer(default), downgrade the outcome at inference
1. downgrade_in_infer(default), downgrade the outcome at inference
train: out = input * mask
inference: out = input * dropout_prob
- train: out = input * mask
(make is a tensor same shape with input, value is 0 or 1
- inference: out = input * dropout_prob
(mask is a tensor same shape with input, value is 0 or 1
ratio of 0 is dropout_prob)
ratio of 0 is dropout_prob)
2. upscale_in_train, upscale the outcome at training time
2. upscale_in_train, upscale the outcome at training time
train: out = input * mask / ( 1.0 - dropout_prob )
inference: out = input
(make is a tensor same shape with input, value is 0 or 1
ratio of 0 is dropout_prob)
dropout op can be removed from the program.
the program will be efficient
- train: out = input * mask / ( 1.0 - dropout_prob )
- inference: out = input
(mask is a tensor same shape with input, value is 0 or 1
ratio of 0 is dropout_prob)
Returns:
Returns:
...
@@ -1333,11 +1343,15 @@ def cross_entropy(input, label, soft_label=False, ignore_index=kIgnoreIndex):
...
@@ -1333,11 +1343,15 @@ def cross_entropy(input, label, soft_label=False, ignore_index=kIgnoreIndex):
A 2-D tensor with shape [N x 1], the cross entropy loss.
A 2-D tensor with shape [N x 1], the cross entropy loss.
Raises:
Raises:
`ValueError`: 1) the 1st dimension of `input` and `label` are not equal.
ValueError:
2) when `soft_label == True`, and the 2nd dimension of
`input` and `label` are not equal.
1. the 1st dimension of ``input`` and ``label`` are not equal.
3) when `soft_label == False`, and the 2nd dimension of
`label` is not 1.
2. when ``soft_label == True``, and the 2nd dimension of
``input`` and ``label`` are not equal.
3. when ``soft_label == False``, and the 2nd dimension of
``label`` is not 1.
Examples:
Examples:
.. code-block:: python
.. code-block:: python
...
@@ -1458,7 +1472,7 @@ def chunk_eval(input,
...
@@ -1458,7 +1472,7 @@ def chunk_eval(input,
F1-score of chunk detection.
F1-score of chunk detection.
For some basics of chunking, please refer to
For some basics of chunking, please refer to
'Chunking with Support Vector Machines <https://aclanthology.info/pdf/N/N01/N01-1025.pdf>'
.
`Chunking with Support Vector Machines <https://aclanthology.info/pdf/N/N01/N01-1025.pdf>`_
.
ChunkEvalOp computes the precision, recall, and F1-score of chunk detection,
ChunkEvalOp computes the precision, recall, and F1-score of chunk detection,
and supports IOB, IOE, IOBES and IO (also known as plain) tagging schemes.
and supports IOB, IOE, IOBES and IO (also known as plain) tagging schemes.
...
@@ -2292,7 +2306,8 @@ def sequence_slice(input, offset, length, name=None):
...
@@ -2292,7 +2306,8 @@ def sequence_slice(input, offset, length, name=None):
out.lod = [[2, 1]],
out.lod = [[2, 1]],
out.dims = (3, 2).
out.dims = (3, 2).
NOTE: The first dimension size of **input**, **offset** and **length**
Note:
The first dimension size of **input**, **offset** and **length**
should be equal. The **offset** should start from 0.
should be equal. The **offset** should start from 0.
Args:
Args:
...
@@ -3013,7 +3028,7 @@ def group_norm(input,
...
@@ -3013,7 +3028,7 @@ def group_norm(input,
"""
"""
**Group Normalization Layer**
**Group Normalization Layer**
Refer to `Group Normalization <https://arxiv.org/abs/1803.08494>`
Refer to `Group Normalization <https://arxiv.org/abs/1803.08494>`
_ .
Args:
Args:
input(Variable): The input tensor variable.
input(Variable): The input tensor variable.
...
@@ -3140,8 +3155,8 @@ def conv2d_transpose(input,
...
@@ -3140,8 +3155,8 @@ def conv2d_transpose(input,
H^\prime_{out} &= (H_{in} - 1) * strides[0] - 2 * paddings[0] + dilations[0] * (H_f - 1) + 1
\\\\
H^\prime_{out} &= (H_{in} - 1) * strides[0] - 2 * paddings[0] + dilations[0] * (H_f - 1) + 1
\\\\
W^\prime_{out} &= (W_{in} - 1) * strides[1] - 2 * paddings[1] + dilations[1] * (W_f - 1) + 1
\\\\
W^\prime_{out} &= (W_{in} - 1) * strides[1] - 2 * paddings[1] + dilations[1] * (W_f - 1) + 1
\\\\
H_{out} \in [ H^\prime_{out}, H^\prime_{out} + strides[0] )
\\\\
H_{out}
&
\in [ H^\prime_{out}, H^\prime_{out} + strides[0] )
\\\\
W_{out} \in [ W^\prime_{out}, W^\prime_{out} + strides[1] )
W_{out}
&
\in [ W^\prime_{out}, W^\prime_{out} + strides[1] )
Args:
Args:
input(Variable): The input image with [N, C, H, W] format.
input(Variable): The input image with [N, C, H, W] format.
...
@@ -4704,9 +4719,9 @@ def ctc_greedy_decoder(input, blank, name=None):
...
@@ -4704,9 +4719,9 @@ def ctc_greedy_decoder(input, blank, name=None):
name (str): The name of this layer. It is optional.
name (str): The name of this layer. It is optional.
Returns:
Returns:
Variable: CTC greedy decode result which is a 2-D tensor with shape [Lp, 1].
Variable: CTC greedy decode result which is a 2-D tensor with shape [Lp, 1].
\
'Lp' is the sum if all output sequences' length. If all the sequences
'Lp' is the sum if all output sequences' length. If all the sequences
\
in result were empty, the result LoDTensor will be [-1] with
in result were empty, the result LoDTensor will be [-1] with
\
LoD [[]] and dims [1, 1].
LoD [[]] and dims [1, 1].
Examples:
Examples:
...
@@ -5072,6 +5087,7 @@ def hsigmoid(input,
...
@@ -5072,6 +5087,7 @@ def hsigmoid(input,
<http://www.iro.umontreal.ca/~lisa/pointeurs/hierarchical-nnlm-aistats05.pdf>`_
<http://www.iro.umontreal.ca/~lisa/pointeurs/hierarchical-nnlm-aistats05.pdf>`_
And if you want to use the costumed tree by set 'is_custom' as true you may need to do following things first:
And if you want to use the costumed tree by set 'is_custom' as true you may need to do following things first:
1. using your word dict to build a binary tree, each leaf node should be an word of your word dict
1. using your word dict to build a binary tree, each leaf node should be an word of your word dict
2. build a dict to store word_id -> word's leaf to root path, we call it path_table.
2. build a dict to store word_id -> word's leaf to root path, we call it path_table.
3. build a dict to store word_id -> code of word's leaf to root path, we call it path_code. Code
3. build a dict to store word_id -> code of word's leaf to root path, we call it path_code. Code
...
@@ -5079,7 +5095,6 @@ def hsigmoid(input,
...
@@ -5079,7 +5095,6 @@ def hsigmoid(input,
4. now, each word should has its path and code along the path, you can pass a batch of path and code
4. now, each word should has its path and code along the path, you can pass a batch of path and code
related to the same batch of inputs.
related to the same batch of inputs.
Args:
Args:
input (Variable): The input tensor variable with shape
input (Variable): The input tensor variable with shape
:math:`[N
\\
times D]`, where :math:`N` is the size of mini-batch,
:math:`[N
\\
times D]`, where :math:`N` is the size of mini-batch,
...
@@ -5485,11 +5500,11 @@ def softmax_with_cross_entropy(logits,
...
@@ -5485,11 +5500,11 @@ def softmax_with_cross_entropy(logits,
.. math::
.. math::
max_j =
\\
max_{i=0}^{K}{
\\
text{logit}_i}
max_j
&
=
\\
max_{i=0}^{K}{
\\
text{logit}_i}
log
\\
_max
\\
_sum_j =
\\
log
\\
sum_{i=0}^{K}
\\
exp(logit_i - max_j)
log
\\
_max
\\
_sum_j
&
=
\\
log
\\
sum_{i=0}^{K}
\\
exp(logit_i - max_j)
softmax_j =
\\
exp(logit_j - max_j - {log
\\
_max
\\
_sum}_j)
softmax_j
&
=
\\
exp(logit_j - max_j - {log
\\
_max
\\
_sum}_j)
and then cross entropy loss is calculated by softmax and label.
and then cross entropy loss is calculated by softmax and label.
...
@@ -5515,10 +5530,10 @@ def softmax_with_cross_entropy(logits,
...
@@ -5515,10 +5530,10 @@ def softmax_with_cross_entropy(logits,
along with the cross entropy loss. Default: False
along with the cross entropy loss. Default: False
Returns:
Returns:
Variable or Tuple of two Variables: Return the cross entropy loss if
Variable or Tuple of two Variables: Return the cross entropy loss if
\
`return_softmax` is False, otherwise the tuple
`return_softmax` is False, otherwise the tuple
\
(loss, softmax), where the cross entropy loss is
(loss, softmax), where the cross entropy loss is
\
a 2-D tensor with shape [N x 1], and softmax is a
a 2-D tensor with shape [N x 1], and softmax is a
\
2-D tensor with shape [N x K].
2-D tensor with shape [N x K].
Examples:
Examples:
...
@@ -5792,15 +5807,21 @@ def squeeze(input, axes, name=None):
...
@@ -5792,15 +5807,21 @@ def squeeze(input, axes, name=None):
the single dimensions will be removed from the shape. If an axis is
the single dimensions will be removed from the shape. If an axis is
selected with shape entry not equal to one, an error is raised.
selected with shape entry not equal to one, an error is raised.
Examples:
For example:
.. code-block:: text
Case 1:
Case 1:
Given
Given
X.shape = (1, 3, 1, 5)
X.shape = (1, 3, 1, 5)
and
and
axes = [0]
axes = [0]
we get:
we get:
Out.shape = (3, 1, 5)
Out.shape = (3, 1, 5)
Case 2:
Case 2:
Given
Given
X.shape = (1, 3, 1, 5)
X.shape = (1, 3, 1, 5)
and
and
...
@@ -5842,6 +5863,9 @@ def unsqueeze(input, axes, name=None):
...
@@ -5842,6 +5863,9 @@ def unsqueeze(input, axes, name=None):
Dimension indices in axes are as seen in the output tensor.
Dimension indices in axes are as seen in the output tensor.
For example:
For example:
.. code-block:: text
Given a tensor such that tensor with shape [3, 4, 5],
Given a tensor such that tensor with shape [3, 4, 5],
then Unsqueezed tensor with axes=[0, 4] has shape [1, 3, 4, 5, 1].
then Unsqueezed tensor with axes=[0, 4] has shape [1, 3, 4, 5, 1].
...
@@ -6729,8 +6753,11 @@ def sequence_scatter(input, index, updates, name=None):
...
@@ -6729,8 +6753,11 @@ def sequence_scatter(input, index, updates, name=None):
the columns to update in each row of X.
the columns to update in each row of X.
Here is an example:
Here is an example:
Given the following input:
Given the following input:
.. code-block:: text
.. code-block:: text
input.data = [[1.0, 1.0, 1.0, 1.0, 1.0, 1.0],
input.data = [[1.0, 1.0, 1.0, 1.0, 1.0, 1.0],
[1.0, 1.0, 1.0, 1.0, 1.0, 1.0],
[1.0, 1.0, 1.0, 1.0, 1.0, 1.0],
[1.0, 1.0, 1.0, 1.0, 1.0, 1.0]]
[1.0, 1.0, 1.0, 1.0, 1.0, 1.0]]
...
@@ -6743,7 +6770,9 @@ def sequence_scatter(input, index, updates, name=None):
...
@@ -6743,7 +6770,9 @@ def sequence_scatter(input, index, updates, name=None):
updates.lod = [[ 0, 3, 8, 12]]
updates.lod = [[ 0, 3, 8, 12]]
Then we have the output:
Then we have the output:
.. code-block:: text
.. code-block:: text
out.data = [[1.3, 1.3, 1.4, 1.0, 1.0, 1.0],
out.data = [[1.3, 1.3, 1.4, 1.0, 1.0, 1.0],
[1.0, 1.0, 1.4, 1.3, 1.2, 1.1],
[1.0, 1.0, 1.4, 1.3, 1.2, 1.1],
[1.0, 1.0, 1.3, 1.2, 1.4, 1.1]]
[1.0, 1.0, 1.3, 1.2, 1.4, 1.1]]
...
@@ -6759,7 +6788,7 @@ def sequence_scatter(input, index, updates, name=None):
...
@@ -6759,7 +6788,7 @@ def sequence_scatter(input, index, updates, name=None):
name (str|None): The output variable name. Default None.
name (str|None): The output variable name. Default None.
Returns:
Returns:
output (Variable)
: The output is a tensor with the same shape as input.
Variable
: The output is a tensor with the same shape as input.
Examples:
Examples:
...
@@ -6933,7 +6962,7 @@ def mean_iou(input, label, num_classes):
...
@@ -6933,7 +6962,7 @@ def mean_iou(input, label, num_classes):
.. math::
.. math::
IOU =
\\
frac{true\_positiv}{(true\_positive + false\_positive + false\_negative)}.
IOU =
\\
frac{true\_positiv
e
}{(true\_positive + false\_positive + false\_negative)}.
The predictions are accumulated in a confusion matrix and mean-IOU
The predictions are accumulated in a confusion matrix and mean-IOU
is then calculated from it.
is then calculated from it.
...
@@ -6946,9 +6975,13 @@ def mean_iou(input, label, num_classes):
...
@@ -6946,9 +6975,13 @@ def mean_iou(input, label, num_classes):
num_classes (int): The possible number of labels.
num_classes (int): The possible number of labels.
Returns:
Returns:
mean_iou (Variable): A Tensor representing the mean intersection-over-union with shape [1].
mean_iou (Variable),out_wrong(Variable),out_correct(Variable):
out_wrong(Variable): A Tensor with shape [num_classes]. The wrong numbers of each class.
out_correct(Variable): A Tensor with shape [num_classes]. The correct numbers of each class.
Three variables:
- mean_iou : A Tensor representing the mean intersection-over-union with shape [1].
- out_wrong: A Tensor with shape [num_classes]. The wrong numbers of each class.
- out_correct: A Tensor with shape [num_classes]. The correct numbers of each class.
Examples:
Examples:
...
@@ -7144,7 +7177,7 @@ def affine_grid(theta, out_shape, name=None):
...
@@ -7144,7 +7177,7 @@ def affine_grid(theta, out_shape, name=None):
Args:
Args:
theta (Variable): A batch of affine transform parameters with shape [N, 2, 3].
theta (Variable): A batch of affine transform parameters with shape [N, 2, 3].
out_shape (Variable | list | tuple): The shape of target output with format [N, C, H, W].
out_shape (Variable | list | tuple): The shape of target output with format [N, C, H, W].
out_shape
can be a Variable or a list or tuple.
``out_shape``
can be a Variable or a list or tuple.
name(str|None): A name for this layer(optional). If set None, the layer
name(str|None): A name for this layer(optional). If set None, the layer
will be named automatically.
will be named automatically.
...
@@ -7157,6 +7190,7 @@ def affine_grid(theta, out_shape, name=None):
...
@@ -7157,6 +7190,7 @@ def affine_grid(theta, out_shape, name=None):
Examples:
Examples:
.. code-block:: python
.. code-block:: python
theta = fluid.layers.data(name="x", shape=[2, 3], dtype="float32")
theta = fluid.layers.data(name="x", shape=[2, 3], dtype="float32")
out_shape = fluid.layers.data(name="y", shape=[-1], dtype="float32")
out_shape = fluid.layers.data(name="y", shape=[-1], dtype="float32")
data = fluid.layers.affine_grid(theta, out_shape)
data = fluid.layers.affine_grid(theta, out_shape)
...
@@ -7192,9 +7226,10 @@ def affine_grid(theta, out_shape, name=None):
...
@@ -7192,9 +7226,10 @@ def affine_grid(theta, out_shape, name=None):
def
rank_loss
(
label
,
left
,
right
,
name
=
None
):
def
rank_loss
(
label
,
left
,
right
,
name
=
None
):
"""
"""
**Rank loss layer for RankNet**
**Rank loss layer for RankNet**
RankNet(http://icml.cc/2015/wp-content/uploads/2015/06/icml_ranking.pdf)
`RankNet <http://icml.cc/2015/wp-content/uploads/2015/06/icml_ranking.pdf>`_
is a pairwise ranking model with a training sample consisting of a pair
is a pairwise ranking model with a training sample consisting of a pair
of documents, A and B. Label P indicates whether A is ranked higher than B
of documents, A and B. Label P indicates whether A is ranked higher than B
or not:
or not:
...
@@ -7202,16 +7237,19 @@ def rank_loss(label, left, right, name=None):
...
@@ -7202,16 +7237,19 @@ def rank_loss(label, left, right, name=None):
P = {0, 1} or {0, 0.5, 1}, where 0.5 means that there is no information
P = {0, 1} or {0, 0.5, 1}, where 0.5 means that there is no information
about the rank of the input pair.
about the rank of the input pair.
Rank loss layer takes three inputs: left (
o_i), right (o_j
) and
Rank loss layer takes three inputs: left (
:math:`o_i` ), right ( :math:`o_j`
) and
label (
P_{i,j}
). The inputs respectively represent RankNet's output scores
label (
:math:`P_{i,j}`
). The inputs respectively represent RankNet's output scores
for documents A and B and the value of label P. The following equation
for documents A and B and the value of label P. The following equation
computes rank loss C_{i,j} from the inputs:
computes rank loss C_{i,j} from the inputs:
$$
.. math::
C_{i,j} = -
\t
ilde{P_{ij}} * o_{i,j} + \log(1 + e^{o_{i,j}})
\\
o_{i,j} = o_i - o_j
\\
C_{i,j} &= -
\\
tilde{P_{ij}} * o_{i,j} + \log(1 + e^{o_{i,j}})
\\\\
\t
ilde{P_{i,j}} = \left \{0, 0.5, 1
\r
ight \} \ or \ \left \{0, 1
\r
ight \}
$$
o_{i,j} &= o_i - o_j
\\\\
\\
tilde{P_{i,j}} &=
\\
left \{0, 0.5, 1
\\
right \} \ or \
\\
left \{0, 1
\\
right \}
Rank loss layer takes batch inputs with size batch_size (batch_size >= 1).
Rank loss layer takes batch inputs with size batch_size (batch_size >= 1).
...
@@ -7237,7 +7275,6 @@ def rank_loss(label, left, right, name=None):
...
@@ -7237,7 +7275,6 @@ def rank_loss(label, left, right, name=None):
right = fluid.layers.data(name="right", shape=[4, 1], dtype="float32")
right = fluid.layers.data(name="right", shape=[4, 1], dtype="float32")
out = fluid.layers.rank_loss(label, left, right)
out = fluid.layers.rank_loss(label, left, right)
"""
"""
helper
=
LayerHelper
(
'rank_loss'
,
**
locals
())
helper
=
LayerHelper
(
'rank_loss'
,
**
locals
())
...
@@ -7269,7 +7306,7 @@ def margin_rank_loss(label, left, right, margin=0.1, name=None):
...
@@ -7269,7 +7306,7 @@ def margin_rank_loss(label, left, right, margin=0.1, name=None):
.. math::
.. math::
rank\_loss
&
= max(0, -label * (left - right) + margin)
rank\_loss = max(0, -label * (left - right) + margin)
Args:
Args:
label (Variable): Indicates whether the left is ranked higher than the right or not.
label (Variable): Indicates whether the left is ranked higher than the right or not.
...
@@ -7278,12 +7315,17 @@ def margin_rank_loss(label, left, right, margin=0.1, name=None):
...
@@ -7278,12 +7315,17 @@ def margin_rank_loss(label, left, right, margin=0.1, name=None):
margin (float): Indicates the given margin.
margin (float): Indicates the given margin.
name (str|None): A name for this layer (optional). If set None, the layer
name (str|None): A name for this layer (optional). If set None, the layer
will be named automatically.
will be named automatically.
Returns:
Returns:
Variable: The ranking loss.
Variable: The ranking loss.
Raises:
Raises:
ValueError: Any of label, left, and right is not a Variable.
ValueError: Any of label, left, and right is not a Variable.
Examples:
Examples:
.. code-block:: python
.. code-block:: python
label = fluid.layers.data(name="label", shape=[4, 1], dtype="float32")
label = fluid.layers.data(name="label", shape=[4, 1], dtype="float32")
left = fluid.layers.data(name="left", shape=[4, 1], dtype="float32")
left = fluid.layers.data(name="left", shape=[4, 1], dtype="float32")
right = fluid.layers.data(name="right", shape=[4, 1], dtype="float32")
right = fluid.layers.data(name="right", shape=[4, 1], dtype="float32")
...
@@ -7587,7 +7629,8 @@ def prelu(x, mode, param_attr=None, name=None):
...
@@ -7587,7 +7629,8 @@ def prelu(x, mode, param_attr=None, name=None):
"""
"""
Equation:
Equation:
y = \max(0, x) + alpha * \min(0, x)
.. math::
y = \max(0, x) +
\\
alpha * \min(0, x)
Args:
Args:
x (Variable): The input tensor.
x (Variable): The input tensor.
...
@@ -7730,20 +7773,29 @@ def flatten(x, axis=1, name=None):
...
@@ -7730,20 +7773,29 @@ def flatten(x, axis=1, name=None):
**Flatten layer**
**Flatten layer**
Flattens the input tensor into a 2D matrix.
Flattens the input tensor into a 2D matrix.
Examples:
For Example:
.. code-block:: text
Case 1:
Case 1:
Given
Given
X.shape = (3, 100, 100, 4)
X.shape = (3, 100, 100, 4)
and
and
axis = 2
axis = 2
We get:
We get:
Out.shape = (3 * 100, 4 * 100)
Out.shape = (3 * 100, 4 * 100)
Case 2:
Case 2:
Given
Given
X.shape = (3, 100, 100, 4)
X.shape = (3, 100, 100, 4)
and
and
axis = 0
axis = 0
We get:
We get:
Out.shape = (1, 3 * 100 * 100 * 4)
Out.shape = (1, 3 * 100 * 100 * 4)
...
@@ -7759,9 +7811,9 @@ def flatten(x, axis=1, name=None):
...
@@ -7759,9 +7811,9 @@ def flatten(x, axis=1, name=None):
will be named automatically.
will be named automatically.
Returns:
Returns:
Variable: A 2D tensor with the contents of the input tensor, with input
Variable: A 2D tensor with the contents of the input tensor, with input
\
dimensions up to axis flattened to the outer dimension of
dimensions up to axis flattened to the outer dimension of
\
the output and remaining input dimensions flattened into the
the output and remaining input dimensions flattened into the
\
inner dimension of the output.
inner dimension of the output.
Raises:
Raises:
...
@@ -7801,15 +7853,19 @@ def sequence_enumerate(input, win_size, pad_value=0, name=None):
...
@@ -7801,15 +7853,19 @@ def sequence_enumerate(input, win_size, pad_value=0, name=None):
The enumerated sequence has the same 1st dimension with variable `input`, and
The enumerated sequence has the same 1st dimension with variable `input`, and
the 2nd dimension is `win_size`, padded by `pad_value` if necessary in generation.
the 2nd dimension is `win_size`, padded by `pad_value` if necessary in generation.
Examples:
.. code-block:: text
Case 1:
Case 1:
Input:
Input:
X.lod = [[0, 3, 5]]
X.lod = [[0, 3, 5]]
X.data = [[1], [2], [3], [4], [5]]
X.data = [[1], [2], [3], [4], [5]]
X.dims = [5, 1]
X.dims = [5, 1]
Attrs:
Attrs:
win_size = 2
win_size = 2
pad_value = 0
pad_value = 0
Output:
Output:
Out.lod = [[0, 3, 5]]
Out.lod = [[0, 3, 5]]
Out.data = [[1, 2], [2, 3], [3, 0], [4, 5], [5, 0]]
Out.data = [[1, 2], [2, 3], [3, 0], [4, 5], [5, 0]]
...
@@ -8896,6 +8952,7 @@ def similarity_focus(input, axis, indexes, name=None):
...
@@ -8896,6 +8952,7 @@ def similarity_focus(input, axis, indexes, name=None):
SimilarityFocus Operator
SimilarityFocus Operator
Generate a similarity focus mask with the same shape of input using the following method:
Generate a similarity focus mask with the same shape of input using the following method:
1. Extract the 3-D tensor(here the first dimension is BatchSize) corresponding
1. Extract the 3-D tensor(here the first dimension is BatchSize) corresponding
to the axis according to the indexes. For example, if axis=1 and indexes=[a],
to the axis according to the indexes. For example, if axis=1 and indexes=[a],
it will get the matrix T=X[:, a, :, :]. In this case, if the shape of input X
it will get the matrix T=X[:, a, :, :]. In this case, if the shape of input X
...
@@ -8969,14 +9026,16 @@ def similarity_focus(input, axis, indexes, name=None):
...
@@ -8969,14 +9026,16 @@ def similarity_focus(input, axis, indexes, name=None):
indexes(list): Indicating the indexes of the selected dimension.
indexes(list): Indicating the indexes of the selected dimension.
Returns:
Returns:
Variable: A tensor variable with the same shape and same type
Variable: A tensor variable with the same shape and same type
\
as the input.
as the input.
Examples:
Examples:
.. code-block:: python
.. code-block:: python
data = fluid.layers.data(
data = fluid.layers.data(
name='data', shape=[2, 3, 2, 2], dtype='float32')
name='data', shape=[2, 3, 2, 2], dtype='float32')
x = fluid.layers.layer_norm(input=data, axis=1, indexes=[0])
x = fluid.layers.layer_norm(input=data, axis=1, indexes=[0])
"""
"""
helper
=
LayerHelper
(
'similarity_focus'
,
**
locals
())
helper
=
LayerHelper
(
'similarity_focus'
,
**
locals
())
# check attrs
# check attrs
...
@@ -9055,6 +9114,7 @@ def hash(input, hash_size, num_hash=1, name=None):
...
@@ -9055,6 +9114,7 @@ def hash(input, hash_size, num_hash=1, name=None):
Examples:
Examples:
.. code-block:: python
.. code-block:: python
word_dict = paddle.dataset.imdb.word_dict()
word_dict = paddle.dataset.imdb.word_dict()
x = fluid.layers.data(shape[1], dtype='int32', lod_level=1)
x = fluid.layers.data(shape[1], dtype='int32', lod_level=1)
out = fluid.layers.hash(input=x, num_hash=4, hash_size=1000)
out = fluid.layers.hash(input=x, num_hash=4, hash_size=1000)
...
@@ -9075,13 +9135,15 @@ def hash(input, hash_size, num_hash=1, name=None):
...
@@ -9075,13 +9135,15 @@ def hash(input, hash_size, num_hash=1, name=None):
def
grid_sampler
(
x
,
grid
,
name
=
None
):
def
grid_sampler
(
x
,
grid
,
name
=
None
):
"""
"""
This operation samples input X by using bilinear interpolation based on
This operation samples input X by using bilinear interpolation based on
flow field grid, which is usually gennerated by
affine_grid
. The grid of
flow field grid, which is usually gennerated by
:code:`affine_grid`
. The grid of
shape [N, H, W, 2] is the concatenation of (grid_x, grid_y) coordinates
shape [N, H, W, 2] is the concatenation of (grid_x, grid_y) coordinates
with shape [N, H, W] each, where grid_x is indexing the 4th dimension
with shape [N, H, W] each, where grid_x is indexing the 4th dimension
(in width dimension) of input data x and grid_y is indexng the 3rd
(in width dimension) of input data x and grid_y is indexng the 3rd
dimention (in height dimension), finally results is the bilinear
dimention (in height dimension), finally results is the bilinear
interpolation value of 4 nearest corner points.
interpolation value of 4 nearest corner points.
.. code-block:: text
Step 1:
Step 1:
Get (x, y) grid coordinates and scale to [0, H-1/W-1].
Get (x, y) grid coordinates and scale to [0, H-1/W-1].
...
@@ -9126,16 +9188,18 @@ def grid_sampler(x, grid, name=None):
...
@@ -9126,16 +9188,18 @@ def grid_sampler(x, grid, name=None):
name (str, default None): The name of this layer.
name (str, default None): The name of this layer.
Returns:
Returns:
out(Variable)
: Output of shape [N, C, H, W] data samples input X
Variable
: Output of shape [N, C, H, W] data samples input X
using bilnear interpolation based on input grid.
using bilnear interpolation based on input grid.
Exmples:
Examples:
.. code-block:: python
.. code-block:: python
x = fluid.layers.data(name='x', shape=[3, 10, 32, 32], dtype='float32')
x = fluid.layers.data(name='x', shape=[3, 10, 32, 32], dtype='float32')
theta = fluid.layers.data(name='theta', shape=[3, 2, 3], dtype='float32')
theta = fluid.layers.data(name='theta', shape=[3, 2, 3], dtype='float32')
grid = fluid.layers.affine_grid(input=theta, size=[3, 10, 32, 32]})
grid = fluid.layers.affine_grid(input=theta, size=[3, 10, 32, 32]})
out = fluid.layers.grid_sampler(x=x, grid=grid)
out = fluid.layers.grid_sampler(x=x, grid=grid)
"""
"""
helper
=
LayerHelper
(
"grid_sampler"
,
**
locals
())
helper
=
LayerHelper
(
"grid_sampler"
,
**
locals
())
...
@@ -9203,19 +9267,19 @@ def add_position_encoding(input, alpha, beta, name=None):
...
@@ -9203,19 +9267,19 @@ def add_position_encoding(input, alpha, beta, name=None):
"""
"""
**Add Position Encoding Layer**
**Add Position Encoding Layer**
This layer accepts an input 3D-Tensor of shape [N x M x P], and return an
This layer accepts an input 3D-Tensor of shape [N x M x P], and return
s
an
output Tensor of shape [N x M x P] with positional encoding value.
output Tensor of shape [N x M x P] with positional encoding value.
Refer to `Attention Is All You Need<http://arxiv.org/pdf/1706.03762.pdf>`_ .
Refer to `Attention Is All You Need
<http://arxiv.org/pdf/1706.03762.pdf>`_ .
.. math::
.. math::
PE(pos, 2i) =
\\
sin{(pos / 10000^{2i / P})}
\\\\
PE(pos, 2i)
&
=
\\
sin{(pos / 10000^{2i / P})}
\\\\
PE(pos, 2i + 1) =
\\
cos{(pos / 10000^{2i / P})}
\\\\
PE(pos, 2i + 1)
&
=
\\
cos{(pos / 10000^{2i / P})}
\\\\
Out(:, pos, i) =
\\
alpha * input(:, pos, i) +
\\
beta * PE(pos, i)
Out(:, pos, i)
&
=
\\
alpha * input(:, pos, i) +
\\
beta * PE(pos, i)
Where:
Where:
* PE(pos, 2i)
: the increment for the number at even position
- :math:`PE(pos, 2i)`
: the increment for the number at even position
* PE(pos, 2i + 1)
: the increment for the number at odd position
- :math:`PE(pos, 2i + 1)`
: the increment for the number at odd position
Args:
Args:
input (Variable): 3-D input tensor with shape [N x M x P]
input (Variable): 3-D input tensor with shape [N x M x P]
...
@@ -9230,6 +9294,7 @@ def add_position_encoding(input, alpha, beta, name=None):
...
@@ -9230,6 +9294,7 @@ def add_position_encoding(input, alpha, beta, name=None):
.. code-block:: python
.. code-block:: python
position_tensor = fluid.layers.add_position_encoding(input=tensor)
position_tensor = fluid.layers.add_position_encoding(input=tensor)
"""
"""
helper
=
LayerHelper
(
'add_position_encoding'
,
**
locals
())
helper
=
LayerHelper
(
'add_position_encoding'
,
**
locals
())
dtype
=
helper
.
input_dtype
()
dtype
=
helper
.
input_dtype
()
...
@@ -9262,13 +9327,13 @@ def bilinear_tensor_product(x,
...
@@ -9262,13 +9327,13 @@ def bilinear_tensor_product(x,
For example:
For example:
.. math::
.. math::
out{i} = x * W_{i} * {y^\mathrm{T}}, i=0,1,...,size-1
out
_
{i} = x * W_{i} * {y^\mathrm{T}}, i=0,1,...,size-1
In this formula:
In this formula:
- :math:`x`: the first input contains M elements, shape is [batch_size, M].
- :math:`x`: the first input contains M elements, shape is [batch_size, M].
- :math:`y`: the second input contains N elements, shape is [batch_size, N].
- :math:`y`: the second input contains N elements, shape is [batch_size, N].
- :math:`W_{i}`: the i-th learned weight, shape is [M, N]
- :math:`W_{i}`: the i-th learned weight, shape is [M, N]
- :math:`out{i}`: the i-th element of out, shape is [batch_size, size].
- :math:`out
_
{i}`: the i-th element of out, shape is [batch_size, size].
- :math:`y^\mathrm{T}`: the transpose of :math:`y_{2}`.
- :math:`y^\mathrm{T}`: the transpose of :math:`y_{2}`.
Args:
Args:
...
...
python/paddle/fluid/layers/tensor.py
浏览文件 @
66ea7184
...
@@ -393,9 +393,6 @@ def fill_constant_batch_size_like(input,
...
@@ -393,9 +393,6 @@ def fill_constant_batch_size_like(input,
It also sets *stop_gradient* to True.
It also sets *stop_gradient* to True.
>>> data = fluid.layers.fill_constant_batch_size_like(
>>> input=like, shape=[1], value=0, dtype='int64')
Args:
Args:
input(${input_type}): ${input_comment}.
input(${input_type}): ${input_comment}.
...
@@ -411,6 +408,14 @@ def fill_constant_batch_size_like(input,
...
@@ -411,6 +408,14 @@ def fill_constant_batch_size_like(input,
Returns:
Returns:
${out_comment}.
${out_comment}.
Examples:
.. code-block:: python
data = fluid.layers.fill_constant_batch_size_like(
input=like, shape=[1], value=0, dtype='int64')
"""
"""
helper
=
LayerHelper
(
"fill_constant_batch_size_like"
,
**
locals
())
helper
=
LayerHelper
(
"fill_constant_batch_size_like"
,
**
locals
())
out
=
helper
.
create_variable_for_type_inference
(
dtype
=
dtype
)
out
=
helper
.
create_variable_for_type_inference
(
dtype
=
dtype
)
...
...
python/paddle/fluid/metrics.py
浏览文件 @
66ea7184
...
@@ -362,7 +362,7 @@ class ChunkEvaluator(MetricBase):
...
@@ -362,7 +362,7 @@ class ChunkEvaluator(MetricBase):
compute the precision recall and F1-score using the accumulated counter
compute the precision recall and F1-score using the accumulated counter
numbers.
numbers.
For some basics of chunking, please refer to
For some basics of chunking, please refer to
'Chunking with Support Vector Machines <https://aclanthology.info/pdf/N/N01/N01-1025.pdf>'
.
`Chunking with Support Vector Machines <https://aclanthology.info/pdf/N/N01/N01-1025.pdf>`_
.
ChunkEvalEvaluator computes the precision, recall, and F1-score of chunk detection,
ChunkEvalEvaluator computes the precision, recall, and F1-score of chunk detection,
and supports IOB, IOE, IOBES and IO (also known as plain) tagging schemes.
and supports IOB, IOE, IOBES and IO (also known as plain) tagging schemes.
...
@@ -391,6 +391,7 @@ class ChunkEvaluator(MetricBase):
...
@@ -391,6 +391,7 @@ class ChunkEvaluator(MetricBase):
def
update
(
self
,
num_infer_chunks
,
num_label_chunks
,
num_correct_chunks
):
def
update
(
self
,
num_infer_chunks
,
num_label_chunks
,
num_correct_chunks
):
"""
"""
Update the states based on the layers.chunk_eval() ouputs.
Update the states based on the layers.chunk_eval() ouputs.
Args:
Args:
num_infer_chunks(int|numpy.array): The number of chunks in Inference on the given minibatch.
num_infer_chunks(int|numpy.array): The number of chunks in Inference on the given minibatch.
num_label_chunks(int|numpy.array): The number of chunks in Label on the given mini-batch.
num_label_chunks(int|numpy.array): The number of chunks in Label on the given mini-batch.
...
@@ -450,9 +451,9 @@ class EditDistance(MetricBase):
...
@@ -450,9 +451,9 @@ class EditDistance(MetricBase):
distance, instance_error = distance_evaluator.eval()
distance, instance_error = distance_evaluator.eval()
In the above example:
In the above example:
'distance' is the average of the edit distance in a pass.
'instance_error' is the instance error rate in a pass.
- 'distance' is the average of the edit distance in a pass.
- 'instance_error' is the instance error rate in a pass.
"""
"""
...
@@ -567,12 +568,15 @@ class DetectionMAP(object):
...
@@ -567,12 +568,15 @@ class DetectionMAP(object):
Calculate the detection mean average precision (mAP).
Calculate the detection mean average precision (mAP).
The general steps are as follows:
The general steps are as follows:
1. calculate the true positive and false positive according to the input
1. calculate the true positive and false positive according to the input
of detection and labels.
of detection and labels.
2. calculate mAP value, support two versions: '11 point' and 'integral'.
2. calculate mAP value, support two versions: '11 point' and 'integral'.
Please get more information from the following articles:
Please get more information from the following articles:
https://sanchom.wordpress.com/tag/average-precision/
https://sanchom.wordpress.com/tag/average-precision/
https://arxiv.org/abs/1512.02325
https://arxiv.org/abs/1512.02325
Args:
Args:
...
@@ -615,8 +619,10 @@ class DetectionMAP(object):
...
@@ -615,8 +619,10 @@ class DetectionMAP(object):
In the above example:
In the above example:
'cur_map_v' is the mAP of current mini-batch.
- 'cur_map_v' is the mAP of current mini-batch.
'accum_map_v' is the accumulative mAP of one pass.
- 'accum_map_v' is the accumulative mAP of one pass.
"""
"""
def
__init__
(
self
,
def
__init__
(
self
,
...
...
python/paddle/fluid/transpiler/distribute_transpiler.py
浏览文件 @
66ea7184
...
@@ -125,14 +125,23 @@ def slice_variable(var_list, slice_count, min_block_size):
...
@@ -125,14 +125,23 @@ def slice_variable(var_list, slice_count, min_block_size):
class
DistributeTranspilerConfig
(
object
):
class
DistributeTranspilerConfig
(
object
):
"""
"""
Args:
.. py:attribute:: slice_var_up (bool)
slice_var_up (bool): Do Tensor slice for pservers, default is True.
split_method (PSDispatcher): RoundRobin or HashName can be used
Do Tensor slice for pservers, default is True.
try to choose the best method to balance loads for pservers.
min_block_size (int): Minimum splitted element number in block.
.. py:attribute:: split_method (PSDispatcher)
According:https://github.com/PaddlePaddle/Paddle/issues/8638#issuecomment-369912156
RoundRobin or HashName can be used.
Try to choose the best method to balance loads for pservers.
.. py:attribute:: min_block_size (int)
Minimum number of splitted elements in block.
According to : https://github.com/PaddlePaddle/Paddle/issues/8638#issuecomment-369912156
We can use bandwidth effiently when data size is larger than 2MB.If you
We can use bandwidth effiently when data size is larger than 2MB.If you
want to change it, please be sure you see the slice_variable function.
want to change it, please be sure you have read the slice_variable function.
"""
"""
slice_var_up
=
True
slice_var_up
=
True
...
...
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