Skip to content
GitLab
Menu
Projects
Groups
Snippets
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in / Register
Toggle navigation
Menu
Open sidebar
dcuai
dlexamples
Commits
555464ee
Commit
555464ee
authored
Jan 14, 2023
by
unknown
Browse files
修改mmseg train.md
parent
a024b5d7
Changes
2
Hide whitespace changes
Inline
Side-by-side
Showing
2 changed files
with
33 additions
and
84 deletions
+33
-84
openmmlab_test/mmdetection-2.25.2/configs/dcn/faster_rcnn_r50_fpn_mdpool_1x_coco.py
...-2.25.2/configs/dcn/faster_rcnn_r50_fpn_mdpool_1x_coco.py
+12
-0
openmmlab_test/mmsegmentation-0.29.1/train.md
openmmlab_test/mmsegmentation-0.29.1/train.md
+21
-84
No files found.
openmmlab_test/mmdetection-2.25.2/configs/dcn/faster_rcnn_r50_fpn_mdpool_1x_coco.py
0 → 100644
View file @
555464ee
_base_
=
'../faster_rcnn/faster_rcnn_r50_fpn_1x_coco.py'
model
=
dict
(
roi_head
=
dict
(
bbox_roi_extractor
=
dict
(
type
=
'SingleRoIExtractor'
,
roi_layer
=
dict
(
_delete_
=
True
,
type
=
'ModulatedDeformRoIPoolPack'
,
output_size
=
7
,
output_channels
=
256
),
out_channels
=
256
,
featmap_strides
=
[
4
,
8
,
16
,
32
])))
openmmlab_test/mmsegmentation-0.29.1/train.md
View file @
555464ee
...
...
@@ -23,15 +23,15 @@ pip3 install opencv-python
### 安装python依赖包
```
python
pip3
install
torch
-
1.10
.
0
a0
+
git
cc7c9c7
-
cp3
6
-
cp3
6
m
-
linux_x86_64
.
whl
-
i
https
:
//
pypi
.
tuna
.
tsinghua
.
edu
.
cn
/
simple
pip3
install
torchvision
-
0.10
.
0
a0
+
300
a8a4
-
cp3
6
-
cp3
6
m
-
linux_x86_64
.
whl
-
i
https
:
//
pypi
.
tuna
.
tsinghua
.
edu
.
cn
/
simple
pip3
install
mmcv_full
-
1.
3
.
1
6
-
cp3
6
-
cp3
6
m
-
linux_x86_64
.
whl
-
i
https
:
//
pypi
.
tuna
.
tsinghua
.
edu
.
cn
/
simple
mm
po
se安装
:
cd
mmsegmentation
pip3
install
torch
-
1.10
.
0
a0
+
git
2040069
.
dtk2210
-
cp3
7
-
cp3
7
m
-
many
linux
2014
_x86_64
.
whl
-
i
https
:
//
pypi
.
tuna
.
tsinghua
.
edu
.
cn
/
simple
pip3
install
torchvision
-
0.10
.
0
a0
+
e04d001
.
dtk2210
-
cp3
7
-
cp3
7
m
-
many
linux
2014
_x86_64
.
whl
-
i
https
:
//
pypi
.
tuna
.
tsinghua
.
edu
.
cn
/
simple
pip3
install
mmcv_full
-
1.
6
.
1
+
gitdebbc80
.
dtk2210
-
cp3
7
-
cp3
7
m
-
many
linux
2014
_x86_64
.
whl
-
i
https
:
//
pypi
.
tuna
.
tsinghua
.
edu
.
cn
/
simple
mmse
g
安装
:
cd
mmsegmentation
-
0.29
.
1
pip3
install
-
e
.
```
注:测试不同版本的dtk,需安装对应版本的库whl包,
dtk22.04.1使用python3.7,dtk21.10.1使用python3.6,
如果测试优化后的版本,需要设置export HIP_UPSAMPLE_OPTIMIZE=1
注:测试不同版本的dtk,需安装对应版本的库whl包,如果测试优化后的版本,需要设置export HIP_UPSAMPLE_OPTIMIZE=1
## PSPNet R50测试
...
...
@@ -41,101 +41,38 @@ pip3 install -e .
.
/
sing_test
.
sh
configs
/
pspnet
/
pspnet_r50
-
d8_512x1024_40k_cityscapes
.
py
```
若使用dtk22.04.1测试,需设置
`export MIOPEN_FIND_MODE=1,export MIOPEN_USE_APPROXIMATE_PERFORMANCE=0`
以下测试均相同.
#### 参数说明
configs/_base_/datasets/cityscapes.py中batch_size=samples_per_gpu
*
卡数,性能计算方法:batch_size/time

#### 性能关注:time
### 多卡测试(单精度)
```
python
.
/
multi_test
.
sh
configs
/
pspnet
/
pspnet_r50
-
d8_512x1024_40k_cityscapes
.
py
```
## DeepLabV3 R50测试
#### 单机多卡训练
### 单卡测试(单精度)
1.
pytorch单机多卡训练
```
python
.
/
sing_test
.
sh
configs
/
deeplabv3
/
deeplabv3_r50
-
d8_512x1024_40k_cityscapes
.
py
```
#### 参数说明
configs/_base_/datasets/cityscapes.py中batch_size=samples_per_gpu
*
卡数,性能计算方法:batch_size/time

#### 性能关注:time
### 多卡测试(单精度)
```
python
.
/
multi_test
.
sh
configs
/
deeplabv3
/
deeplabv3_r50
-
d8_512x1024_40k_cityscapes
.
py
```
## FCN R50测试
### 单卡测试(单精度)
```
python
.
/
sing_test
.
sh
configs
/
fcn
/
fcn_r50
-
d8_512x1024_40k_cityscapes
.
py
```
#### 参数说明
configs/_base_/datasets/cityscapes.py中batch_size=samples_per_gpu
*
卡数,性能计算方法:batch_size/time

#### 性能关注:time
### 多卡测试(单精度)
```
python
.
/
multi_test
.
sh
configs
/
fcn
/
fcn_r50
-
d8_512x1024_40k_cityscapes
.
py
```
## UperNet R50测试
### 单卡测试(单精度)
```
python
.
/
sing_test
.
sh
configs
/
upernet
/
upernet_r50_512x1024_40k_cityscapes
.
py
```
#### 参数说明
configs/_base_/datasets/cityscapes.py中batch_size=samples_per_gpu
*
卡数,性能计算方法:batch_size/time

#### 性能关注:time
### 多卡测试(单精度)
```
python
.
/
multi_test
.
sh
configs
/
upernet
/
upernet_r50_512x1024_40k_cityscapes
.
py
.
/
multi_test
.
sh
configs
/
pspnet
/
pspnet_r50
-
d8_512x1024_40k_cityscapes
.
py
```
##
DeepLabV3plus_R50测试
##
##
多机多卡训练
### 单卡测试(单精度)
1.
pytorch多机多卡训练 在第一台机器上: NODES=2 NODE_RANK=0 PORT=12345 MASTER_ADDR=10.1.3.56 sh tools/dist_train.sh configs/pspnet/pspnet_r50-d8_512x1024_40k_cityscapes.py 4 在第二台机器上: NODES=2 NODE_RANK=1 PORT=12345 MASTER_ADDR=10.1.3.56 sh tools/dist_train.sh configs/pspnet/pspnet_r50-d8_512x1024_40k_cityscapes.py 4
```
python
.
/
sing_test
.
sh
configs
/
deeplabv3plus
/
deeplabv3plus_r50
-
d8_512x1024_40k_cityscapes
.
py
```
### 半精度测试
#### 参数说明
修改configs文件,添加fp16 = dict(loss_scale=512.),单机多卡和多机多卡测试与单精度测试方法相同。
configs/_base_/datasets/cityscapes.py中batch_size=samples_per_gpu
*
卡数,性能计算方法:batch_size/time

### 其他模型测试
#### 性能关注:time
其他模型的测试步骤和pspnet_r50相同,只需修改对应的config文件即可,下面列出相关模型对应的config文件列表:
### 多卡测试(单精度)
| 模型 | configs |
| ----------------- | ------------------------------------------------------------ |
| DeepLabV3 R50 | configs/deeplabv3/deeplabv3_r50-d8_512x1024_40k_cityscapes.py |
| FCN R50 | configs/fcn/fcn_r50-d8_512x1024_40k_cityscapes.py |
| UperNet R50 | configs/upernet/upernet_r50_512x1024_40k_cityscapes.py |
| DeepLabV3plus_R50 | configs/deeplabv3plus/deeplabv3plus_r50-d8_512x1024_40k_cityscapes.py |
```
python
.
/
multi_test
.
sh
configs
/
deeplabv3plus
/
deeplabv3plus_r50
-
d8_512x1024_40k_cityscapes
.
py
```
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
.
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment