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
s920243400
PaddleDetection
提交
b1e56994
P
PaddleDetection
项目概览
s920243400
/
PaddleDetection
与 Fork 源项目一致
Fork自
PaddlePaddle / PaddleDetection
通知
2
Star
0
Fork
0
代码
文件
提交
分支
Tags
贡献者
分支图
Diff
Issue
0
列表
看板
标记
里程碑
合并请求
0
Wiki
0
Wiki
分析
仓库
DevOps
项目成员
Pages
P
PaddleDetection
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
0
Issue
0
列表
看板
标记
里程碑
合并请求
0
合并请求
0
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
未验证
提交
b1e56994
编写于
2月 25, 2019
作者:
K
Kaipeng Deng
提交者:
GitHub
2月 25, 2019
浏览文件
操作
浏览文件
下载
差异文件
Merge pull request #15901 from heavengate/release/1.3
cherry-pick pool/adaptive_pool/yolov3_loss doc fix
上级
4b3f9e5c
64e33566
变更
4
显示空白变更内容
内联
并排
Showing
4 changed file
with
201 addition
and
100 deletion
+201
-100
paddle/fluid/operators/detection/yolov3_loss_op.cc
paddle/fluid/operators/detection/yolov3_loss_op.cc
+23
-17
paddle/fluid/operators/pool_op.cc
paddle/fluid/operators/pool_op.cc
+87
-57
python/paddle/fluid/layers/detection.py
python/paddle/fluid/layers/detection.py
+10
-9
python/paddle/fluid/layers/nn.py
python/paddle/fluid/layers/nn.py
+81
-17
未找到文件。
paddle/fluid/operators/detection/yolov3_loss_op.cc
浏览文件 @
b1e56994
...
@@ -144,34 +144,40 @@ class Yolov3LossOpMaker : public framework::OpProtoAndCheckerMaker {
...
@@ -144,34 +144,40 @@ class Yolov3LossOpMaker : public framework::OpProtoAndCheckerMaker {
"The ignore threshold to ignore confidence loss."
)
"The ignore threshold to ignore confidence loss."
)
.
SetDefault
(
0.7
);
.
SetDefault
(
0.7
);
AddComment
(
R"DOC(
AddComment
(
R"DOC(
This operator generate
yolov3 loss by
given predict result and ground
This operator generate
s yolov3 loss based on
given predict result and ground
truth boxes.
truth boxes.
The output of previous network is in shape [N, C, H, W], while H and W
The output of previous network is in shape [N, C, H, W], while H and W
should be the same,
specify the grid size, each grid point predict given
should be the same,
H and W specify the grid size, each grid point predict
number boxes, this given number is specified by anchors, it should be
given number boxes, this given number, which following will be represented as S,
half anchors length, which following will be represented as S. In the
is specified by the number of anchors, In the second dimension(the channel
second dimention(the channel dimention), C should be S * (class_num + 5),
dimension), C should be equal to S * (class_num + 5), class_num is the object
c
lass_num is the box categoriy number of source dataset(such as coco),
c
ategory number of source dataset(such as 80 in coco dataset), so in the
s
o in the second dimention, stores 4 box location coordinates x, y, w, h
s
econd(channel) dimension, apart from 4 box location coordinates x, y, w, h,
a
nd
confidence score of the box and class one-hot key of each anchor box.
a
lso includes
confidence score of the box and class one-hot key of each anchor box.
While the 4 location coordinates if $$tx, ty, tw, th$$
, the box predictions
Assume the 4 location coordinates are :math:`t_x, t_y, t_w, t_h`
, the box predictions
correspnd to
:
should be as follows
:
$$
$$
b_x = \sigma(t_x) + c_x
b_x = \\sigma(t_x) + c_x
b_y = \sigma(t_y) + c_y
$$
$$
b_y = \\sigma(t_y) + c_y
$$
$$
b_w = p_w e^{t_w}
b_w = p_w e^{t_w}
$$
$$
b_h = p_h e^{t_h}
b_h = p_h e^{t_h}
$$
$$
While $$c_x, c_y$$ is the left top corner of current grid and $$p_w, p_h$$
In the equation above, :math:`c_x, c_y` is the left top corner of current grid
is specified by anchors.
and :math:`p_w, p_h`
is specified by anchors.
As for confidence score, it is the logistic regression value of IoU between
As for confidence score, it is the logistic regression value of IoU between
anchor boxes and ground truth boxes, the score of the anchor box which has
anchor boxes and ground truth boxes, the score of the anchor box which has
the max IoU should be 1, and if the anchor box has IoU bigger th
e
n ignore
the max IoU should be 1, and if the anchor box has IoU bigger th
a
n ignore
thresh, the confidence score loss of this anchor box will be ignored.
thresh, the confidence score loss of this anchor box will be ignored.
Therefore, the yolov3 loss consist of three major parts, box location loss,
Therefore, the yolov3 loss consist of three major parts, box location loss,
...
@@ -186,13 +192,13 @@ class Yolov3LossOpMaker : public framework::OpProtoAndCheckerMaker {
...
@@ -186,13 +192,13 @@ class Yolov3LossOpMaker : public framework::OpProtoAndCheckerMaker {
In order to trade off box coordinate losses between big boxes and small
In order to trade off box coordinate losses between big boxes and small
boxes, box coordinate losses will be mutiplied by scale weight, which is
boxes, box coordinate losses will be mutiplied by scale weight, which is
calculated as follow.
calculated as follow
s
.
$$
$$
weight_{box} = 2.0 - t_w * t_h
weight_{box} = 2.0 - t_w * t_h
$$
$$
Final loss will be represented as follow.
Final loss will be represented as follow
s
.
$$
$$
loss = (loss_{xy} + loss_{wh}) * weight_{box}
loss = (loss_{xy} + loss_{wh}) * weight_{box}
...
...
paddle/fluid/operators/pool_op.cc
浏览文件 @
b1e56994
...
@@ -168,9 +168,10 @@ void Pool2dOpMaker::Make() {
...
@@ -168,9 +168,10 @@ void Pool2dOpMaker::Make() {
"be ignored."
);
// TODO(Chengduo): Add checker.
"be ignored."
);
// TODO(Chengduo): Add checker.
// (Currently,
// (Currently,
// TypedAttrChecker don't support vector type.)
// TypedAttrChecker don't support vector type.)
AddAttr
<
bool
>
(
"global_pooling"
,
AddAttr
<
bool
>
(
"global_pooling"
,
"(bool, default false) Whether to use the global pooling. "
"(bool, default false) Whether to use the global pooling. "
"If global_pooling = true, k
size and paddings will be ignored."
)
"If global_pooling = true, kernel
size and paddings will be ignored."
)
.
SetDefault
(
false
);
.
SetDefault
(
false
);
AddAttr
<
std
::
vector
<
int
>>
(
"strides"
,
AddAttr
<
std
::
vector
<
int
>>
(
"strides"
,
"(vector<int>, default {1, 1}), strides(height, "
"(vector<int>, default {1, 1}), strides(height, "
...
@@ -182,7 +183,7 @@ void Pool2dOpMaker::Make() {
...
@@ -182,7 +183,7 @@ void Pool2dOpMaker::Make() {
"paddings"
,
"paddings"
,
"(vector<int>, default {0,0}), paddings(height, width) of pooling "
"(vector<int>, default {0,0}), paddings(height, width) of pooling "
"operator."
"operator."
"If global_pooling = true, paddings and ksize will be ignored."
)
"If global_pooling = true, paddings and k
ernel
size will be ignored."
)
.
SetDefault
({
0
,
0
});
.
SetDefault
({
0
,
0
});
AddAttr
<
bool
>
(
AddAttr
<
bool
>
(
"exclusive"
,
"exclusive"
,
...
@@ -204,7 +205,7 @@ void Pool2dOpMaker::Make() {
...
@@ -204,7 +205,7 @@ void Pool2dOpMaker::Make() {
.
SetDefault
(
false
);
.
SetDefault
(
false
);
AddAttr
<
bool
>
(
AddAttr
<
bool
>
(
"ceil_mode"
,
"ceil_mode"
,
"(bool, default false) Wether to use the ceil function to calculate "
"(bool, default false) W
h
ether to use the ceil function to calculate "
"output height and width. False is the default. If it is set to False, "
"output height and width. False is the default. If it is set to False, "
"the floor function will be used."
)
"the floor function will be used."
)
.
SetDefault
(
false
);
.
SetDefault
(
false
);
...
@@ -259,31 +260,40 @@ Example:
...
@@ -259,31 +260,40 @@ Example:
W_{out} = \\frac{(W_{in} - ksize[1] + 2 * paddings[1] + strides[1] - 1)}{strides[1]} + 1
W_{out} = \\frac{(W_{in} - ksize[1] + 2 * paddings[1] + strides[1] - 1)}{strides[1]} + 1
$$
$$
For exclusive =
tru
e:
For exclusive =
fals
e:
$$
$$
hstart = i * strides[0] - paddings[0]
hstart = i * strides[0] - paddings[0]
$$
$$
hend = hstart + ksize[0]
hend = hstart + ksize[0]
$$
$$
wstart = j * strides[1] - paddings[1]
wstart = j * strides[1] - paddings[1]
$$
$$
wend = wstart + ksize[1]
wend = wstart + ksize[1]
$$
$$
Output(i ,j) = \\frac{sum(Input[hstart:hend, wstart:wend])}{ksize[0] * ksize[1]}
Output(i ,j) = \\frac{sum(Input[hstart:hend, wstart:wend])}{ksize[0] * ksize[1]}
$$
$$
For exclusive = false:
For exclusive = true:
$$
$$
hstart = max(0, i * strides[0] - paddings[0])
hstart = max(0, i * strides[0] - paddings[0])
$$
$$
hend = min(H, hstart + ksize[0])
hend = min(H, hstart + ksize[0])
$$
$$
wstart = max(0, j * strides[1] - paddings[1])
wstart = max(0, j * strides[1] - paddings[1])
$$
$$
wend = min(W, wstart + ksize[1])
wend = min(W, wstart + ksize[1])
Output(i ,j) = \\frac{sum(Input[hstart:hend, wstart:wend])}{(hend - hstart) * (wend - wstart)}
$$
$$
For adaptive = true:
$$
$$
hstart = floor(i * H_{in} / H_{out})
hend = ceil((i + 1) * H_{in} / H_{out})
wstart = floor(j * W_{in} / W_{out})
wend = ceil((j + 1) * W_{in} / W_{out})
Output(i ,j) = \\frac{sum(Input[hstart:hend, wstart:wend])}{(hend - hstart) * (wend - wstart)}
Output(i ,j) = \\frac{sum(Input[hstart:hend, wstart:wend])}{(hend - hstart) * (wend - wstart)}
$$
$$
)DOC"
);
)DOC"
);
}
}
...
@@ -324,7 +334,7 @@ void Pool3dOpMaker::Make() {
...
@@ -324,7 +334,7 @@ void Pool3dOpMaker::Make() {
AddAttr
<
bool
>
(
AddAttr
<
bool
>
(
"global_pooling"
,
"global_pooling"
,
"(bool, default false) Whether to use the global pooling. "
"(bool, default false) Whether to use the global pooling. "
"If global_pooling = true, k
size and paddings wille
be ignored."
)
"If global_pooling = true, k
ernel size and paddings will
be ignored."
)
.
SetDefault
(
false
);
.
SetDefault
(
false
);
AddAttr
<
std
::
vector
<
int
>>
(
AddAttr
<
std
::
vector
<
int
>>
(
"strides"
,
"strides"
,
...
@@ -359,7 +369,7 @@ void Pool3dOpMaker::Make() {
...
@@ -359,7 +369,7 @@ void Pool3dOpMaker::Make() {
.
SetDefault
(
false
);
.
SetDefault
(
false
);
AddAttr
<
bool
>
(
AddAttr
<
bool
>
(
"ceil_mode"
,
"ceil_mode"
,
"(bool, default false) Wether to use the ceil function to calculate "
"(bool, default false) W
h
ether to use the ceil function to calculate "
"output height and width. False is the default. If it is set to False, "
"output height and width. False is the default. If it is set to False, "
"the floor function will be used."
)
"the floor function will be used."
)
.
SetDefault
(
false
);
.
SetDefault
(
false
);
...
@@ -393,45 +403,65 @@ Example:
...
@@ -393,45 +403,65 @@ Example:
Out shape: $(N, C, D_{out}, H_{out}, W_{out})$
Out shape: $(N, C, D_{out}, H_{out}, W_{out})$
For ceil_mode = false:
For ceil_mode = false:
$$
$$
D_{out} = \frac{(D_{in} - ksize[0] + 2 * paddings[0])}{strides[0]} + 1 \\
D_{out} = \\frac{(D_{in} - ksize[0] + 2 * paddings[0])}{strides[0]} + 1
H_{out} = \frac{(H_{in} - ksize[1] + 2 * paddings[1])}{strides[1]} + 1 \\
$$
W_{out} = \frac{(W_{in} - ksize[2] + 2 * paddings[2])}{strides[2]} + 1
$$
H_{out} = \\frac{(H_{in} - ksize[1] + 2 * paddings[1])}{strides[2]} + 1
$$
$$
W_{out} = \\frac{(W_{in} - ksize[2] + 2 * paddings[2])}{strides[2]} + 1
$$
$$
For ceil_mode = true:
For ceil_mode = true:
$$
$$
D_{out} = \frac{(D_{in} - ksize[0] + 2 * paddings[0] + strides[0] -1)}{strides[0]} + 1 \\
D_{out} = \\frac{(D_{in} - ksize[0] + 2 * paddings[0] + strides[0] -1)}{strides[0]} + 1
H_{out} = \frac{(H_{in} - ksize[1] + 2 * paddings[1] + strides[1] -1)}{strides[1]} + 1 \\
W_{out} = \frac{(W_{in} - ksize[2] + 2 * paddings[2] + strides[2] -1)}{strides[2]} + 1
$$
$$
For exclusive = true:
$$
H_{out} = \\frac{(H_{in} - ksize[1] + 2 * paddings[1] + strides[1] -1)}{strides[1]} + 1
$$
$$
W_{out} = \\frac{(W_{in} - ksize[2] + 2 * paddings[2] + strides[2] -1)}{strides[2]} + 1
$$
For exclusive = false:
$$
$$
dstart = i * strides[0] - paddings[0]
dstart = i * strides[0] - paddings[0]
$$
$$
dend = dstart + ksize[0]
dend = dstart + ksize[0]
$$
$$
hstart = j * strides[1] - paddings[1]
hstart = j * strides[1] - paddings[1]
$$
$$
hend = hstart + ksize[1]
hend = hstart + ksize[1]
$$
$$
wstart = k * strides[2] - paddings[2]
wstart = k * strides[2] - paddings[2]
$$
$$
wend = wstart + ksize[2]
wend = wstart + ksize[2]
$$
$$
Output(i ,j, k) = \\frac{sum(Input[dstart:dend, hstart:hend, wstart:wend])}{ksize[0] * ksize[1] * ksize[2]}
Output(i ,j, k) = \\frac{sum(Input[dstart:dend, hstart:hend, wstart:wend])}{ksize[0] * ksize[1] * ksize[2]}
$$
$$
For exclusive = false:
For exclusive = true:
$$
$$
dstart = max(0, i * strides[0] - paddings[0])
dstart = max(0, i * strides[0] - paddings[0])
$$
$$
dend = min(D, dstart + ksize[0])
dend = min(D, dstart + ksize[0])
hstart = max(0, j * strides[1] - paddings[1])
$$
$$
hend = min(H, hstart + ksize[1])
hend = min(H, hstart + ksize[1])
$$
$$
wstart = max(0, k * strides[2] - paddings[2])
wstart = max(0, k * strides[2] - paddings[2])
$$
$$
wend = min(W, wstart + ksize[2])
wend = min(W, wstart + ksize[2])
Output(i ,j, k) = \\frac{sum(Input[dstart:dend, hstart:hend, wstart:wend])}{(dend - dstart) * (hend - hstart) * (wend - wstart)}
$$
$$
For adaptive = true:
$$
$$
dstart = floor(i * D_{in} / D_{out})
dend = ceil((i + 1) * D_{in} / D_{out})
hstart = floor(j * H_{in} / H_{out})
hend = ceil((j + 1) * H_{in} / H_{out})
wstart = floor(k * W_{in} / W_{out})
wend = ceil((k + 1) * W_{in} / W_{out})
Output(i ,j, k) = \\frac{sum(Input[dstart:dend, hstart:hend, wstart:wend])}{(dend - dstart) * (hend - hstart) * (wend - wstart)}
Output(i ,j, k) = \\frac{sum(Input[dstart:dend, hstart:hend, wstart:wend])}{(dend - dstart) * (hend - hstart) * (wend - wstart)}
$$
$$
...
...
python/paddle/fluid/layers/detection.py
浏览文件 @
b1e56994
...
@@ -551,9 +551,10 @@ def yolov3_loss(x,
...
@@ -551,9 +551,10 @@ def yolov3_loss(x,
gtbox = fluid.layers.data(name='gtbox', shape=[6, 5], dtype='float32')
gtbox = fluid.layers.data(name='gtbox', shape=[6, 5], dtype='float32')
gtlabel = fluid.layers.data(name='gtlabel', shape=[6, 1], dtype='int32')
gtlabel = fluid.layers.data(name='gtlabel', shape=[6, 1], dtype='int32')
anchors = [10, 13, 16, 30, 33, 23, 30, 61, 62, 45, 59, 119, 116, 90, 156, 198, 373, 326]
anchors = [10, 13, 16, 30, 33, 23, 30, 61, 62, 45, 59, 119, 116, 90, 156, 198, 373, 326]
anchors = [0, 1, 2]
anchor_mask = [0, 1, 2]
loss = fluid.layers.yolov3_loss(x=x, gtbox=gtbox, class_num=80, anchors=anchors,
loss = fluid.layers.yolov3_loss(x=x, gtbox=gtbox, gtlabel=gtlabel, anchors=anchors,
ignore_thresh=0.5, downsample_ratio=32)
anchor_mask=anchor_mask, class_num=80,
ignore_thresh=0.7, downsample_ratio=32)
"""
"""
helper
=
LayerHelper
(
'yolov3_loss'
,
**
locals
())
helper
=
LayerHelper
(
'yolov3_loss'
,
**
locals
())
...
...
python/paddle/fluid/layers/nn.py
浏览文件 @
b1e56994
...
@@ -2441,7 +2441,7 @@ def pool2d(input,
...
@@ -2441,7 +2441,7 @@ def pool2d(input,
data = fluid.layers.data(
data = fluid.layers.data(
name='data', shape=[3, 32, 32], dtype='float32')
name='data', shape=[3, 32, 32], dtype='float32')
conv
2d = fluid.layers.pool2d(
pool
2d = fluid.layers.pool2d(
input=data,
input=data,
pool_size=2,
pool_size=2,
pool_type='max',
pool_type='max',
...
@@ -2490,6 +2490,7 @@ def pool2d(input,
...
@@ -2490,6 +2490,7 @@ def pool2d(input,
return
pool_out
return
pool_out
@
templatedoc
()
def
pool3d
(
input
,
def
pool3d
(
input
,
pool_size
=-
1
,
pool_size
=-
1
,
pool_type
=
"max"
,
pool_type
=
"max"
,
...
@@ -2501,13 +2502,19 @@ def pool3d(input,
...
@@ -2501,13 +2502,19 @@ def pool3d(input,
name
=
None
,
name
=
None
,
exclusive
=
True
):
exclusive
=
True
):
"""
"""
This function adds the operator for pooling in 3-dimensions, using the
${comment}
pooling configurations mentioned in input parameters.
Args:
Args:
input (Variable): ${input_comment}
input (Variable): The input tensor of pooling operator. The format of
pool_size (int): ${ksize_comment}
input tensor is NCDHW, where N is batch size, C is
pool_type (str): ${pooling_type_comment}
the number of channels, D is the depth of the feature,
H is the height of the feature, and W is the width
of the feature.
pool_size (int|list|tuple): The pool kernel size. If pool kernel size
is a tuple or list, it must contain three integers,
(pool_size_Depth, pool_size_Height, pool_size_Width).
Otherwise, the pool kernel size will be the cube of an int.
pool_type (string): ${pooling_type_comment}
pool_stride (int): stride of the pooling layer.
pool_stride (int): stride of the pooling layer.
pool_padding (int): padding size.
pool_padding (int): padding size.
global_pooling (bool): ${global_pooling_comment}
global_pooling (bool): ${global_pooling_comment}
...
@@ -2520,6 +2527,19 @@ def pool3d(input,
...
@@ -2520,6 +2527,19 @@ def pool3d(input,
Returns:
Returns:
Variable: output of pool3d layer.
Variable: output of pool3d layer.
Examples:
.. code-block:: python
data = fluid.layers.data(
name='data', shape=[3, 32, 32, 32], dtype='float32')
pool3d = fluid.layers.pool3d(
input=data,
pool_size=2,
pool_type='max',
pool_stride=1,
global_pooling=False)
"""
"""
if
pool_type
not
in
[
"max"
,
"avg"
]:
if
pool_type
not
in
[
"max"
,
"avg"
]:
raise
ValueError
(
raise
ValueError
(
...
@@ -2569,7 +2589,27 @@ def adaptive_pool2d(input,
...
@@ -2569,7 +2589,27 @@ def adaptive_pool2d(input,
require_index
=
False
,
require_index
=
False
,
name
=
None
):
name
=
None
):
"""
"""
${comment}
**Adaptive Pool2d Operator**
The adaptive_pool2d operation calculates the output based on the input, pool_size,
pool_type parameters. Input(X) and output(Out) are in NCHW format, where N is batch
size, C is the number of channels, H is the height of the feature, and W is
the width of the feature. Parameters(pool_size) should contain two elements which
represent height and width, respectively. Also the H and W dimensions of output(Out)
is same as Parameter(pool_size).
For average adaptive pool2d:
.. math::
hstart &= floor(i * H_{in} / H_{out})
hend &= ceil((i + 1) * H_{in} / H_{out})
wstart &= floor(j * W_{in} / W_{out})
wend &= ceil((j + 1) * W_{in} / W_{out})
Output(i ,j) &=
\\
frac{sum(Input[hstart:hend, wstart:wend])}{(hend - hstart) * (wend - wstart)}
Args:
Args:
input (Variable): The input tensor of pooling operator. The format of
input (Variable): The input tensor of pooling operator. The format of
...
@@ -2579,8 +2619,8 @@ def adaptive_pool2d(input,
...
@@ -2579,8 +2619,8 @@ def adaptive_pool2d(input,
pool_size (int|list|tuple): The pool kernel size. If pool kernel size is a tuple or list,
pool_size (int|list|tuple): The pool kernel size. If pool kernel size is a tuple or list,
it must contain two integers, (pool_size_Height, pool_size_Width).
it must contain two integers, (pool_size_Height, pool_size_Width).
pool_type: ${pooling_type_comment}
pool_type: ${pooling_type_comment}
require_index (bool): If true, the index of max pooling point
along with outputs.
require_index (bool): If true, the index of max pooling point
will be returned along
i
t cannot be set in average pooling type.
with outputs. I
t cannot be set in average pooling type.
name (str|None): A name for this layer(optional). If set None, the
name (str|None): A name for this layer(optional). If set None, the
layer will be named automatically.
layer will be named automatically.
...
@@ -2661,18 +2701,42 @@ def adaptive_pool3d(input,
...
@@ -2661,18 +2701,42 @@ def adaptive_pool3d(input,
require_index
=
False
,
require_index
=
False
,
name
=
None
):
name
=
None
):
"""
"""
${comment}
**Adaptive Pool3d Operator**
The adaptive_pool3d operation calculates the output based on the input, pool_size,
pool_type parameters. Input(X) and output(Out) are in NCDHW format, where N is batch
size, C is the number of channels, D is the depth of the feature, H is the height of
the feature, and W is the width of the feature. Parameters(pool_size) should contain
three elements which represent height and width, respectively. Also the D, H and W
dimensions of output(Out) is same as Parameter(pool_size).
For average adaptive pool3d:
.. math::
dstart &= floor(i * D_{in} / D_{out})
dend &= ceil((i + 1) * D_{in} / D_{out})
hstart &= floor(j * H_{in} / H_{out})
hend &= ceil((j + 1) * H_{in} / H_{out})
wstart &= floor(k * W_{in} / W_{out})
wend &= ceil((k + 1) * W_{in} / W_{out})
Output(i ,j, k) &=
\\
frac{sum(Input[dstart:dend, hstart:hend, wstart:wend])}{(dend - dstart) * (hend - hstart) * (wend - wstart)}
Args:
Args:
input (Variable): The input tensor of pooling operator. The format of
input (Variable): The input tensor of pooling operator. The format of
input tensor is NCHW, where N is batch size, C is
input tensor is NC
D
HW, where N is batch size, C is
the number of channels,
H is the height of the
the number of channels,
D is the depth of the feature,
feature, and W is the width of the feature.
H is the height of the
feature, and W is the width of the feature.
pool_size (int|list|tuple): The pool kernel size. If pool kernel size is a tuple or list,
pool_size (int|list|tuple): The pool kernel size. If pool kernel size is a tuple or list,
it must contain t
wo
integers, (Depth, Height, Width).
it must contain t
hree
integers, (Depth, Height, Width).
pool_type: ${pooling_type_comment}
pool_type: ${pooling_type_comment}
require_index (bool): If true, the index of max pooling point
along with outputs.
require_index (bool): If true, the index of max pooling point
will be returned along
i
t cannot be set in average pooling type.
with outputs. I
t cannot be set in average pooling type.
name (str|None): A name for this layer(optional). If set None, the
name (str|None): A name for this layer(optional). If set None, the
layer will be named automatically.
layer will be named automatically.
...
@@ -2709,7 +2773,7 @@ def adaptive_pool3d(input,
...
@@ -2709,7 +2773,7 @@ def adaptive_pool3d(input,
name='data', shape=[3, 32, 32], dtype='float32')
name='data', shape=[3, 32, 32], dtype='float32')
pool_out, mask = fluid.layers.adaptive_pool3d(
pool_out, mask = fluid.layers.adaptive_pool3d(
input=data,
input=data,
pool_size=[3, 3],
pool_size=[3, 3
, 3
],
pool_type='avg')
pool_type='avg')
"""
"""
if
pool_type
not
in
[
"max"
,
"avg"
]:
if
pool_type
not
in
[
"max"
,
"avg"
]:
...
...
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