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
PaddleClas
提交
5a1c4210
P
PaddleClas
项目概览
PaddlePaddle
/
PaddleClas
接近 2 年 前同步成功
通知
116
Star
4999
Fork
1114
代码
文件
提交
分支
Tags
贡献者
分支图
Diff
Issue
19
列表
看板
标记
里程碑
合并请求
6
Wiki
0
Wiki
分析
仓库
DevOps
项目成员
Pages
P
PaddleClas
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
19
Issue
19
列表
看板
标记
里程碑
合并请求
6
合并请求
6
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
5a1c4210
编写于
6月 11, 2020
作者:
W
wqz960
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
add ghostnet
上级
bd67368c
变更
4
隐藏空白更改
内联
并排
Showing
4 changed file
with
544 addition
and
0 deletion
+544
-0
configs/GhostNet/GhostNet_0_5.yaml
configs/GhostNet/GhostNet_0_5.yaml
+74
-0
configs/GhostNet/GhostNet_1_0.yaml
configs/GhostNet/GhostNet_1_0.yaml
+74
-0
configs/GhostNet/GhostNet_1_3.yaml
configs/GhostNet/GhostNet_1_3.yaml
+75
-0
ppcls/modeling/architectures/ghostnet.py
ppcls/modeling/architectures/ghostnet.py
+321
-0
未找到文件。
configs/GhostNet/GhostNet_0_5.yaml
0 → 100644
浏览文件 @
5a1c4210
mode
:
'
train'
ARCHITECTURE
:
name
:
'
GhostNet_0_5'
pretrained_model
:
"
"
model_save_dir
:
"
./output/"
classes_num
:
1000
total_images
:
1281167
save_interval
:
1
validate
:
True
valid_interval
:
1
epochs
:
360
topk
:
5
image_shape
:
[
3
,
224
,
224
]
use_mix
:
False
ls_epsilon
:
0.1
LEARNING_RATE
:
function
:
'
CosineWarmup'
params
:
lr
:
0.8
OPTIMIZER
:
function
:
'
Momentum'
params
:
momentum
:
0.9
regularizer
:
function
:
'
L2'
factor
:
0.0000400
TRAIN
:
batch_size
:
2048
num_workers
:
4
file_list
:
"
./dataset/ILSVRC2012/train_list.txt"
data_dir
:
"
./dataset/ILSVRC2012/"
shuffle_seed
:
0
transforms
:
-
DecodeImage
:
to_rgb
:
True
to_np
:
False
channel_first
:
False
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
1./255.
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
-
ToCHWImage
:
VALID
:
batch_size
:
64
num_workers
:
4
file_list
:
"
./dataset/ILSVRC2012/val_list.txt"
data_dir
:
"
./dataset/ILSVRC2012/"
shuffle_seed
:
0
transforms
:
-
DecodeImage
:
to_rgb
:
True
to_np
:
False
channel_first
:
False
-
ResizeImage
:
resize_short
:
256
-
CropImage
:
size
:
224
-
NormalizeImage
:
scale
:
1.0/255.0
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
-
ToCHWImage
:
configs/GhostNet/GhostNet_1_0.yaml
0 → 100644
浏览文件 @
5a1c4210
mode
:
'
train'
ARCHITECTURE
:
name
:
'
GhostNet_1_0'
pretrained_model
:
"
"
model_save_dir
:
"
./output/"
classes_num
:
1000
total_images
:
1281167
save_interval
:
1
validate
:
True
valid_interval
:
1
epochs
:
360
topk
:
5
image_shape
:
[
3
,
224
,
224
]
use_mix
:
False
ls_epsilon
:
0.1
LEARNING_RATE
:
function
:
'
CosineWarmup'
params
:
lr
:
0.4
OPTIMIZER
:
function
:
'
Momentum'
params
:
momentum
:
0.9
regularizer
:
function
:
'
L2'
factor
:
0.0000400
TRAIN
:
batch_size
:
1024
num_workers
:
4
file_list
:
"
./dataset/ILSVRC2012/train_list.txt"
data_dir
:
"
./dataset/ILSVRC2012/"
shuffle_seed
:
0
transforms
:
-
DecodeImage
:
to_rgb
:
True
to_np
:
False
channel_first
:
False
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
1./255.
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
-
ToCHWImage
:
VALID
:
batch_size
:
64
num_workers
:
4
file_list
:
"
./dataset/ILSVRC2012/val_list.txt"
data_dir
:
"
./dataset/ILSVRC2012/"
shuffle_seed
:
0
transforms
:
-
DecodeImage
:
to_rgb
:
True
to_np
:
False
channel_first
:
False
-
ResizeImage
:
resize_short
:
256
-
CropImage
:
size
:
224
-
NormalizeImage
:
scale
:
1.0/255.0
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
-
ToCHWImage
:
configs/GhostNet/GhostNet_1_3.yaml
0 → 100644
浏览文件 @
5a1c4210
mode
:
'
train'
ARCHITECTURE
:
name
:
'
GhostNet_1_3'
pretrained_model
:
"
"
model_save_dir
:
"
./output/"
classes_num
:
1000
total_images
:
1281167
save_interval
:
1
validate
:
True
valid_interval
:
1
epochs
:
360
topk
:
5
image_shape
:
[
3
,
224
,
224
]
use_mix
:
False
ls_epsilon
:
0.1
LEARNING_RATE
:
function
:
'
CosineWarmup'
params
:
lr
:
0.4
OPTIMIZER
:
function
:
'
Momentum'
params
:
momentum
:
0.9
regularizer
:
function
:
'
L2'
factor
:
0.0000400
TRAIN
:
batch_size
:
1024
num_workers
:
4
file_list
:
"
./dataset/ILSVRC2012/train_list.txt"
data_dir
:
"
./dataset/ILSVRC2012/"
shuffle_seed
:
0
transforms
:
-
DecodeImage
:
to_rgb
:
True
to_np
:
False
channel_first
:
False
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
AutoAugment
:
-
NormalizeImage
:
scale
:
1./255.
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
-
ToCHWImage
:
VALID
:
batch_size
:
64
num_workers
:
4
file_list
:
"
./dataset/ILSVRC2012/val_list.txt"
data_dir
:
"
./dataset/ILSVRC2012/"
shuffle_seed
:
0
transforms
:
-
DecodeImage
:
to_rgb
:
True
to_np
:
False
channel_first
:
False
-
ResizeImage
:
resize_short
:
256
-
CropImage
:
size
:
224
-
NormalizeImage
:
scale
:
1.0/255.0
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
-
ToCHWImage
:
ppcls/modeling/architectures/ghostnet.py
0 → 100644
浏览文件 @
5a1c4210
from
__future__
import
absolute_import
from
__future__
import
division
from
__future__
import
print_function
import
math
import
paddle
import
paddle.fluid
as
fluid
from
paddle.fluid.param_attr
import
ParamAttr
from
paddle.fluid.initializer
import
MSRA
from
paddle.fluid.contrib.model_stat
import
summary
__all__
=
[
"GhostNet"
,
"GhostNet_0_5"
,
"GhostNet_1_0"
,
"GhostNet_1_3"
]
class
GhostNet
():
def
__init__
(
self
,
width_mult
):
cfgs
=
[
# k, t, c, SE, s
[
3
,
16
,
16
,
0
,
1
],
[
3
,
48
,
24
,
0
,
2
],
[
3
,
72
,
24
,
0
,
1
],
[
5
,
72
,
40
,
1
,
2
],
[
5
,
120
,
40
,
1
,
1
],
[
3
,
240
,
80
,
0
,
2
],
[
3
,
200
,
80
,
0
,
1
],
[
3
,
184
,
80
,
0
,
1
],
[
3
,
184
,
80
,
0
,
1
],
[
3
,
480
,
112
,
1
,
1
],
[
3
,
672
,
112
,
1
,
1
],
[
5
,
672
,
160
,
1
,
2
],
[
5
,
960
,
160
,
0
,
1
],
[
5
,
960
,
160
,
1
,
1
],
[
5
,
960
,
160
,
0
,
1
],
[
5
,
960
,
160
,
1
,
1
]
]
self
.
cfgs
=
cfgs
self
.
width_mult
=
width_mult
def
_make_divisible
(
self
,
v
,
divisor
,
min_value
=
None
):
"""
This function is taken from the original tf repo.
It ensures that all layers have a channel number that is divisible by 8
It can be seen here:
https://github.com/tensorflow/models/blob/master/research/slim/nets/mobilenet/mobilenet.py
"""
if
min_value
is
None
:
min_value
=
divisor
new_v
=
max
(
min_value
,
int
(
v
+
divisor
/
2
)
//
divisor
*
divisor
)
# Make sure that round down does not go down by more than 10%.
if
new_v
<
0.9
*
v
:
new_v
+=
divisor
return
new_v
def
conv_bn_layer
(
self
,
input
,
num_filters
,
filter_size
,
stride
=
1
,
groups
=
1
,
act
=
None
,
name
=
None
,
data_format
=
"NCHW"
):
x
=
fluid
.
layers
.
conv2d
(
input
=
input
,
num_filters
=
num_filters
,
filter_size
=
filter_size
,
stride
=
stride
,
padding
=
(
filter_size
-
1
)
//
2
,
groups
=
groups
,
act
=
None
,
param_attr
=
ParamAttr
(
initializer
=
fluid
.
initializer
.
MSRA
(),
name
=
name
+
"_weights"
),
bias_attr
=
False
,
name
=
name
+
"_conv_op"
,
data_format
=
data_format
)
x
=
fluid
.
layers
.
batch_norm
(
input
=
x
,
act
=
act
,
name
=
name
+
"_bn"
,
param_attr
=
ParamAttr
(
name
=
name
+
"_bn_scale"
,
regularizer
=
fluid
.
regularizer
.
L2DecayRegularizer
(
regularization_coeff
=
0.0
)),
bias_attr
=
ParamAttr
(
name
=
name
+
"_bn_offset"
,
regularizer
=
fluid
.
regularizer
.
L2DecayRegularizer
(
regularization_coeff
=
0.0
)),
moving_mean_name
=
name
+
"_bn_mean"
,
moving_variance_name
=
name
+
"_bn_variance"
,
data_layout
=
data_format
)
return
x
def
SElayer
(
self
,
input
,
num_channels
,
reduction_ratio
=
4
,
name
=
None
):
pool
=
fluid
.
layers
.
pool2d
(
input
=
input
,
pool_size
=
0
,
pool_type
=
'avg'
,
global_pooling
=
True
)
stdv
=
1.0
/
math
.
sqrt
(
pool
.
shape
[
1
]
*
1.0
)
squeeze
=
fluid
.
layers
.
fc
(
input
=
pool
,
size
=
num_channels
//
reduction_ratio
,
act
=
'relu'
,
param_attr
=
fluid
.
param_attr
.
ParamAttr
(
initializer
=
fluid
.
initializer
.
Uniform
(
-
stdv
,
stdv
),
name
=
name
+
'_sqz_weights'
),
bias_attr
=
ParamAttr
(
name
=
name
+
'_sqz_offset'
))
stdv
=
1.0
/
math
.
sqrt
(
squeeze
.
shape
[
1
]
*
1.0
)
excitation
=
fluid
.
layers
.
fc
(
input
=
squeeze
,
size
=
num_channels
,
act
=
None
,
param_attr
=
fluid
.
param_attr
.
ParamAttr
(
initializer
=
fluid
.
initializer
.
Uniform
(
-
stdv
,
stdv
),
name
=
name
+
'_exc_weights'
),
bias_attr
=
ParamAttr
(
name
=
name
+
'_exc_offset'
))
#ones = fluid.layers.fill_constant(excitation.shape, "float32", 1)
#zeros = fluid.layers.fill_constant(excitation.shape, "float32", 0)
#excitation = fluid.layers.elementwise_max(excitation, zeros)
# excitation = fluid.layers.elementwise_min(excitation, ones)
excitation
=
fluid
.
layers
.
clip
(
x
=
excitation
,
min
=
0
,
max
=
1
,
name
=
name
+
'_clip'
)
scale
=
fluid
.
layers
.
elementwise_mul
(
x
=
input
,
y
=
excitation
,
axis
=
0
)
return
scale
def
depthwise_conv
(
self
,
inp
,
oup
,
kernel_size
,
stride
=
1
,
relu
=
False
,
name
=
None
,
data_format
=
"NCHW"
):
return
self
.
conv_bn_layer
(
input
=
inp
,
num_filters
=
oup
,
filter_size
=
kernel_size
,
stride
=
stride
,
groups
=
inp
.
shape
[
1
]
if
data_format
==
"NCHW"
else
inp
.
shape
[
-
1
],
act
=
"relu"
if
relu
else
None
,
name
=
name
+
"_dw"
,
data_format
=
data_format
)
def
GhostModule
(
self
,
inp
,
oup
,
kernel_size
=
1
,
ratio
=
2
,
dw_size
=
3
,
stride
=
1
,
relu
=
True
,
name
=
None
,
data_format
=
"NCHW"
):
self
.
oup
=
oup
init_channels
=
int
(
math
.
ceil
(
oup
/
ratio
))
new_channels
=
int
(
init_channels
*
(
ratio
-
1
))
primary_conv
=
self
.
conv_bn_layer
(
input
=
inp
,
num_filters
=
init_channels
,
filter_size
=
kernel_size
,
stride
=
stride
,
groups
=
1
,
act
=
"relu"
if
relu
else
None
,
name
=
name
+
"_primary_conv"
,
data_format
=
"NCHW"
)
cheap_operation
=
self
.
conv_bn_layer
(
input
=
primary_conv
,
num_filters
=
new_channels
,
filter_size
=
dw_size
,
stride
=
1
,
groups
=
init_channels
,
act
=
"relu"
if
relu
else
None
,
name
=
name
+
"_cheap_operation"
,
data_format
=
data_format
)
out
=
fluid
.
layers
.
concat
([
primary_conv
,
cheap_operation
],
axis
=
1
,
name
=
name
+
"_concat"
)
# return out[:, :self.oup, :, :]
print
(
self
.
oup
)
print
(
out
.
shape
)
return
fluid
.
layers
.
slice
(
out
,
axes
=
[
1
],
starts
=
[
0
],
ends
=
[
self
.
oup
])
def
GhostBottleneck
(
self
,
inp
,
hidden_dim
,
oup
,
kernel_size
,
stride
,
use_se
,
name
=
None
,
data_format
=
"NCHW"
):
inp_channels
=
inp
.
shape
[
1
]
x
=
self
.
GhostModule
(
inp
=
inp
,
oup
=
hidden_dim
,
kernel_size
=
1
,
stride
=
1
,
relu
=
True
,
name
=
name
+
"GhostBottle_1"
,
data_format
=
"NCHW"
)
if
stride
==
2
:
x
=
self
.
depthwise_conv
(
inp
=
x
,
oup
=
hidden_dim
,
kernel_size
=
kernel_size
,
stride
=
stride
,
relu
=
False
,
name
=
name
+
"_dw2"
,
data_format
=
"NCHW"
)
if
use_se
:
x
=
self
.
SElayer
(
input
=
x
,
num_channels
=
hidden_dim
,
name
=
name
+
"SElayer"
)
x
=
self
.
GhostModule
(
inp
=
x
,
oup
=
oup
,
kernel_size
=
1
,
relu
=
False
,
name
=
name
+
"GhostModule_2"
)
if
stride
==
1
and
inp_channels
==
oup
:
shortcut
=
inp
else
:
shortcut
=
self
.
depthwise_conv
(
inp
=
inp
,
oup
=
inp_channels
,
kernel_size
=
kernel_size
,
stride
=
stride
,
relu
=
False
,
name
=
name
+
"shortcut_depthwise_conv"
,
data_format
=
"NCHW"
)
shortcut
=
self
.
conv_bn_layer
(
input
=
shortcut
,
num_filters
=
oup
,
filter_size
=
1
,
stride
=
1
,
groups
=
1
,
act
=
None
,
name
=
name
+
"shortcut_conv_bn"
,
data_format
=
"NCHW"
)
return
fluid
.
layers
.
elementwise_add
(
x
=
x
,
y
=
shortcut
,
axis
=-
1
,
act
=
None
,
name
=
name
+
"elementwise_add"
)
def
net
(
self
,
input
,
class_dim
=
1000
):
#build first layer:
output_channel
=
int
(
self
.
_make_divisible
(
16
*
self
.
width_mult
,
4
))
#print(output_channel)
x
=
self
.
conv_bn_layer
(
input
=
input
,
num_filters
=
output_channel
,
filter_size
=
3
,
stride
=
2
,
groups
=
1
,
act
=
"relu"
,
name
=
"firstlayer"
,
data_format
=
"NCHW"
)
input_channel
=
output_channel
#build inverted residual blocks
idx
=
0
fm
=
{}
for
k
,
exp_size
,
c
,
use_se
,
s
in
self
.
cfgs
:
output_channel
=
int
(
self
.
_make_divisible
(
c
*
self
.
width_mult
,
4
))
hidden_channel
=
int
(
self
.
_make_divisible
(
exp_size
*
self
.
width_mult
,
4
))
#print(output_channel)
#print(hidden_channel)
x
=
self
.
GhostBottleneck
(
inp
=
x
,
hidden_dim
=
hidden_channel
,
oup
=
output_channel
,
kernel_size
=
k
,
stride
=
s
,
use_se
=
use_se
,
name
=
"GhostBottle_"
+
str
(
idx
),
data_format
=
"NCHW"
)
input_channel
=
output_channel
fm
[
str
(
idx
)]
=
x
idx
+=
1
#build last several layers
output_channel
=
int
(
self
.
_make_divisible
(
exp_size
*
self
.
width_mult
,
4
))
x
=
self
.
conv_bn_layer
(
input
=
x
,
num_filters
=
output_channel
,
filter_size
=
1
,
stride
=
1
,
groups
=
1
,
act
=
"relu"
,
name
=
"lastlayer"
,
data_format
=
"NCHW"
)
x
=
fluid
.
layers
.
pool2d
(
input
=
x
,
pool_type
=
'avg'
,
global_pooling
=
True
,
data_format
=
"NCHW"
)
input_channel
=
output_channel
output_channel
=
1280
stdv
=
1.0
/
math
.
sqrt
(
x
.
shape
[
1
]
*
1.0
)
out
=
fluid
.
layers
.
conv2d
(
input
=
x
,
num_filters
=
output_channel
,
filter_size
=
1
,
groups
=
1
,
param_attr
=
ParamAttr
(
name
=
"fc_0_w"
,
initializer
=
fluid
.
initializer
.
Uniform
(
-
stdv
,
stdv
)),
bias_attr
=
False
,
name
=
"fc_0"
)
out
=
fluid
.
layers
.
batch_norm
(
input
=
out
,
act
=
"relu"
,
name
=
"fc_0_bn"
,
param_attr
=
ParamAttr
(
name
=
"fc_0_bn_scale"
,
regularizer
=
fluid
.
regularizer
.
L2DecayRegularizer
(
regularization_coeff
=
0.0
)),
bias_attr
=
ParamAttr
(
name
=
"fc_0_bn_offset"
,
regularizer
=
fluid
.
regularizer
.
L2DecayRegularizer
(
regularization_coeff
=
0.0
)),
moving_mean_name
=
"fc_0_bn_mean"
,
moving_variance_name
=
"fc_0_bn_variance"
,
data_layout
=
"NCHW"
)
out
=
fluid
.
layers
.
dropout
(
x
=
out
,
dropout_prob
=
0.2
)
stdv
=
1.0
/
math
.
sqrt
(
out
.
shape
[
1
]
*
1.0
)
out
=
fluid
.
layers
.
fc
(
input
=
out
,
size
=
class_dim
,
param_attr
=
ParamAttr
(
name
=
"fc_1_w"
,
initializer
=
fluid
.
initializer
.
Uniform
(
-
stdv
,
stdv
)),
bias_attr
=
ParamAttr
(
name
=
"fc_1_bias"
))
return
out
,
fm
def
GhostNet_0_5
():
model
=
GhostNet
(
width_mult
=
0.5
)
return
model
def
GhostNet_1_0
():
model
=
GhostNet
(
width_mult
=
1.0
)
return
model
def
GhostNet_1_3
():
model
=
GhostNet
(
width_mult
=
1.3
)
return
model
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