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
49bf3d3f
P
PaddleClas
项目概览
PaddlePaddle
/
PaddleClas
大约 1 年 前同步成功
通知
115
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
49bf3d3f
编写于
9月 11, 2020
作者:
littletomatodonkey
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
add resnest metrics
上级
911611d9
变更
4
隐藏空白更改
内联
并排
Showing
4 changed file
with
12 addition
and
19 deletion
+12
-19
README.md
README.md
+2
-2
docs/en/models/ResNeSt_RegNet_en.md
docs/en/models/ResNeSt_RegNet_en.md
+2
-2
docs/zh_CN/models/ResNeSt_RegNet.md
docs/zh_CN/models/ResNeSt_RegNet.md
+2
-2
ppcls/modeling/architectures/resnest.py
ppcls/modeling/architectures/resnest.py
+6
-13
未找到文件。
README.md
浏览文件 @
49bf3d3f
...
@@ -286,8 +286,8 @@ ResNeSt与RegNet系列模型的精度、速度指标如下表所示,更多关
...
@@ -286,8 +286,8 @@ ResNeSt与RegNet系列模型的精度、速度指标如下表所示,更多关
| 模型 | Top-1 Acc | Top-5 Acc | time(ms)
<br>
bs=1 | time(ms)
<br>
bs=4 | Flops(G) | Params(M) | 下载地址 |
| 模型 | Top-1 Acc | Top-5 Acc | time(ms)
<br>
bs=1 | time(ms)
<br>
bs=4 | Flops(G) | Params(M) | 下载地址 |
|------------------------|-----------|-----------|------------------|------------------|----------|-----------|------------------------------------------------------------------------------------------------------|
|------------------------|-----------|-----------|------------------|------------------|----------|-----------|------------------------------------------------------------------------------------------------------|
| ResNeSt50_
<br>
fast_1s1x64d | 0.8035 | 0.9528 |
- | -
| 8.68 | 26.3 |
[
下载链接
](
https://paddle-imagenet-models-name.bj.bcebos.com/ResNeSt50_fast_1s1x64d_pretrained.pdparams
)
|
| ResNeSt50_
<br>
fast_1s1x64d | 0.8035 | 0.9528 |
3.45405 | 8.72680
| 8.68 | 26.3 |
[
下载链接
](
https://paddle-imagenet-models-name.bj.bcebos.com/ResNeSt50_fast_1s1x64d_pretrained.pdparams
)
|
| ResNeSt50 | 0.8102 | 0.9542 |
- | -
| 10.78 | 27.5 |
[
下载链接
](
https://paddle-imagenet-models-name.bj.bcebos.com/ResNeSt50_pretrained.pdparams
)
|
| ResNeSt50 | 0.8102 | 0.9542 |
6.69042 | 8.01664
| 10.78 | 27.5 |
[
下载链接
](
https://paddle-imagenet-models-name.bj.bcebos.com/ResNeSt50_pretrained.pdparams
)
|
| RegNetX_4GF | 0.785 | 0.9416 | 6.46478 | 11.19862 | 8 | 22.1 |
[
下载链接
](
https://paddle-imagenet-models-name.bj.bcebos.com/RegNetX_4GF_pretrained.pdparams
)
|
| RegNetX_4GF | 0.785 | 0.9416 | 6.46478 | 11.19862 | 8 | 22.1 |
[
下载链接
](
https://paddle-imagenet-models-name.bj.bcebos.com/RegNetX_4GF_pretrained.pdparams
)
|
...
...
docs/en/models/ResNeSt_RegNet_en.md
浏览文件 @
49bf3d3f
...
@@ -17,6 +17,6 @@ RegNet was proposed in 2020 by Facebook to deepen the concept of design space. B
...
@@ -17,6 +17,6 @@ RegNet was proposed in 2020 by Facebook to deepen the concept of design space. B
| Models | Crop Size | Resize Short Size | FP16
<br>
Batch Size=1
<br>
(ms) | FP16
<br>
Batch Size=4
<br>
(ms) | FP16
<br>
Batch Size=8
<br>
(ms) | FP32
<br>
Batch Size=1
<br>
(ms) | FP32
<br>
Batch Size=4
<br>
(ms) | FP32
<br>
Batch Size=8
<br>
(ms) |
| Models | Crop Size | Resize Short Size | FP16
<br>
Batch Size=1
<br>
(ms) | FP16
<br>
Batch Size=4
<br>
(ms) | FP16
<br>
Batch Size=8
<br>
(ms) | FP32
<br>
Batch Size=1
<br>
(ms) | FP32
<br>
Batch Size=4
<br>
(ms) | FP32
<br>
Batch Size=8
<br>
(ms) |
|--------------------|-----------|-------------------|------------------------------|------------------------------|------------------------------|------------------------------|------------------------------|------------------------------|
|--------------------|-----------|-------------------|------------------------------|------------------------------|------------------------------|------------------------------|------------------------------|------------------------------|
| ResNeSt50_fast_1s1x64d | 224 | 256 |
- | - | - | - | - | -
|
| ResNeSt50_fast_1s1x64d | 224 | 256 |
3.46466 | 5.56647 | 9.11848 | 3.45405 | 8.72680 | 15.48710
|
| ResNeSt50 | 224 | 256 |
- | - | - | - | - | -
|
| ResNeSt50 | 224 | 256 |
7.05851 | 8.97676 | 13.34704 | 6.16248 | 12.0633 | 21.49936
|
| RegNetX_4GF | 224 | 256 | 6.69042 | 8.01664 | 11.60608 | 6.46478 | 11.19862 | 16.89089 |
| RegNetX_4GF | 224 | 256 | 6.69042 | 8.01664 | 11.60608 | 6.46478 | 11.19862 | 16.89089 |
docs/zh_CN/models/ResNeSt_RegNet.md
浏览文件 @
49bf3d3f
...
@@ -20,6 +20,6 @@ RegNet是由facebook于2020年提出,旨在深化设计空间理念的概念
...
@@ -20,6 +20,6 @@ RegNet是由facebook于2020年提出,旨在深化设计空间理念的概念
| Models | Crop Size | Resize Short Size | FP16
<br>
Batch Size=1
<br>
(ms) | FP16
<br>
Batch Size=4
<br>
(ms) | FP16
<br>
Batch Size=8
<br>
(ms) | FP32
<br>
Batch Size=1
<br>
(ms) | FP32
<br>
Batch Size=4
<br>
(ms) | FP32
<br>
Batch Size=8
<br>
(ms) |
| Models | Crop Size | Resize Short Size | FP16
<br>
Batch Size=1
<br>
(ms) | FP16
<br>
Batch Size=4
<br>
(ms) | FP16
<br>
Batch Size=8
<br>
(ms) | FP32
<br>
Batch Size=1
<br>
(ms) | FP32
<br>
Batch Size=4
<br>
(ms) | FP32
<br>
Batch Size=8
<br>
(ms) |
|--------------------|-----------|-------------------|------------------------------|------------------------------|------------------------------|------------------------------|------------------------------|------------------------------|
|--------------------|-----------|-------------------|------------------------------|------------------------------|------------------------------|------------------------------|------------------------------|------------------------------|
| ResNeSt50_fast_1s1x64d | 224 | 256 |
- | - | - | - | - | -
|
| ResNeSt50_fast_1s1x64d | 224 | 256 |
3.46466 | 5.56647 | 9.11848 | 3.45405 | 8.72680 | 15.48710
|
| ResNeSt50 | 224 | 256 |
- | - | - | - | - | -
|
| ResNeSt50 | 224 | 256 |
7.05851 | 8.97676 | 13.34704 | 6.16248 | 12.0633 | 21.49936
|
| RegNetX_4GF | 224 | 256 | 6.69042 | 8.01664 | 11.60608 | 6.46478 | 11.19862 | 16.89089 |
| RegNetX_4GF | 224 | 256 | 6.69042 | 8.01664 | 11.60608 | 6.46478 | 11.19862 | 16.89089 |
ppcls/modeling/architectures/resnest.py
浏览文件 @
49bf3d3f
...
@@ -104,11 +104,7 @@ class ResNeSt():
...
@@ -104,11 +104,7 @@ class ResNeSt():
is_first
=
False
,
is_first
=
False
,
name
=
"layer1"
)
name
=
"layer1"
)
x
=
self
.
resnest_layer
(
x
=
self
.
resnest_layer
(
x
=
x
,
x
=
x
,
planes
=
128
,
blocks
=
self
.
layers
[
1
],
stride
=
2
,
name
=
"layer2"
)
planes
=
128
,
blocks
=
self
.
layers
[
1
],
stride
=
2
,
name
=
"layer2"
)
if
self
.
dilated
or
self
.
dilation
==
4
:
if
self
.
dilated
or
self
.
dilation
==
4
:
x
=
self
.
resnest_layer
(
x
=
self
.
resnest_layer
(
x
=
x
,
x
=
x
,
...
@@ -152,10 +148,8 @@ class ResNeSt():
...
@@ -152,10 +148,8 @@ class ResNeSt():
blocks
=
self
.
layers
[
3
],
blocks
=
self
.
layers
[
3
],
stride
=
2
,
stride
=
2
,
name
=
"layer4"
)
name
=
"layer4"
)
x
=
fluid
.
layers
.
pool2d
(
x
=
fluid
.
layers
.
pool2d
(
input
=
x
,
pool_type
=
"avg"
,
global_pooling
=
True
)
input
=
x
,
pool_type
=
"avg"
,
global_pooling
=
True
)
x
=
fluid
.
layers
.
dropout
(
x
=
x
,
dropout_prob
=
self
.
final_drop
)
x
=
fluid
.
layers
.
dropout
(
x
=
x
,
dropout_prob
=
self
.
final_drop
)
stdv
=
1.0
/
math
.
sqrt
(
x
.
shape
[
1
]
*
1.0
)
stdv
=
1.0
/
math
.
sqrt
(
x
.
shape
[
1
]
*
1.0
)
x
=
fluid
.
layers
.
fc
(
x
=
fluid
.
layers
.
fc
(
input
=
x
,
input
=
x
,
...
@@ -266,8 +260,7 @@ class ResNeSt():
...
@@ -266,8 +260,7 @@ class ResNeSt():
param_attr
=
ParamAttr
(
param_attr
=
ParamAttr
(
name
=
name
+
"_splat_weights"
,
initializer
=
MSRA
()),
name
=
name
+
"_splat_weights"
,
initializer
=
MSRA
()),
bias_attr
=
False
)
bias_attr
=
False
)
atten
=
self
.
rsoftmax
(
atten
=
self
.
rsoftmax
(
x
=
atten
,
radix
=
radix
,
cardinality
=
groups
)
x
=
atten
,
radix
=
radix
,
cardinality
=
groups
)
atten
=
fluid
.
layers
.
reshape
(
x
=
atten
,
shape
=
[
-
1
,
atten
.
shape
[
1
],
1
,
1
])
atten
=
fluid
.
layers
.
reshape
(
x
=
atten
,
shape
=
[
-
1
,
atten
.
shape
[
1
],
1
,
1
])
if
radix
>
1
:
if
radix
>
1
:
...
@@ -275,10 +268,10 @@ class ResNeSt():
...
@@ -275,10 +268,10 @@ class ResNeSt():
input
=
atten
,
num_or_sections
=
radix
,
dim
=
1
)
input
=
atten
,
num_or_sections
=
radix
,
dim
=
1
)
out
=
fluid
.
layers
.
sum
([
out
=
fluid
.
layers
.
sum
([
fluid
.
layers
.
elementwise_mul
(
fluid
.
layers
.
elementwise_mul
(
x
=
att
,
y
=
spli
t
)
for
(
att
,
split
)
in
zip
(
attens
,
splited
)
x
=
split
,
y
=
at
t
)
for
(
att
,
split
)
in
zip
(
attens
,
splited
)
])
])
else
:
else
:
out
=
fluid
.
layers
.
elementwise_mul
(
atten
,
x
)
out
=
fluid
.
layers
.
elementwise_mul
(
x
,
atten
)
return
out
return
out
def
bottleneck
(
self
,
def
bottleneck
(
self
,
...
...
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