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
f0804433
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
提交
f0804433
编写于
3月 05, 2019
作者:
D
dengkaipeng
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
add mixup score and label_smooth for yolov3_loss. test=develop
上级
4e8c03bd
变更
2
隐藏空白更改
内联
并排
Showing
2 changed file
with
76 addition
and
26 deletion
+76
-26
paddle/fluid/operators/detection/yolov3_loss_op.cc
paddle/fluid/operators/detection/yolov3_loss_op.cc
+21
-0
paddle/fluid/operators/detection/yolov3_loss_op.h
paddle/fluid/operators/detection/yolov3_loss_op.h
+55
-26
未找到文件。
paddle/fluid/operators/detection/yolov3_loss_op.cc
浏览文件 @
f0804433
...
...
@@ -72,6 +72,18 @@ class Yolov3LossOp : public framework::OperatorWithKernel {
PADDLE_ENFORCE_GT
(
class_num
,
0
,
"Attr(class_num) should be an integer greater then 0."
);
if
(
ctx
->
HasInput
(
"GTScore"
))
{
auto
dim_gtscore
=
ctx
->
GetInputDim
(
"GTScore"
);
PADDLE_ENFORCE_EQ
(
dim_gtscore
.
size
(),
2
,
"Input(GTScore) should be a 2-D tensor"
);
PADDLE_ENFORCE_EQ
(
dim_gtscore
[
0
],
dim_gtbox
[
0
],
"Input(GTBox) and Input(GTScore) dim[0] should be same"
);
PADDLE_ENFORCE_EQ
(
dim_gtscore
[
1
],
dim_gtbox
[
1
],
"Input(GTBox) and Input(GTScore) dim[1] should be same"
);
}
std
::
vector
<
int64_t
>
dim_out
({
dim_x
[
0
]});
ctx
->
SetOutputDim
(
"Loss"
,
framework
::
make_ddim
(
dim_out
));
...
...
@@ -112,6 +124,11 @@ class Yolov3LossOpMaker : public framework::OpProtoAndCheckerMaker {
"This is a 2-D tensor with shape of [N, max_box_num], "
"and each element should be an integer to indicate the "
"box class id."
);
AddInput
(
"GTScore"
,
"The score of GTLabel, This is a 2-D tensor in same shape "
"GTLabel, and score values should in range (0, 1). This "
"input is for GTLabel score can be not 1.0 in image mixup "
"augmentation."
);
AddOutput
(
"Loss"
,
"The output yolov3 loss tensor, "
"This is a 1-D tensor with shape of [N]"
);
...
...
@@ -143,6 +160,8 @@ class Yolov3LossOpMaker : public framework::OpProtoAndCheckerMaker {
AddAttr
<
float
>
(
"ignore_thresh"
,
"The ignore threshold to ignore confidence loss."
)
.
SetDefault
(
0.7
);
AddAttr
<
bool
>
(
"use_label_smooth"
,
"bool,default True"
,
"use label smooth"
)
.
SetDefault
(
true
);
AddComment
(
R"DOC(
This operator generates yolov3 loss based on given predict result and ground
truth boxes.
...
...
@@ -240,6 +259,7 @@ class Yolov3LossGradMaker : public framework::SingleGradOpDescMaker {
op
->
SetInput
(
"X"
,
Input
(
"X"
));
op
->
SetInput
(
"GTBox"
,
Input
(
"GTBox"
));
op
->
SetInput
(
"GTLabel"
,
Input
(
"GTLabel"
));
op
->
SetInput
(
"GTScore"
,
Input
(
"GTScore"
));
op
->
SetInput
(
framework
::
GradVarName
(
"Loss"
),
OutputGrad
(
"Loss"
));
op
->
SetInput
(
"ObjectnessMask"
,
Output
(
"ObjectnessMask"
));
op
->
SetInput
(
"GTMatchMask"
,
Output
(
"GTMatchMask"
));
...
...
@@ -249,6 +269,7 @@ class Yolov3LossGradMaker : public framework::SingleGradOpDescMaker {
op
->
SetOutput
(
framework
::
GradVarName
(
"X"
),
InputGrad
(
"X"
));
op
->
SetOutput
(
framework
::
GradVarName
(
"GTBox"
),
{});
op
->
SetOutput
(
framework
::
GradVarName
(
"GTLabel"
),
{});
op
->
SetOutput
(
framework
::
GradVarName
(
"GTScore"
),
{});
return
std
::
unique_ptr
<
framework
::
OpDesc
>
(
op
);
}
};
...
...
paddle/fluid/operators/detection/yolov3_loss_op.h
浏览文件 @
f0804433
...
...
@@ -37,8 +37,8 @@ static T SigmoidCrossEntropy(T x, T label) {
}
template
<
typename
T
>
static
T
L
2
Loss
(
T
x
,
T
y
)
{
return
0.5
*
(
y
-
x
)
*
(
y
-
x
);
static
T
L
1
Loss
(
T
x
,
T
y
)
{
return
std
::
abs
(
y
-
x
);
}
template
<
typename
T
>
...
...
@@ -47,8 +47,8 @@ static T SigmoidCrossEntropyGrad(T x, T label) {
}
template
<
typename
T
>
static
T
L
2
LossGrad
(
T
x
,
T
y
)
{
return
x
-
y
;
static
T
L
1
LossGrad
(
T
x
,
T
y
)
{
return
x
>
y
?
1.0
:
-
1.0
;
}
static
int
GetMaskIndex
(
std
::
vector
<
int
>
mask
,
int
val
)
{
...
...
@@ -121,47 +121,49 @@ template <typename T>
static
void
CalcBoxLocationLoss
(
T
*
loss
,
const
T
*
input
,
Box
<
T
>
gt
,
std
::
vector
<
int
>
anchors
,
int
an_idx
,
int
box_idx
,
int
gi
,
int
gj
,
int
grid_size
,
int
input_size
,
int
stride
)
{
int
input_size
,
int
stride
,
T
score
)
{
T
tx
=
gt
.
x
*
grid_size
-
gi
;
T
ty
=
gt
.
y
*
grid_size
-
gj
;
T
tw
=
std
::
log
(
gt
.
w
*
input_size
/
anchors
[
2
*
an_idx
]);
T
th
=
std
::
log
(
gt
.
h
*
input_size
/
anchors
[
2
*
an_idx
+
1
]);
T
scale
=
(
2.0
-
gt
.
w
*
gt
.
h
);
T
scale
=
(
2.0
-
gt
.
w
*
gt
.
h
)
*
score
;
loss
[
0
]
+=
SigmoidCrossEntropy
<
T
>
(
input
[
box_idx
],
tx
)
*
scale
;
loss
[
0
]
+=
SigmoidCrossEntropy
<
T
>
(
input
[
box_idx
+
stride
],
ty
)
*
scale
;
loss
[
0
]
+=
L
2
Loss
<
T
>
(
input
[
box_idx
+
2
*
stride
],
tw
)
*
scale
;
loss
[
0
]
+=
L
2
Loss
<
T
>
(
input
[
box_idx
+
3
*
stride
],
th
)
*
scale
;
loss
[
0
]
+=
L
1
Loss
<
T
>
(
input
[
box_idx
+
2
*
stride
],
tw
)
*
scale
;
loss
[
0
]
+=
L
1
Loss
<
T
>
(
input
[
box_idx
+
3
*
stride
],
th
)
*
scale
;
}
template
<
typename
T
>
static
void
CalcBoxLocationLossGrad
(
T
*
input_grad
,
const
T
loss
,
const
T
*
input
,
Box
<
T
>
gt
,
std
::
vector
<
int
>
anchors
,
int
an_idx
,
int
box_idx
,
int
gi
,
int
gj
,
int
grid_size
,
int
input_size
,
int
stride
)
{
int
grid_size
,
int
input_size
,
int
stride
,
T
score
)
{
T
tx
=
gt
.
x
*
grid_size
-
gi
;
T
ty
=
gt
.
y
*
grid_size
-
gj
;
T
tw
=
std
::
log
(
gt
.
w
*
input_size
/
anchors
[
2
*
an_idx
]);
T
th
=
std
::
log
(
gt
.
h
*
input_size
/
anchors
[
2
*
an_idx
+
1
]);
T
scale
=
(
2.0
-
gt
.
w
*
gt
.
h
);
T
scale
=
(
2.0
-
gt
.
w
*
gt
.
h
)
*
score
;
input_grad
[
box_idx
]
=
SigmoidCrossEntropyGrad
<
T
>
(
input
[
box_idx
],
tx
)
*
scale
*
loss
;
input_grad
[
box_idx
+
stride
]
=
SigmoidCrossEntropyGrad
<
T
>
(
input
[
box_idx
+
stride
],
ty
)
*
scale
*
loss
;
input_grad
[
box_idx
+
2
*
stride
]
=
L
2
LossGrad
<
T
>
(
input
[
box_idx
+
2
*
stride
],
tw
)
*
scale
*
loss
;
L
1
LossGrad
<
T
>
(
input
[
box_idx
+
2
*
stride
],
tw
)
*
scale
*
loss
;
input_grad
[
box_idx
+
3
*
stride
]
=
L
2
LossGrad
<
T
>
(
input
[
box_idx
+
3
*
stride
],
th
)
*
scale
*
loss
;
L
1
LossGrad
<
T
>
(
input
[
box_idx
+
3
*
stride
],
th
)
*
scale
*
loss
;
}
template
<
typename
T
>
static
inline
void
CalcLabelLoss
(
T
*
loss
,
const
T
*
input
,
const
int
index
,
const
int
label
,
const
int
class_num
,
const
int
stride
)
{
const
int
stride
,
const
T
pos
,
const
T
neg
,
T
score
)
{
for
(
int
i
=
0
;
i
<
class_num
;
i
++
)
{
T
pred
=
input
[
index
+
i
*
stride
];
loss
[
0
]
+=
SigmoidCrossEntropy
<
T
>
(
pred
,
(
i
==
label
)
?
1.0
:
0.0
)
;
loss
[
0
]
+=
SigmoidCrossEntropy
<
T
>
(
pred
,
(
i
==
label
)
?
pos
:
neg
)
*
score
;
}
}
...
...
@@ -169,11 +171,13 @@ template <typename T>
static
inline
void
CalcLabelLossGrad
(
T
*
input_grad
,
const
T
loss
,
const
T
*
input
,
const
int
index
,
const
int
label
,
const
int
class_num
,
const
int
stride
)
{
const
int
stride
,
const
T
pos
,
const
T
neg
,
T
score
)
{
for
(
int
i
=
0
;
i
<
class_num
;
i
++
)
{
T
pred
=
input
[
index
+
i
*
stride
];
input_grad
[
index
+
i
*
stride
]
=
SigmoidCrossEntropyGrad
<
T
>
(
pred
,
(
i
==
label
)
?
1.0
:
0.0
)
*
loss
;
SigmoidCrossEntropyGrad
<
T
>
(
pred
,
(
i
==
label
)
?
pos
:
neg
)
*
score
*
loss
;
}
}
...
...
@@ -188,8 +192,8 @@ static inline void CalcObjnessLoss(T* loss, const T* input, const T* objness,
for
(
int
l
=
0
;
l
<
w
;
l
++
)
{
T
obj
=
objness
[
k
*
w
+
l
];
if
(
obj
>
1e-5
)
{
// positive sample: obj =
1
loss
[
i
]
+=
SigmoidCrossEntropy
<
T
>
(
input
[
k
*
w
+
l
],
1.0
);
// positive sample: obj =
mixup score
loss
[
i
]
+=
SigmoidCrossEntropy
<
T
>
(
input
[
k
*
w
+
l
],
1.0
)
*
obj
;
}
else
if
(
obj
>
-
0.5
)
{
// negetive sample: obj = 0
loss
[
i
]
+=
SigmoidCrossEntropy
<
T
>
(
input
[
k
*
w
+
l
],
0.0
);
...
...
@@ -215,7 +219,8 @@ static inline void CalcObjnessLossGrad(T* input_grad, const T* loss,
T
obj
=
objness
[
k
*
w
+
l
];
if
(
obj
>
1e-5
)
{
input_grad
[
k
*
w
+
l
]
=
SigmoidCrossEntropyGrad
<
T
>
(
input
[
k
*
w
+
l
],
1.0
)
*
loss
[
i
];
SigmoidCrossEntropyGrad
<
T
>
(
input
[
k
*
w
+
l
],
1.0
)
*
obj
*
loss
[
i
];
}
else
if
(
obj
>
-
0.5
)
{
input_grad
[
k
*
w
+
l
]
=
SigmoidCrossEntropyGrad
<
T
>
(
input
[
k
*
w
+
l
],
0.0
)
*
loss
[
i
];
...
...
@@ -252,6 +257,7 @@ class Yolov3LossKernel : public framework::OpKernel<T> {
auto
*
input
=
ctx
.
Input
<
Tensor
>
(
"X"
);
auto
*
gt_box
=
ctx
.
Input
<
Tensor
>
(
"GTBox"
);
auto
*
gt_label
=
ctx
.
Input
<
Tensor
>
(
"GTLabel"
);
auto
*
gt_score
=
ctx
.
Input
<
Tensor
>
(
"GTScore"
);
auto
*
loss
=
ctx
.
Output
<
Tensor
>
(
"Loss"
);
auto
*
objness_mask
=
ctx
.
Output
<
Tensor
>
(
"ObjectnessMask"
);
auto
*
gt_match_mask
=
ctx
.
Output
<
Tensor
>
(
"GTMatchMask"
);
...
...
@@ -260,6 +266,7 @@ class Yolov3LossKernel : public framework::OpKernel<T> {
int
class_num
=
ctx
.
Attr
<
int
>
(
"class_num"
);
float
ignore_thresh
=
ctx
.
Attr
<
float
>
(
"ignore_thresh"
);
int
downsample_ratio
=
ctx
.
Attr
<
int
>
(
"downsample_ratio"
);
bool
use_label_smooth
=
ctx
.
Attr
<
bool
>
(
"use_label_smooth"
);
const
int
n
=
input
->
dims
()[
0
];
const
int
h
=
input
->
dims
()[
2
];
...
...
@@ -272,9 +279,17 @@ class Yolov3LossKernel : public framework::OpKernel<T> {
const
int
stride
=
h
*
w
;
const
int
an_stride
=
(
class_num
+
5
)
*
stride
;
T
label_pos
=
1.0
;
T
label_neg
=
0.0
;
if
(
use_label_smooth
)
{
label_pos
=
1.0
-
1.0
/
static_cast
<
T
>
(
class_num
);
label_neg
=
1.0
/
static_cast
<
T
>
(
class_num
);
}
const
T
*
input_data
=
input
->
data
<
T
>
();
const
T
*
gt_box_data
=
gt_box
->
data
<
T
>
();
const
int
*
gt_label_data
=
gt_label
->
data
<
int
>
();
const
T
*
gt_score_data
=
gt_score
->
data
<
T
>
();
T
*
loss_data
=
loss
->
mutable_data
<
T
>
({
n
},
ctx
.
GetPlace
());
memset
(
loss_data
,
0
,
loss
->
numel
()
*
sizeof
(
T
));
T
*
obj_mask_data
=
...
...
@@ -355,19 +370,20 @@ class Yolov3LossKernel : public framework::OpKernel<T> {
int
mask_idx
=
GetMaskIndex
(
anchor_mask
,
best_n
);
gt_match_mask_data
[
i
*
b
+
t
]
=
mask_idx
;
if
(
mask_idx
>=
0
)
{
T
score
=
gt_score_data
[
i
*
b
+
t
];
int
box_idx
=
GetEntryIndex
(
i
,
mask_idx
,
gj
*
w
+
gi
,
mask_num
,
an_stride
,
stride
,
0
);
CalcBoxLocationLoss
<
T
>
(
loss_data
+
i
,
input_data
,
gt
,
anchors
,
best_n
,
box_idx
,
gi
,
gj
,
h
,
input_size
,
stride
);
box_idx
,
gi
,
gj
,
h
,
input_size
,
stride
,
score
);
int
obj_idx
=
(
i
*
mask_num
+
mask_idx
)
*
stride
+
gj
*
w
+
gi
;
obj_mask_data
[
obj_idx
]
=
1.0
;
obj_mask_data
[
obj_idx
]
=
score
;
int
label
=
gt_label_data
[
i
*
b
+
t
];
int
label_idx
=
GetEntryIndex
(
i
,
mask_idx
,
gj
*
w
+
gi
,
mask_num
,
an_stride
,
stride
,
5
);
CalcLabelLoss
<
T
>
(
loss_data
+
i
,
input_data
,
label_idx
,
label
,
class_num
,
stride
);
class_num
,
stride
,
label_pos
,
label_neg
,
score
);
}
}
}
...
...
@@ -384,6 +400,7 @@ class Yolov3LossGradKernel : public framework::OpKernel<T> {
auto
*
input
=
ctx
.
Input
<
Tensor
>
(
"X"
);
auto
*
gt_box
=
ctx
.
Input
<
Tensor
>
(
"GTBox"
);
auto
*
gt_label
=
ctx
.
Input
<
Tensor
>
(
"GTLabel"
);
auto
*
gt_score
=
ctx
.
Input
<
Tensor
>
(
"GTScore"
);
auto
*
input_grad
=
ctx
.
Output
<
Tensor
>
(
framework
::
GradVarName
(
"X"
));
auto
*
loss_grad
=
ctx
.
Input
<
Tensor
>
(
framework
::
GradVarName
(
"Loss"
));
auto
*
objness_mask
=
ctx
.
Input
<
Tensor
>
(
"ObjectnessMask"
);
...
...
@@ -392,6 +409,7 @@ class Yolov3LossGradKernel : public framework::OpKernel<T> {
auto
anchor_mask
=
ctx
.
Attr
<
std
::
vector
<
int
>>
(
"anchor_mask"
);
int
class_num
=
ctx
.
Attr
<
int
>
(
"class_num"
);
int
downsample_ratio
=
ctx
.
Attr
<
int
>
(
"downsample_ratio"
);
bool
use_label_smooth
=
ctx
.
Attr
<
bool
>
(
"use_label_smooth"
);
const
int
n
=
input_grad
->
dims
()[
0
];
const
int
c
=
input_grad
->
dims
()[
1
];
...
...
@@ -404,9 +422,17 @@ class Yolov3LossGradKernel : public framework::OpKernel<T> {
const
int
stride
=
h
*
w
;
const
int
an_stride
=
(
class_num
+
5
)
*
stride
;
T
label_pos
=
1.0
;
T
label_neg
=
0.0
;
if
(
use_label_smooth
)
{
label_pos
=
1.0
-
1.0
/
static_cast
<
T
>
(
class_num
);
label_neg
=
1.0
/
static_cast
<
T
>
(
class_num
);
}
const
T
*
input_data
=
input
->
data
<
T
>
();
const
T
*
gt_box_data
=
gt_box
->
data
<
T
>
();
const
int
*
gt_label_data
=
gt_label
->
data
<
int
>
();
const
T
*
gt_score_data
=
gt_score
->
data
<
T
>
();
const
T
*
loss_grad_data
=
loss_grad
->
data
<
T
>
();
const
T
*
obj_mask_data
=
objness_mask
->
data
<
T
>
();
const
int
*
gt_match_mask_data
=
gt_match_mask
->
data
<
int
>
();
...
...
@@ -418,21 +444,24 @@ class Yolov3LossGradKernel : public framework::OpKernel<T> {
for
(
int
t
=
0
;
t
<
b
;
t
++
)
{
int
mask_idx
=
gt_match_mask_data
[
i
*
b
+
t
];
if
(
mask_idx
>=
0
)
{
T
score
=
gt_score_data
[
i
*
b
+
t
];
Box
<
T
>
gt
=
GetGtBox
(
gt_box_data
,
i
,
b
,
t
);
int
gi
=
static_cast
<
int
>
(
gt
.
x
*
w
);
int
gj
=
static_cast
<
int
>
(
gt
.
y
*
h
);
int
box_idx
=
GetEntryIndex
(
i
,
mask_idx
,
gj
*
w
+
gi
,
mask_num
,
an_stride
,
stride
,
0
);
CalcBoxLocationLossGrad
<
T
>
(
input_grad_data
,
loss_grad_data
[
i
],
input_data
,
gt
,
anchors
,
anchor_mask
[
mask_idx
],
box_idx
,
gi
,
gj
,
h
,
input_size
,
stride
);
CalcBoxLocationLossGrad
<
T
>
(
input_grad_data
,
loss_grad_data
[
i
],
input_data
,
gt
,
anchors
,
anchor_mask
[
mask_idx
],
box_idx
,
gi
,
gj
,
h
,
input_size
,
stride
,
score
);
int
label
=
gt_label_data
[
i
*
b
+
t
];
int
label_idx
=
GetEntryIndex
(
i
,
mask_idx
,
gj
*
w
+
gi
,
mask_num
,
an_stride
,
stride
,
5
);
CalcLabelLossGrad
<
T
>
(
input_grad_data
,
loss_grad_data
[
i
],
input_data
,
label_idx
,
label
,
class_num
,
stride
);
label_idx
,
label
,
class_num
,
stride
,
label_pos
,
label_neg
,
score
);
}
}
}
...
...
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