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
d58fd3b7
P
PaddleClas
项目概览
PaddlePaddle
/
PaddleClas
1 年多 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
未验证
提交
d58fd3b7
编写于
6月 02, 2021
作者:
W
Wei Shengyu
提交者:
GitHub
6月 02, 2021
浏览文件
操作
浏览文件
下载
差异文件
Merge pull request #765 from cuicheng01/develop_reg
Add legendary_model's congigs and fix some trainer's bug
上级
aecbf40b
d13cb223
变更
45
隐藏空白更改
内联
并排
Showing
45 changed file
with
4485 addition
and
64 deletion
+4485
-64
ppcls/arch/backbone/__init__.py
ppcls/arch/backbone/__init__.py
+7
-7
ppcls/configs/ImageNet/HRNet/HRNet_W18_C.yaml
ppcls/configs/ImageNet/HRNet/HRNet_W18_C.yaml
+124
-0
ppcls/configs/ImageNet/HRNet/HRNet_W30_C.yaml
ppcls/configs/ImageNet/HRNet/HRNet_W30_C.yaml
+124
-0
ppcls/configs/ImageNet/HRNet/HRNet_W32_C.yaml
ppcls/configs/ImageNet/HRNet/HRNet_W32_C.yaml
+124
-0
ppcls/configs/ImageNet/HRNet/HRNet_W40_C.yaml
ppcls/configs/ImageNet/HRNet/HRNet_W40_C.yaml
+124
-0
ppcls/configs/ImageNet/HRNet/HRNet_W44_C.yaml
ppcls/configs/ImageNet/HRNet/HRNet_W44_C.yaml
+124
-0
ppcls/configs/ImageNet/HRNet/HRNet_W48_C.yaml
ppcls/configs/ImageNet/HRNet/HRNet_W48_C.yaml
+124
-0
ppcls/configs/ImageNet/HRNet/HRNet_W64_C.yaml
ppcls/configs/ImageNet/HRNet/HRNet_W64_C.yaml
+124
-0
ppcls/configs/ImageNet/Ineption/InceptionV3.yaml
ppcls/configs/ImageNet/Ineption/InceptionV3.yaml
+124
-0
ppcls/configs/ImageNet/MobileNetV1/MobileNetV1.yaml
ppcls/configs/ImageNet/MobileNetV1/MobileNetV1.yaml
+124
-0
ppcls/configs/ImageNet/MobileNetV1/MobileNetV1_x0_25.yaml
ppcls/configs/ImageNet/MobileNetV1/MobileNetV1_x0_25.yaml
+124
-0
ppcls/configs/ImageNet/MobileNetV1/MobileNetV1_x0_5.yaml
ppcls/configs/ImageNet/MobileNetV1/MobileNetV1_x0_5.yaml
+124
-0
ppcls/configs/ImageNet/MobileNetV1/MobileNetV1_x0_75.yaml
ppcls/configs/ImageNet/MobileNetV1/MobileNetV1_x0_75.yaml
+124
-0
ppcls/configs/ImageNet/MobileNetV3/MobileNetV3_large_x0_35.yaml
...configs/ImageNet/MobileNetV3/MobileNetV3_large_x0_35.yaml
+122
-0
ppcls/configs/ImageNet/MobileNetV3/MobileNetV3_large_x0_5.yaml
.../configs/ImageNet/MobileNetV3/MobileNetV3_large_x0_5.yaml
+122
-0
ppcls/configs/ImageNet/MobileNetV3/MobileNetV3_large_x0_75.yaml
...configs/ImageNet/MobileNetV3/MobileNetV3_large_x0_75.yaml
+122
-0
ppcls/configs/ImageNet/MobileNetV3/MobileNetV3_large_x1_0.yaml
.../configs/ImageNet/MobileNetV3/MobileNetV3_large_x1_0.yaml
+122
-0
ppcls/configs/ImageNet/MobileNetV3/MobileNetV3_large_x1_25.yaml
...configs/ImageNet/MobileNetV3/MobileNetV3_large_x1_25.yaml
+122
-0
ppcls/configs/ImageNet/MobileNetV3/MobileNetV3_small_x0_35.yaml
...configs/ImageNet/MobileNetV3/MobileNetV3_small_x0_35.yaml
+122
-0
ppcls/configs/ImageNet/MobileNetV3/MobileNetV3_small_x0_5.yaml
.../configs/ImageNet/MobileNetV3/MobileNetV3_small_x0_5.yaml
+122
-0
ppcls/configs/ImageNet/MobileNetV3/MobileNetV3_small_x0_75.yaml
...configs/ImageNet/MobileNetV3/MobileNetV3_small_x0_75.yaml
+122
-0
ppcls/configs/ImageNet/MobileNetV3/MobileNetV3_small_x1_0.yaml
.../configs/ImageNet/MobileNetV3/MobileNetV3_small_x1_0.yaml
+122
-0
ppcls/configs/ImageNet/MobileNetV3/MobileNetV3_small_x1_25.yaml
...configs/ImageNet/MobileNetV3/MobileNetV3_small_x1_25.yaml
+122
-0
ppcls/configs/ImageNet/ResNet/ResNet101.yaml
ppcls/configs/ImageNet/ResNet/ResNet101.yaml
+124
-0
ppcls/configs/ImageNet/ResNet/ResNet101_vd.yaml
ppcls/configs/ImageNet/ResNet/ResNet101_vd.yaml
+122
-0
ppcls/configs/ImageNet/ResNet/ResNet152.yaml
ppcls/configs/ImageNet/ResNet/ResNet152.yaml
+124
-0
ppcls/configs/ImageNet/ResNet/ResNet152_vd.yaml
ppcls/configs/ImageNet/ResNet/ResNet152_vd.yaml
+122
-0
ppcls/configs/ImageNet/ResNet/ResNet18.yaml
ppcls/configs/ImageNet/ResNet/ResNet18.yaml
+124
-0
ppcls/configs/ImageNet/ResNet/ResNet18_vd.yaml
ppcls/configs/ImageNet/ResNet/ResNet18_vd.yaml
+122
-0
ppcls/configs/ImageNet/ResNet/ResNet200_vd.yaml
ppcls/configs/ImageNet/ResNet/ResNet200_vd.yaml
+122
-0
ppcls/configs/ImageNet/ResNet/ResNet34.yaml
ppcls/configs/ImageNet/ResNet/ResNet34.yaml
+124
-0
ppcls/configs/ImageNet/ResNet/ResNet34_vd.yaml
ppcls/configs/ImageNet/ResNet/ResNet34_vd.yaml
+122
-0
ppcls/configs/ImageNet/ResNet/ResNet50.yaml
ppcls/configs/ImageNet/ResNet/ResNet50.yaml
+44
-44
ppcls/configs/ImageNet/ResNet/ResNet50_vd.yaml
ppcls/configs/ImageNet/ResNet/ResNet50_vd.yaml
+122
-0
ppcls/configs/ImageNet/VGG/VGG11.yaml
ppcls/configs/ImageNet/VGG/VGG11.yaml
+122
-0
ppcls/configs/ImageNet/VGG/VGG13.yaml
ppcls/configs/ImageNet/VGG/VGG13.yaml
+122
-0
ppcls/configs/ImageNet/VGG/VGG16.yaml
ppcls/configs/ImageNet/VGG/VGG16.yaml
+122
-0
ppcls/configs/ImageNet/VGG/VGG19.yaml
ppcls/configs/ImageNet/VGG/VGG19.yaml
+122
-0
ppcls/data/__init__.py
ppcls/data/__init__.py
+5
-4
ppcls/data/dataset/common_dataset.py
ppcls/data/dataset/common_dataset.py
+0
-1
ppcls/data/postprocess/__init__.py
ppcls/data/postprocess/__init__.py
+2
-2
ppcls/data/preprocess/__init__.py
ppcls/data/preprocess/__init__.py
+0
-1
ppcls/engine/trainer.py
ppcls/engine/trainer.py
+2
-2
ppcls/losses/triplet.py
ppcls/losses/triplet.py
+0
-2
tools/run.sh
tools/run.sh
+1
-1
未找到文件。
ppcls/arch/backbone/__init__.py
浏览文件 @
d58fd3b7
...
...
@@ -12,9 +12,14 @@
# See the License for the specific language governing permissions and
# limitations under the License.
from
ppcls.arch.backbone.model_zoo.resnet
import
ResNet18
,
ResNet34
,
ResNet50
,
ResNet101
,
ResNet152
from
ppcls.arch.backbone.legendary_models.mobilenet_v1
import
MobileNetV1_x0_25
,
MobileNetV1_x0_5
,
MobileNetV1_x0_75
,
MobileNetV1
from
ppcls.arch.backbone.legendary_models.mobilenet_v3
import
MobileNetV3_small_x0_35
,
MobileNetV3_small_x0_5
,
MobileNetV3_small_x0_75
,
MobileNetV3_small_x1_0
,
MobileNetV3_small_x1_25
,
MobileNetV3_large_x0_35
,
MobileNetV3_large_x0_5
,
MobileNetV3_large_x0_75
,
MobileNetV3_large_x1_0
,
MobileNetV3_large_x1_25
from
ppcls.arch.backbone.legendary_models.resnet
import
ResNet18
,
ResNet18_vd
,
ResNet34
,
ResNet34_vd
,
ResNet50
,
ResNet50_vd
,
ResNet101
,
ResNet101_vd
,
ResNet152
,
ResNet152_vd
,
ResNet200_vd
from
ppcls.arch.backbone.legendary_models.vgg
import
VGG11
,
VGG13
,
VGG16
,
VGG19
from
ppcls.arch.backbone.legendary_models.inception_v3
import
InceptionV3
from
ppcls.arch.backbone.legendary_models.hrnet
import
HRNet_W18_C
,
HRNet_W30_C
,
HRNet_W32_C
,
HRNet_W40_C
,
HRNet_W44_C
,
HRNet_W48_C
,
HRNet_W60_C
,
HRNet_W64_C
,
SE_HRNet_W64_C
from
ppcls.arch.backbone.model_zoo.resnet_vc
import
ResNet18_vc
,
ResNet34_vc
,
ResNet50_vc
,
ResNet101_vc
,
ResNet152_vc
from
ppcls.arch.backbone.model_zoo.resnet_vd
import
ResNet18_vd
,
ResNet34_vd
,
ResNet50_vd
,
ResNet101_vd
,
ResNet152_vd
,
ResNet200_vd
from
ppcls.arch.backbone.model_zoo.resnext
import
ResNeXt50_32x4d
,
ResNeXt50_64x4d
,
ResNeXt101_32x4d
,
ResNeXt101_64x4d
,
ResNeXt152_32x4d
,
ResNeXt152_64x4d
from
ppcls.arch.backbone.model_zoo.res2net
import
Res2Net50_48w_2s
,
Res2Net50_26w_4s
,
Res2Net50_14w_8s
,
Res2Net50_48w_2s
,
Res2Net50_26w_6s
,
Res2Net50_26w_8s
,
Res2Net101_26w_4s
,
Res2Net152_26w_4s
,
Res2Net200_26w_4s
from
ppcls.arch.backbone.model_zoo.res2net_vd
import
Res2Net50_vd_48w_2s
,
Res2Net50_vd_26w_4s
,
Res2Net50_vd_14w_8s
,
Res2Net50_vd_48w_2s
,
Res2Net50_vd_26w_6s
,
Res2Net50_vd_26w_8s
,
Res2Net101_vd_26w_4s
,
Res2Net152_vd_26w_4s
,
Res2Net200_vd_26w_4s
...
...
@@ -23,22 +28,17 @@ from ppcls.arch.backbone.model_zoo.se_resnext_vd import SE_ResNeXt50_vd_32x4d, S
from
ppcls.arch.backbone.model_zoo.se_resnext
import
SE_ResNeXt50_32x4d
,
SE_ResNeXt101_32x4d
,
SE_ResNeXt152_64x4d
from
ppcls.arch.backbone.model_zoo.dpn
import
DPN68
,
DPN92
,
DPN98
,
DPN107
,
DPN131
from
ppcls.arch.backbone.model_zoo.densenet
import
DenseNet121
,
DenseNet161
,
DenseNet169
,
DenseNet201
,
DenseNet264
from
ppcls.arch.backbone.model_zoo.hrnet
import
HRNet_W18_C
,
HRNet_W30_C
,
HRNet_W32_C
,
HRNet_W40_C
,
HRNet_W44_C
,
HRNet_W48_C
,
HRNet_W60_C
,
HRNet_W64_C
,
SE_HRNet_W18_C
,
SE_HRNet_W30_C
,
SE_HRNet_W32_C
,
SE_HRNet_W40_C
,
SE_HRNet_W44_C
,
SE_HRNet_W48_C
,
SE_HRNet_W60_C
,
SE_HRNet_W64_C
from
ppcls.arch.backbone.model_zoo.efficientnet
import
EfficientNetB0
,
EfficientNetB1
,
EfficientNetB2
,
EfficientNetB3
,
EfficientNetB4
,
EfficientNetB5
,
EfficientNetB6
,
EfficientNetB7
,
EfficientNetB0_small
from
ppcls.arch.backbone.model_zoo.resnest
import
ResNeSt50_fast_1s1x64d
,
ResNeSt50
,
ResNeSt101
from
ppcls.arch.backbone.model_zoo.googlenet
import
GoogLeNet
from
ppcls.arch.backbone.model_zoo.mobilenet_v1
import
MobileNetV1_x0_25
,
MobileNetV1_x0_5
,
MobileNetV1_x0_75
,
MobileNetV1
from
ppcls.arch.backbone.model_zoo.mobilenet_v2
import
MobileNetV2_x0_25
,
MobileNetV2_x0_5
,
MobileNetV2_x0_75
,
MobileNetV2
,
MobileNetV2_x1_5
,
MobileNetV2_x2_0
from
ppcls.arch.backbone.model_zoo.mobilenet_v3
import
MobileNetV3_small_x0_35
,
MobileNetV3_small_x0_5
,
MobileNetV3_small_x0_75
,
MobileNetV3_small_x1_0
,
MobileNetV3_small_x1_25
,
MobileNetV3_large_x0_35
,
MobileNetV3_large_x0_5
,
MobileNetV3_large_x0_75
,
MobileNetV3_large_x1_0
,
MobileNetV3_large_x1_25
from
ppcls.arch.backbone.model_zoo.shufflenet_v2
import
ShuffleNetV2_x0_25
,
ShuffleNetV2_x0_33
,
ShuffleNetV2_x0_5
,
ShuffleNetV2_x1_0
,
ShuffleNetV2_x1_5
,
ShuffleNetV2_x2_0
,
ShuffleNetV2_swish
from
ppcls.arch.backbone.model_zoo.alexnet
import
AlexNet
from
ppcls.arch.backbone.model_zoo.inception_v3
import
InceptionV3
from
ppcls.arch.backbone.model_zoo.inception_v4
import
InceptionV4
from
ppcls.arch.backbone.model_zoo.xception
import
Xception41
,
Xception65
,
Xception71
from
ppcls.arch.backbone.model_zoo.xception_deeplab
import
Xception41_deeplab
,
Xception65_deeplab
,
Xception71_deeplab
from
ppcls.arch.backbone.model_zoo.resnext101_wsl
import
ResNeXt101_32x8d_wsl
,
ResNeXt101_32x16d_wsl
,
ResNeXt101_32x32d_wsl
,
ResNeXt101_32x48d_wsl
from
ppcls.arch.backbone.model_zoo.squeezenet
import
SqueezeNet1_0
,
SqueezeNet1_1
from
ppcls.arch.backbone.model_zoo.vgg
import
VGG11
,
VGG13
,
VGG16
,
VGG19
from
ppcls.arch.backbone.model_zoo.darknet
import
DarkNet53
from
ppcls.arch.backbone.model_zoo.regnet
import
RegNetX_200MF
,
RegNetX_4GF
,
RegNetX_32GF
,
RegNetY_200MF
,
RegNetY_4GF
,
RegNetY_32GF
from
ppcls.arch.backbone.model_zoo.vision_transformer
import
ViT_small_patch16_224
,
ViT_base_patch16_224
,
ViT_base_patch16_384
,
ViT_base_patch32_384
,
ViT_large_patch16_224
,
ViT_large_patch16_384
,
ViT_large_patch32_384
,
ViT_huge_patch16_224
,
ViT_huge_patch32_384
...
...
ppcls/configs/ImageNet/HRNet/HRNet_W18_C.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
120
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
HRNet_W18_C"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Piecewise
learning_rate
:
0.1
decay_epochs
:
[
30
,
60
,
90
]
values
:
[
0.1
,
0.01
,
0.001
,
0.0001
]
regularizer
:
name
:
'
L2'
coeff
:
0.0001
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/configs/ImageNet/HRNet/HRNet_W30_C.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
120
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
HRNet_W30_C"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Piecewise
learning_rate
:
0.1
decay_epochs
:
[
30
,
60
,
90
]
values
:
[
0.1
,
0.01
,
0.001
,
0.0001
]
regularizer
:
name
:
'
L2'
coeff
:
0.0001
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/configs/ImageNet/HRNet/HRNet_W32_C.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
120
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
HRNet_W32_C"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Piecewise
learning_rate
:
0.1
decay_epochs
:
[
30
,
60
,
90
]
values
:
[
0.1
,
0.01
,
0.001
,
0.0001
]
regularizer
:
name
:
'
L2'
coeff
:
0.0001
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/configs/ImageNet/HRNet/HRNet_W40_C.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
120
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
HRNet_W40_C"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Piecewise
learning_rate
:
0.1
decay_epochs
:
[
30
,
60
,
90
]
values
:
[
0.1
,
0.01
,
0.001
,
0.0001
]
regularizer
:
name
:
'
L2'
coeff
:
0.0001
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/configs/ImageNet/HRNet/HRNet_W44_C.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
120
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
HRNet_W44_C"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Piecewise
learning_rate
:
0.1
decay_epochs
:
[
30
,
60
,
90
]
values
:
[
0.1
,
0.01
,
0.001
,
0.0001
]
regularizer
:
name
:
'
L2'
coeff
:
0.0001
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/configs/ImageNet/HRNet/HRNet_W48_C.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
120
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
HRNet_W48_C"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Piecewise
learning_rate
:
0.1
decay_epochs
:
[
30
,
60
,
90
]
values
:
[
0.1
,
0.01
,
0.001
,
0.0001
]
regularizer
:
name
:
'
L2'
coeff
:
0.0001
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/configs/ImageNet/HRNet/HRNet_W64_C.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
120
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
HRNet_W64_C"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Piecewise
learning_rate
:
0.1
decay_epochs
:
[
30
,
60
,
90
]
values
:
[
0.1
,
0.01
,
0.001
,
0.0001
]
regularizer
:
name
:
'
L2'
coeff
:
0.0001
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/configs/ImageNet/Ineption/InceptionV3.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
200
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
InceptionV3"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Cosine
learning_rate
:
0.045
decay_epochs
:
[
30
,
60
,
90
]
values
:
[
0.1
,
0.01
,
0.001
,
0.0001
]
regularizer
:
name
:
'
L2'
coeff
:
0.0001
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
299
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
channel_first
:
False
-
ResizeImage
:
resize_short
:
320
-
CropImage
:
size
:
299
-
NormalizeImage
:
scale
:
1.0/255.0
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
-
ToCHWImage
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/configs/ImageNet/MobileNetV1/MobileNetV1.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
120
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
MobileNetV1"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Piecewise
learning_rate
:
0.1
decay_epochs
:
[
30
,
60
,
90
]
values
:
[
0.1
,
0.01
,
0.001
,
0.0001
]
regularizer
:
name
:
'
L2'
coeff
:
0.00003
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/configs/ImageNet/MobileNetV1/MobileNetV1_x0_25.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
120
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
MobileNetV1_x0_25"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Piecewise
learning_rate
:
0.1
decay_epochs
:
[
30
,
60
,
90
]
values
:
[
0.1
,
0.01
,
0.001
,
0.0001
]
regularizer
:
name
:
'
L2'
coeff
:
0.00003
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/configs/ImageNet/MobileNetV1/MobileNetV1_x0_5.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
120
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
MobileNetV1_x0_5"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Piecewise
learning_rate
:
0.1
decay_epochs
:
[
30
,
60
,
90
]
values
:
[
0.1
,
0.01
,
0.001
,
0.0001
]
regularizer
:
name
:
'
L2'
coeff
:
0.00003
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/configs/ImageNet/MobileNetV1/MobileNetV1_x0_75.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
120
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
MobileNetV1_x0_75"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Piecewise
learning_rate
:
0.1
decay_epochs
:
[
30
,
60
,
90
]
values
:
[
0.1
,
0.01
,
0.001
,
0.0001
]
regularizer
:
name
:
'
L2'
coeff
:
0.00003
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/configs/ImageNet/MobileNetV3/MobileNetV3_large_x0_35.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
360
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
MobileNetV3_large_x0_35"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Cosine
learning_rate
:
1.3
regularizer
:
name
:
'
L2'
coeff
:
0.00002
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
512
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/configs/ImageNet/MobileNetV3/MobileNetV3_large_x0_5.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
360
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
MobileNetV3_large_x0_5"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Cosine
learning_rate
:
1.3
regularizer
:
name
:
'
L2'
coeff
:
0.00002
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
512
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/configs/ImageNet/MobileNetV3/MobileNetV3_large_x0_75.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
360
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
MobileNetV3_large_x0_75"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Cosine
learning_rate
:
1.3
regularizer
:
name
:
'
L2'
coeff
:
0.00002
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
512
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/configs/ImageNet/MobileNetV3/MobileNetV3_large_x1_0.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
360
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
MobileNetV3_large_x1_0"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Cosine
learning_rate
:
1.3
regularizer
:
name
:
'
L2'
coeff
:
0.00002
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
512
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/configs/ImageNet/MobileNetV3/MobileNetV3_large_x1_25.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
360
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
MobileNetV3_large_x1_25"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Cosine
learning_rate
:
1.3
regularizer
:
name
:
'
L2'
coeff
:
0.00002
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
512
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/configs/ImageNet/MobileNetV3/MobileNetV3_small_x0_35.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
360
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
MobileNetV3_small_x0_35"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Cosine
learning_rate
:
1.3
regularizer
:
name
:
'
L2'
coeff
:
0.00002
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
512
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/configs/ImageNet/MobileNetV3/MobileNetV3_small_x0_5.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
360
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
MobileNetV3_small_x0_5"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Cosine
learning_rate
:
1.3
regularizer
:
name
:
'
L2'
coeff
:
0.00002
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
512
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/configs/ImageNet/MobileNetV3/MobileNetV3_small_x0_75.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
360
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
MobileNetV3_small_x0_75"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Cosine
learning_rate
:
1.3
regularizer
:
name
:
'
L2'
coeff
:
0.00002
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
512
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/configs/ImageNet/MobileNetV3/MobileNetV3_small_x1_0.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
360
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
MobileNetV3_small_x1_0"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Cosine
learning_rate
:
1.3
regularizer
:
name
:
'
L2'
coeff
:
0.00002
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
512
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/configs/ImageNet/MobileNetV3/MobileNetV3_small_x1_25.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
360
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
MobileNetV3_small_x1_25"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Cosine
learning_rate
:
1.3
regularizer
:
name
:
'
L2'
coeff
:
0.00002
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
512
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/configs/ImageNet/ResNet/ResNet101.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
120
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
ResNet101"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Piecewise
learning_rate
:
0.1
decay_epochs
:
[
30
,
60
,
90
]
values
:
[
0.1
,
0.01
,
0.001
,
0.0001
]
regularizer
:
name
:
'
L2'
coeff
:
0.0001
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/configs/ImageNet/ResNet/ResNet101_vd.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
200
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
ResNet101_vd"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Cosine
learning_rate
:
0.1
regularizer
:
name
:
'
L2'
coeff
:
0.0001
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/configs/ImageNet/ResNet/ResNet152.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
120
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
ResNet152"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Piecewise
learning_rate
:
0.1
decay_epochs
:
[
30
,
60
,
90
]
values
:
[
0.1
,
0.01
,
0.001
,
0.0001
]
regularizer
:
name
:
'
L2'
coeff
:
0.0001
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/configs/ImageNet/ResNet/ResNet152_vd.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
200
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
ResNet152_vd"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Cosine
learning_rate
:
0.1
regularizer
:
name
:
'
L2'
coeff
:
0.0001
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/configs/ImageNet/ResNet/ResNet18.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
120
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
ResNet18"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Piecewise
learning_rate
:
0.1
decay_epochs
:
[
30
,
60
,
90
]
values
:
[
0.1
,
0.01
,
0.001
,
0.0001
]
regularizer
:
name
:
'
L2'
coeff
:
0.0001
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/configs/ImageNet/ResNet/ResNet18_vd.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
200
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
ResNet18_vd"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Cosine
learning_rate
:
0.1
regularizer
:
name
:
'
L2'
coeff
:
0.00007
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/configs/ImageNet/ResNet/ResNet200_vd.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
200
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
ResNet200_vd"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Cosine
learning_rate
:
0.1
regularizer
:
name
:
'
L2'
coeff
:
0.0001
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/configs/ImageNet/ResNet/ResNet34.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
120
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
ResNet34"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Piecewise
learning_rate
:
0.1
decay_epochs
:
[
30
,
60
,
90
]
values
:
[
0.1
,
0.01
,
0.001
,
0.0001
]
regularizer
:
name
:
'
L2'
coeff
:
0.0001
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/configs/ImageNet/ResNet/ResNet34_vd.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
200
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
ResNet101_vd"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Cosine
learning_rate
:
0.1
regularizer
:
name
:
'
L2'
coeff
:
0.00007
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/configs/ImageNet/ResNet/ResNet50.yaml
浏览文件 @
d58fd3b7
...
...
@@ -45,52 +45,52 @@ Optimizer:
# data loader for train and eval
DataLoader
:
Train
:
# Dataset:
# Sampler:
# Loader:
batch_size
:
256
num_workers
:
4
file_list
:
"
./dataset/ILSVRC2012/train_list.txt"
data_dir
:
"
./dataset/ILSVRC2012/"
shuffle_seed
:
0
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
# Dataset:
# Sampler:
# Loader:
batch_size
:
128
num_workers
:
4
file_list
:
"
./dataset/ILSVRC2012/val_list.txt"
data_dir
:
"
./dataset/ILSVRC2012/"
shuffle_seed
:
0
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
...
...
ppcls/configs/ImageNet/ResNet/ResNet50_vd.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
200
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
ResNet50_vd"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Cosine
learning_rate
:
0.1
regularizer
:
name
:
'
L2'
coeff
:
0.00007
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/configs/ImageNet/VGG/VGG11.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
90
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
VGG11"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Cosine
learning_rate
:
0.1
regularizer
:
name
:
'
L2'
coeff
:
0.0002
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/configs/ImageNet/VGG/VGG13.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
90
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
VGG13"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Cosine
learning_rate
:
0.01
regularizer
:
name
:
'
L2'
coeff
:
0.0003
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/configs/ImageNet/VGG/VGG16.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
90
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
VGG16"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Cosine
learning_rate
:
0.01
regularizer
:
name
:
'
L2'
coeff
:
0.0004
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/configs/ImageNet/VGG/VGG19.yaml
0 → 100644
浏览文件 @
d58fd3b7
# global configs
Global
:
checkpoints
:
null
pretrained_model
:
null
output_dir
:
"
./output/"
device
:
"
gpu"
class_num
:
1000
save_interval
:
1
eval_during_train
:
True
eval_interval
:
1
epochs
:
150
print_batch_step
:
10
use_visualdl
:
False
# used for static mode and model export
image_shape
:
[
3
,
224
,
224
]
save_inference_dir
:
"
./inference"
# model architecture
Arch
:
name
:
"
VGG19"
# loss function config for traing/eval process
Loss
:
Train
:
-
CELoss
:
weight
:
1.0
Eval
:
-
CELoss
:
weight
:
1.0
Optimizer
:
name
:
Momentum
momentum
:
0.9
lr
:
name
:
Cosine
learning_rate
:
0.01
regularizer
:
name
:
'
L2'
coeff
:
0.0004
# data loader for train and eval
DataLoader
:
Train
:
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/train_list.txt"
transform_ops
:
-
RandCropImage
:
size
:
224
-
RandFlipImage
:
flip_code
:
1
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
True
loader
:
num_workers
:
6
use_shared_memory
:
False
Eval
:
# TOTO: modify to the latest trainer
dataset
:
name
:
ImageNetDataset
image_root
:
"
./dataset/ILSVRC2012/"
cls_label_path
:
"
./dataset/ILSVRC2012/val_list.txt"
transform_ops
:
-
ResizeImage
:
size
:
224
-
NormalizeImage
:
scale
:
0.00392157
mean
:
[
0.485
,
0.456
,
0.406
]
std
:
[
0.229
,
0.224
,
0.225
]
order
:
'
'
sampler
:
name
:
DistributedBatchSampler
batch_size
:
64
drop_last
:
False
shuffle
:
False
loader
:
num_workers
:
6
use_shared_memory
:
False
Infer
:
infer_imgs
:
"
docs/images/whl/demo.jpg"
batch_size
:
10
transforms
:
-
DecodeImage
:
to_rgb
:
True
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
:
PostProcess
:
name
:
Topk
topk
:
5
class_id_map_file
:
"
ppcls/utils/imagenet1k_label_list.txt"
Metric
:
Train
:
-
Topk
:
k
:
[
1
,
5
]
Eval
:
-
Topk
:
k
:
[
1
,
5
]
ppcls/data/__init__.py
浏览文件 @
d58fd3b7
...
...
@@ -18,13 +18,13 @@ from paddle.io import DistributedBatchSampler, BatchSampler, DataLoader
from
ppcls.utils
import
logger
from
.
import
dataset
s
from
.
import
dataset
from
.
import
imaug
from
.
import
samplers
# dataset
from
.dataset
s
.imagenet_dataset
import
ImageNetDataset
from
.dataset.imagenet_dataset
import
ImageNetDataset
from
.dataset.multilabel_dataset
import
MultiLabelDataset
from
.dataset
s
.common_dataset
import
create_operators
from
.dataset.common_dataset
import
create_operators
# sampler
from
.samplers
import
DistributedRandomIdentitySampler
...
...
@@ -35,6 +35,7 @@ def build_dataloader(config, mode, device, seed=None):
assert
mode
in
[
'Train'
,
'Eval'
,
'Test'
],
"Mode should be Train, Eval or Test."
# build dataset
config_dataset
=
config
[
mode
][
'dataset'
]
config_dataset
=
copy
.
deepcopy
(
config_dataset
)
dataset_name
=
config_dataset
.
pop
(
'name'
)
if
'batch_transform_ops'
in
config_dataset
:
batch_transform
=
config_dataset
.
pop
(
'batch_transform_ops'
)
...
...
@@ -105,7 +106,7 @@ def build_dataloader(config, mode, device, seed=None):
logger
.
info
(
"build data_loader({}) success..."
.
format
(
data_loader
))
return
dataloader
return
data
_
loader
'''
# TODO: fix the format
...
...
ppcls/data/dataset/common_dataset.py
浏览文件 @
d58fd3b7
...
...
@@ -40,7 +40,6 @@ def create_operators(params):
assert
isinstance
(
params
,
list
),
(
'operator config should be a list'
)
ops
=
[]
for
operator
in
params
:
print
(
operator
)
assert
isinstance
(
operator
,
dict
)
and
len
(
operator
)
==
1
,
"yaml format error"
op_name
=
list
(
operator
)[
0
]
...
...
ppcls/data/postprocess/__init__.py
浏览文件 @
d58fd3b7
...
...
@@ -14,9 +14,9 @@
import
copy
import
importlib
from
.
import
topk
_process
from
.
import
topk
from
.topk
_process
import
Topk
from
.topk
import
Topk
def
build_postprocess
(
config
):
...
...
ppcls/data/preprocess/__init__.py
浏览文件 @
d58fd3b7
...
...
@@ -40,7 +40,6 @@ def transform(data, ops=[]):
""" transform """
for
op
in
ops
:
data
=
op
(
data
)
#print(data.shape, op)
return
data
...
...
ppcls/engine/trainer.py
浏览文件 @
d58fd3b7
...
...
@@ -104,7 +104,7 @@ class Trainer(object):
metric_func
=
self
.
_build_metric_info
(
self
.
config
[
"Metric"
])
train_dataloader
=
build_dataloader
(
self
.
config
[
"DataLoader"
],
"
t
rain"
,
train_dataloader
=
build_dataloader
(
self
.
config
[
"DataLoader"
],
"
T
rain"
,
self
.
device
)
step_each_epoch
=
len
(
train_dataloader
)
...
...
@@ -217,7 +217,7 @@ class Trainer(object):
def
eval
(
self
,
epoch_id
=
0
):
output_info
=
dict
()
eval_dataloader
=
build_dataloader
(
self
.
config
[
"DataLoader"
],
"
e
val"
,
eval_dataloader
=
build_dataloader
(
self
.
config
[
"DataLoader"
],
"
E
val"
,
self
.
device
)
self
.
model
.
eval
()
...
...
ppcls/losses/triplet.py
浏览文件 @
d58fd3b7
...
...
@@ -48,7 +48,6 @@ class TripletLossV2(nn.Layer):
# `dist_ap` means distance(anchor, positive)
## both `dist_ap` and `relative_p_inds` with shape [N, 1]
#print(is_pos.shape, dist.shape, type(is_pos), type(dist), paddle.reshape(paddle.masked_select(dist, is_pos),(bs, -1)))
'''
dist_ap, relative_p_inds = paddle.max(
paddle.reshape(dist[is_pos], (bs, -1)), axis=1, keepdim=True)
...
...
@@ -98,7 +97,6 @@ class TripletLoss(nn.Layer):
"""
inputs
=
input
[
"features"
]
#print(inputs.shape, targets.shape)
bs
=
inputs
.
shape
[
0
]
# Compute pairwise distance, replace by the official when merged
dist
=
paddle
.
pow
(
inputs
,
2
).
sum
(
axis
=
1
,
keepdim
=
True
).
expand
([
bs
,
bs
])
...
...
tools/run.sh
浏览文件 @
d58fd3b7
...
...
@@ -3,5 +3,5 @@
python3.7
-m
paddle.distributed.launch
\
--gpus
=
"0,1,2,3"
\
tools/train.py
\
-c
./
configs
/ResNet/ResNet50.yaml
\
-c
./
ppcls/configs/ImageNet
/ResNet/ResNet50.yaml
\
-o
print_interval
=
10
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