Merge pull request #819 from cuicheng01/develop_reg

Update configs

ppcls/configs/ImageNet/HRNet/HRNet_W18_C.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "HRNet_W18_C"
+  name: HRNet_W18_C
  
 # loss function config for traing/eval process
 Loss:
@@ -46,80 +46,80 @@ Optimizer:
 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: ''
+      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: 1.0/255.0
+            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
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: True
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
   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:
-              resize_short: 256
-          - CropImage:
-              size: 224
-          - NormalizeImage:
-              scale: 0.00392157
-              mean: [0.485, 0.456, 0.406]
-              std: [0.229, 0.224, 0.225]
-              order: ''
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
+      transform_ops:
+        - 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: ''
     sampler:
-        name: DistributedBatchSampler
-        batch_size: 64
-        drop_last: False
-        shuffle: False
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: False
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  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:
+    - 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"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
-    Train:
+  Train:
     - TopkAcc:
         topk: [1, 5]
-    Eval:
+  Eval:
     - TopkAcc:
         topk: [1, 5]

ppcls/configs/ImageNet/HRNet/HRNet_W30_C.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "HRNet_W30_C"
+  name: HRNet_W30_C
  
 # loss function config for traing/eval process
 Loss:
@@ -46,80 +46,80 @@ Optimizer:
 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: ''
+      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: 1.0/255.0
+            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
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: True
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
   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:
-              resize_short: 256
-          - CropImage:
-              size: 224
-          - NormalizeImage:
-              scale: 0.00392157
-              mean: [0.485, 0.456, 0.406]
-              std: [0.229, 0.224, 0.225]
-              order: ''
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
+      transform_ops:
+        - 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: ''
     sampler:
-        name: DistributedBatchSampler
-        batch_size: 64
-        drop_last: False
-        shuffle: False
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: False
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  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:
+    - 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"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
-    Train:
+  Train:
     - TopkAcc:
         topk: [1, 5]
-    Eval:
+  Eval:
     - TopkAcc:
         topk: [1, 5]

ppcls/configs/ImageNet/HRNet/HRNet_W32_C.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "HRNet_W32_C"
+  name: HRNet_W32_C
  
 # loss function config for traing/eval process
 Loss:
@@ -46,80 +46,80 @@ Optimizer:
 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: ''
+      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: 1.0/255.0
+            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
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: True
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
   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:
-              resize_short: 256
-          - CropImage:
-              size: 224
-          - NormalizeImage:
-              scale: 0.00392157
-              mean: [0.485, 0.456, 0.406]
-              std: [0.229, 0.224, 0.225]
-              order: ''
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
+      transform_ops:
+        - 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: ''
     sampler:
-        name: DistributedBatchSampler
-        batch_size: 64
-        drop_last: False
-        shuffle: False
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: False
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  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:
+    - 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"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
-    Train:
+  Train:
     - TopkAcc:
         topk: [1, 5]
-    Eval:
+  Eval:
     - TopkAcc:
         topk: [1, 5]

ppcls/configs/ImageNet/HRNet/HRNet_W40_C.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "HRNet_W40_C"
+  name: HRNet_W40_C
  
 # loss function config for traing/eval process
 Loss:
@@ -46,80 +46,80 @@ Optimizer:
 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: ''
+      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: 1.0/255.0
+            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
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: True
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
   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:
-              resize_short: 256
-          - CropImage:
-              size: 224
-          - NormalizeImage:
-              scale: 0.00392157
-              mean: [0.485, 0.456, 0.406]
-              std: [0.229, 0.224, 0.225]
-              order: ''
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
+      transform_ops:
+        - 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: ''
     sampler:
-        name: DistributedBatchSampler
-        batch_size: 64
-        drop_last: False
-        shuffle: False
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: False
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  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:
+    - 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"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
-    Train:
+  Train:
     - TopkAcc:
         topk: [1, 5]
-    Eval:
+  Eval:
     - TopkAcc:
         topk: [1, 5]

ppcls/configs/ImageNet/HRNet/HRNet_W44_C.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "HRNet_W44_C"
+  name: HRNet_W44_C
  
 # loss function config for traing/eval process
 Loss:
@@ -46,78 +46,80 @@ Optimizer:
 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: ''
+      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: 1.0/255.0
+            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
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: True
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
   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: ''
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
+      transform_ops:
+        - 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: ''
     sampler:
-        name: DistributedBatchSampler
-        batch_size: 64
-        drop_last: False
-        shuffle: False
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: False
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  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:
+    - 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"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
-    Train:
+  Train:
     - TopkAcc:
         topk: [1, 5]
-    Eval:
+  Eval:
     - TopkAcc:
         topk: [1, 5]

ppcls/configs/ImageNet/HRNet/HRNet_W48_C.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "HRNet_W48_C"
+  name: HRNet_W48_C
  
 # loss function config for traing/eval process
 Loss:
@@ -46,80 +46,80 @@ Optimizer:
 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: ''
+      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: 1.0/255.0
+            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
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: True
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
   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:
-              resize_short: 256
-          - CropImage:
-              size: 224
-          - NormalizeImage:
-              scale: 0.00392157
-              mean: [0.485, 0.456, 0.406]
-              std: [0.229, 0.224, 0.225]
-              order: ''
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
+      transform_ops:
+        - 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: ''
     sampler:
-        name: DistributedBatchSampler
-        batch_size: 64
-        drop_last: False
-        shuffle: False
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: False
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  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:
+    - 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"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
-    Train:
+  Train:
     - TopkAcc:
         topk: [1, 5]
-    Eval:
+  Eval:
     - TopkAcc:
         topk: [1, 5]

ppcls/configs/ImageNet/HRNet/HRNet_W64_C.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "HRNet_W64_C"
+  name: HRNet_W64_C
  
 # loss function config for traing/eval process
 Loss:
@@ -46,80 +46,80 @@ Optimizer:
 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: ''
+      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: 1.0/255.0
+            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
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: True
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
   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:
-              resize_short: 256
-          - CropImage:
-              size: 224
-          - NormalizeImage:
-              scale: 0.00392157
-              mean: [0.485, 0.456, 0.406]
-              std: [0.229, 0.224, 0.225]
-              order: ''
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
+      transform_ops:
+        - 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: ''
     sampler:
-        name: DistributedBatchSampler
-        batch_size: 64
-        drop_last: False
-        shuffle: False
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: False
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  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:
+    - 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"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
-    Train:
+  Train:
     - TopkAcc:
         topk: [1, 5]
-    Eval:
+  Eval:
     - TopkAcc:
         topk: [1, 5]

ppcls/configs/ImageNet/Ineption/InceptionV3.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -12,18 +12,19 @@ Global:
   print_batch_step: 10
   use_visualdl: False
   # used for static mode and model export
-  image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  image_shape: [3, 299, 299]
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "InceptionV3"
+  name: InceptionV3
  
 # loss function config for traing/eval process
 Loss:
   Train:
     - CELoss:
         weight: 1.0
+        epsilon: 0.1
   Eval:
     - CELoss:
         weight: 1.0
@@ -35,8 +36,6 @@ Optimizer:
   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
@@ -46,80 +45,80 @@ Optimizer:
 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: ''
+      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: 1.0/255.0
+            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
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: True
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
   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:
-              resize_short: 320
-          - CropImage:
-              size: 299
-          - NormalizeImage:
-              scale: 0.00392157
-              mean: [0.485, 0.456, 0.406]
-              std: [0.229, 0.224, 0.225]
-              order: ''
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
+      transform_ops:
+        - 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: ''
     sampler:
-        name: DistributedBatchSampler
-        batch_size: 64
-        drop_last: False
-        shuffle: False
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: False
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  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:
+    - 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"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
-    Train:
+  Train:
     - TopkAcc:
         topk: [1, 5]
-    Eval:
+  Eval:
     - TopkAcc:
         topk: [1, 5]

ppcls/configs/ImageNet/MobileNetV1/MobileNetV1.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "MobileNetV1"
+  name: MobileNetV1
  
 # loss function config for traing/eval process
 Loss:
@@ -39,87 +39,87 @@ Optimizer:
     values: [0.1, 0.01, 0.001, 0.0001]
   regularizer:
     name: 'L2'
-    coeff: 0.00003
+    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: ''
+      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: 1.0/255.0
+            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
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: True
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
   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:
-              resize_short: 256
-          - CropImage:
-              size: 224
-          - NormalizeImage:
-              scale: 0.00392157
-              mean: [0.485, 0.456, 0.406]
-              std: [0.229, 0.224, 0.225]
-              order: ''
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
+      transform_ops:
+        - 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: ''
     sampler:
-        name: DistributedBatchSampler
-        batch_size: 64
-        drop_last: False
-        shuffle: False
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: False
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  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:
+    - 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"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
-    Train:
+  Train:
     - TopkAcc:
         topk: [1, 5]
-    Eval:
+  Eval:
     - TopkAcc:
         topk: [1, 5]

ppcls/configs/ImageNet/MobileNetV1/MobileNetV1_x0_25.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "MobileNetV1_x0_25"
+  name: MobileNetV1_x0_25
  
 # loss function config for traing/eval process
 Loss:
@@ -39,87 +39,87 @@ Optimizer:
     values: [0.1, 0.01, 0.001, 0.0001]
   regularizer:
     name: 'L2'
-    coeff: 0.00003
+    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: ''
+      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: 1.0/255.0
+            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
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: True
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
   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:
-              resize_short: 256
-          - CropImage:
-              size: 224
-          - NormalizeImage:
-              scale: 0.00392157
-              mean: [0.485, 0.456, 0.406]
-              std: [0.229, 0.224, 0.225]
-              order: ''
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
+      transform_ops:
+        - 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: ''
     sampler:
-        name: DistributedBatchSampler
-        batch_size: 64
-        drop_last: False
-        shuffle: False
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: False
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  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:
+    - 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"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
-    Train:
+  Train:
     - TopkAcc:
         topk: [1, 5]
-    Eval:
+  Eval:
     - TopkAcc:
         topk: [1, 5]

ppcls/configs/ImageNet/MobileNetV1/MobileNetV1_x0_5.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "MobileNetV1_x0_5"
+  name: MobileNetV1_x0_5
  
 # loss function config for traing/eval process
 Loss:
@@ -39,87 +39,87 @@ Optimizer:
     values: [0.1, 0.01, 0.001, 0.0001]
   regularizer:
     name: 'L2'
-    coeff: 0.00003
+    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: ''
+      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: 1.0/255.0
+            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
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: True
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
   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:
-              resize_short: 256
-          - CropImage:
-              size: 224
-          - NormalizeImage:
-              scale: 0.00392157
-              mean: [0.485, 0.456, 0.406]
-              std: [0.229, 0.224, 0.225]
-              order: ''
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
+      transform_ops:
+        - 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: ''
     sampler:
-        name: DistributedBatchSampler
-        batch_size: 64
-        drop_last: False
-        shuffle: False
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: False
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  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:
+    - 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"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
-    Train:
+  Train:
     - TopkAcc:
         topk: [1, 5]
-    Eval:
+  Eval:
     - TopkAcc:
         topk: [1, 5]

ppcls/configs/ImageNet/MobileNetV1/MobileNetV1_x0_75.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "MobileNetV1_x0_75"
+  name: MobileNetV1_x0_75
  
 # loss function config for traing/eval process
 Loss:
@@ -39,87 +39,87 @@ Optimizer:
     values: [0.1, 0.01, 0.001, 0.0001]
   regularizer:
     name: 'L2'
-    coeff: 0.00003
+    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: ''
+      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: 1.0/255.0
+            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
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: True
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
   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:
-              resize_short: 256
-          - CropImage:
-              size: 224
-          - NormalizeImage:
-              scale: 0.00392157
-              mean: [0.485, 0.456, 0.406]
-              std: [0.229, 0.224, 0.225]
-              order: ''
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
+      transform_ops:
+        - 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: ''
     sampler:
-        name: DistributedBatchSampler
-        batch_size: 64
-        drop_last: False
-        shuffle: False
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: False
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  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:
+    - 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"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
-    Train:
+  Train:
     - TopkAcc:
         topk: [1, 5]
-    Eval:
+  Eval:
     - TopkAcc:
         topk: [1, 5]

ppcls/configs/ImageNet/MobileNetV3/MobileNetV3_large_x0_35.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "MobileNetV3_large_x0_35"
+  name: MobileNetV3_large_x0_35
  
 # loss function config for traing/eval process
 Loss:
@@ -45,80 +45,80 @@ Optimizer:
 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: ''
+      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: 1.0/255.0
+            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
+      name: DistributedBatchSampler
+      batch_size: 512
+      drop_last: False
+      shuffle: True
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
   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:
-              resize_short: 256
-          - CropImage:
-              size: 224
-          - NormalizeImage:
-              scale: 0.00392157
-              mean: [0.485, 0.456, 0.406]
-              std: [0.229, 0.224, 0.225]
-              order: ''
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
+      transform_ops:
+        - 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: ''
     sampler:
-        name: DistributedBatchSampler
-        batch_size: 64
-        drop_last: False
-        shuffle: False
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: False
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  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:
+    - 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"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
-    Train:
+  Train:
     - TopkAcc:
         topk: [1, 5]
-    Eval:
+  Eval:
     - TopkAcc:
         topk: [1, 5]

ppcls/configs/ImageNet/MobileNetV3/MobileNetV3_large_x0_5.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "MobileNetV3_large_x0_5"
+  name: MobileNetV3_large_x0_5
  
 # loss function config for traing/eval process
 Loss:
@@ -45,80 +45,80 @@ Optimizer:
 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: ''
+      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: 1.0/255.0
+            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
+      name: DistributedBatchSampler
+      batch_size: 512
+      drop_last: False
+      shuffle: True
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
   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:
-              resize_short: 256
-          - CropImage:
-              size: 224
-          - NormalizeImage:
-              scale: 0.00392157
-              mean: [0.485, 0.456, 0.406]
-              std: [0.229, 0.224, 0.225]
-              order: ''
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
+      transform_ops:
+        - 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: ''
     sampler:
-        name: DistributedBatchSampler
-        batch_size: 64
-        drop_last: False
-        shuffle: False
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: False
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  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:
+    - 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"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
-    Train:
+  Train:
     - TopkAcc:
         topk: [1, 5]
-    Eval:
+  Eval:
     - TopkAcc:
         topk: [1, 5]

ppcls/configs/ImageNet/MobileNetV3/MobileNetV3_large_x0_75.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "MobileNetV3_large_x0_75"
+  name: MobileNetV3_large_x0_75
  
 # loss function config for traing/eval process
 Loss:
@@ -45,80 +45,80 @@ Optimizer:
 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: ''
+      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: 1.0/255.0
+            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
+      name: DistributedBatchSampler
+      batch_size: 512
+      drop_last: False
+      shuffle: True
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
   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:
-              resize_short: 256
-          - CropImage:
-              size: 224
-          - NormalizeImage:
-              scale: 0.00392157
-              mean: [0.485, 0.456, 0.406]
-              std: [0.229, 0.224, 0.225]
-              order: ''
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
+      transform_ops:
+        - 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: ''
     sampler:
-        name: DistributedBatchSampler
-        batch_size: 64
-        drop_last: False
-        shuffle: False
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: False
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  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:
+    - 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"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
-    Train:
+  Train:
     - TopkAcc:
         topk: [1, 5]
-    Eval:
+  Eval:
     - TopkAcc:
         topk: [1, 5]

ppcls/configs/ImageNet/MobileNetV3/MobileNetV3_large_x1_0.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "MobileNetV3_large_x1_0"
+  name: MobileNetV3_large_x1_0
  
 # loss function config for traing/eval process
 Loss:
@@ -45,81 +45,81 @@ Optimizer:
 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
-          - AutoAugment:
-          - NormalizeImage:
-              scale: 0.00392157
-              mean: [0.485, 0.456, 0.406]
-              std: [0.229, 0.224, 0.225]
-              order: ''
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/train_list.txt
+      transform_ops:
+        - RandCropImage:
+            size: 224
+        - RandFlipImage:
+            flip_code: 1
+        - AutoAugment:
+        - NormalizeImage:
+            scale: 1.0/255.0
+            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
+      name: DistributedBatchSampler
+      batch_size: 512
+      drop_last: False
+      shuffle: True
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
   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:
-              resize_short: 256
-          - CropImage:
-              size: 224
-          - NormalizeImage:
-              scale: 0.00392157
-              mean: [0.485, 0.456, 0.406]
-              std: [0.229, 0.224, 0.225]
-              order: ''
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
+      transform_ops:
+        - 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: ''
     sampler:
-        name: DistributedBatchSampler
-        batch_size: 64
-        drop_last: False
-        shuffle: False
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: False
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  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:
+    - 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"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
-    Train:
+  Train:
     - TopkAcc:
         topk: [1, 5]
-    Eval:
+  Eval:
     - TopkAcc:
         topk: [1, 5]

ppcls/configs/ImageNet/MobileNetV3/MobileNetV3_large_x1_25.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "MobileNetV3_large_x1_25"
+  name: MobileNetV3_large_x1_25
  
 # loss function config for traing/eval process
 Loss:
@@ -38,87 +38,87 @@ Optimizer:
     learning_rate: 1.3
   regularizer:
     name: 'L2'
-    coeff: 0.00002
+    coeff: 0.00004
 
 
 # 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: ''
+      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: 1.0/255.0
+            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
+      name: DistributedBatchSampler
+      batch_size: 512
+      drop_last: False
+      shuffle: True
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
   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:
-              resize_short: 256
-          - CropImage:
-              size: 224
-          - NormalizeImage:
-              scale: 0.00392157
-              mean: [0.485, 0.456, 0.406]
-              std: [0.229, 0.224, 0.225]
-              order: ''
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
+      transform_ops:
+        - 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: ''
     sampler:
-        name: DistributedBatchSampler
-        batch_size: 64
-        drop_last: False
-        shuffle: False
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: False
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  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:
+    - 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"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
-    Train:
+  Train:
     - TopkAcc:
         topk: [1, 5]
-    Eval:
+  Eval:
     - TopkAcc:
         topk: [1, 5]

ppcls/configs/ImageNet/MobileNetV3/MobileNetV3_small_x0_35.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "MobileNetV3_small_x0_35"
+  name: MobileNetV3_small_x0_35
  
 # loss function config for traing/eval process
 Loss:
@@ -38,87 +38,87 @@ Optimizer:
     learning_rate: 1.3
   regularizer:
     name: 'L2'
-    coeff: 0.00002
+    coeff: 0.00001
 
 
 # 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: ''
+      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: 1.0/255.0
+            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
+      name: DistributedBatchSampler
+      batch_size: 512
+      drop_last: False
+      shuffle: True
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
   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:
-              resize_short: 256
-          - CropImage:
-              size: 224
-          - NormalizeImage:
-              scale: 0.00392157
-              mean: [0.485, 0.456, 0.406]
-              std: [0.229, 0.224, 0.225]
-              order: ''
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
+      transform_ops:
+        - 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: ''
     sampler:
-        name: DistributedBatchSampler
-        batch_size: 64
-        drop_last: False
-        shuffle: False
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: False
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  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:
+    - 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"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
-    Train:
+  Train:
     - TopkAcc:
         topk: [1, 5]
-    Eval:
+  Eval:
     - TopkAcc:
         topk: [1, 5]

ppcls/configs/ImageNet/MobileNetV3/MobileNetV3_small_x0_5.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "MobileNetV3_small_x0_5"
+  name: MobileNetV3_small_x0_5
  
 # loss function config for traing/eval process
 Loss:
@@ -38,87 +38,87 @@ Optimizer:
     learning_rate: 1.3
   regularizer:
     name: 'L2'
-    coeff: 0.00002
+    coeff: 0.00001
 
 
 # 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: ''
+      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: 1.0/255.0
+            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
+      name: DistributedBatchSampler
+      batch_size: 512
+      drop_last: False
+      shuffle: True
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
   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:
-              resize_short: 256
-          - CropImage:
-              size: 224
-          - NormalizeImage:
-              scale: 0.00392157
-              mean: [0.485, 0.456, 0.406]
-              std: [0.229, 0.224, 0.225]
-              order: ''
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
+      transform_ops:
+        - 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: ''
     sampler:
-        name: DistributedBatchSampler
-        batch_size: 64
-        drop_last: False
-        shuffle: False
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: False
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  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:
+    - 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"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
-    Train:
+  Train:
     - TopkAcc:
         topk: [1, 5]
-    Eval:
+  Eval:
     - TopkAcc:
         topk: [1, 5]

ppcls/configs/ImageNet/MobileNetV3/MobileNetV3_small_x0_75.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "MobileNetV3_small_x0_75"
+  name: MobileNetV3_small_x0_75
  
 # loss function config for traing/eval process
 Loss:
@@ -45,80 +45,80 @@ Optimizer:
 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: ''
+      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: 1.0/255.0
+            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
+      name: DistributedBatchSampler
+      batch_size: 512
+      drop_last: False
+      shuffle: True
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
   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:
-              resize_short: 256
-          - CropImage:
-              size: 224
-          - NormalizeImage:
-              scale: 0.00392157
-              mean: [0.485, 0.456, 0.406]
-              std: [0.229, 0.224, 0.225]
-              order: ''
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
+      transform_ops:
+        - 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: ''
     sampler:
-        name: DistributedBatchSampler
-        batch_size: 64
-        drop_last: False
-        shuffle: False
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: False
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  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:
+    - 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"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
-    Train:
+  Train:
     - TopkAcc:
         topk: [1, 5]
-    Eval:
+  Eval:
     - TopkAcc:
         topk: [1, 5]

ppcls/configs/ImageNet/MobileNetV3/MobileNetV3_small_x1_0.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "MobileNetV3_small_x1_0"
+  name: MobileNetV3_small_x1_0
  
 # loss function config for traing/eval process
 Loss:
@@ -45,80 +45,80 @@ Optimizer:
 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: ''
+      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: 1.0/255.0
+            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
+      name: DistributedBatchSampler
+      batch_size: 512
+      drop_last: False
+      shuffle: True
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
   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:
-              resize_short: 256
-          - CropImage:
-              size: 224
-          - NormalizeImage:
-              scale: 0.00392157
-              mean: [0.485, 0.456, 0.406]
-              std: [0.229, 0.224, 0.225]
-              order: ''
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
+      transform_ops:
+        - 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: ''
     sampler:
-        name: DistributedBatchSampler
-        batch_size: 64
-        drop_last: False
-        shuffle: False
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: False
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  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:
+    - 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"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
-    Train:
+  Train:
     - TopkAcc:
         topk: [1, 5]
-    Eval:
+  Eval:
     - TopkAcc:
         topk: [1, 5]

ppcls/configs/ImageNet/MobileNetV3/MobileNetV3_small_x1_25.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "MobileNetV3_small_x1_25"
+  name: MobileNetV3_small_x1_25
  
 # loss function config for traing/eval process
 Loss:
@@ -45,80 +45,80 @@ Optimizer:
 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: ''
+      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: 1.0/255.0
+            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
+      name: DistributedBatchSampler
+      batch_size: 512
+      drop_last: False
+      shuffle: True
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
   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:
-              resize_short: 256
-          - CropImage:
-              size: 224
-          - NormalizeImage:
-              scale: 0.00392157
-              mean: [0.485, 0.456, 0.406]
-              std: [0.229, 0.224, 0.225]
-              order: ''
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
+      transform_ops:
+        - 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: ''
     sampler:
-        name: DistributedBatchSampler
-        batch_size: 64
-        drop_last: False
-        shuffle: False
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: False
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  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:
+    - 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"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
-    Train:
+  Train:
     - TopkAcc:
         topk: [1, 5]
-    Eval:
+  Eval:
     - TopkAcc:
         topk: [1, 5]

ppcls/configs/ImageNet/ResNet/ResNet101.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "ResNet101"
+  name: ResNet101
  
 # loss function config for traing/eval process
 Loss:
@@ -30,10 +30,10 @@ Loss:
 
 
 Optimizer:
-  name: "Momentum"
+  name: Momentum
   momentum: 0.9
   lr:
-    name: "Piecewise"
+    name: Piecewise
     learning_rate: 0.1
     decay_epochs: [30, 60, 90]
     values: [0.1, 0.01, 0.001, 0.0001]
@@ -46,9 +46,9 @@ Optimizer:
 DataLoader:
   Train:
     dataset:
-      name: "ImageNetDataset"
-      image_root: "./dataset/ILSVRC2012/"
-      cls_label_path: "./dataset/ILSVRC2012/train_list.txt"
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/train_list.txt
       transform_ops:
         - RandCropImage:
             size: 224
@@ -61,20 +61,20 @@ DataLoader:
             order: ''
 
     sampler:
-      name: "DistributedBatchSampler"
+      name: DistributedBatchSampler
       batch_size: 64
       drop_last: False
       shuffle: True
     loader:
-      num_workers: 6
+      num_workers: 4
       use_shared_memory: True
 
   Eval:
     # TOTO: modify to the latest trainer
     dataset: 
-      name: "ImageNetDataset"
-      image_root: "./dataset/ILSVRC2012/"
-      cls_label_path: "./dataset/ILSVRC2012/val_list.txt"
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
       transform_ops:
         - ResizeImage:
             resize_short: 256
@@ -86,16 +86,16 @@ DataLoader:
             std: [0.229, 0.224, 0.225]
             order: ''
     sampler:
-      name: "DistributedBatchSampler"
+      name: DistributedBatchSampler
       batch_size: 64
       drop_last: False
       shuffle: False
     loader:
-      num_workers: 6
+      num_workers: 4
       use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  infer_imgs: docs/images/whl/demo.jpg
   batch_size: 10
   transforms:
     - DecodeImage:
@@ -112,9 +112,9 @@ Infer:
         order: ''
     - ToCHWImage:
   PostProcess:
-    name: "Topk"
+    name: Topk
     topk: 5
-    class_id_map_file: "ppcls/utils/imagenet1k_label_list.txt"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
   Train:

ppcls/configs/ImageNet/ResNet/ResNet101_vd.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "ResNet101_vd"
+  name: ResNet101_vd
  
 # loss function config for traing/eval process
 Loss:
@@ -31,10 +31,10 @@ Loss:
 
 
 Optimizer:
-  name: "Momentum"
+  name: Momentum
   momentum: 0.9
   lr:
-    name: "Cosine"
+    name: Cosine
     learning_rate: 0.1
   regularizer:
     name: 'L2'
@@ -45,9 +45,9 @@ Optimizer:
 DataLoader:
   Train:
     dataset:
-      name: "ImageNetDataset"
-      image_root: "./dataset/ILSVRC2012/"
-      cls_label_path: "./dataset/ILSVRC2012/train_list.txt"
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/train_list.txt
       transform_ops:
         - RandCropImage:
             size: 224
@@ -63,20 +63,20 @@ DataLoader:
             alpha: 0.2
 
     sampler:
-      name: "DistributedBatchSampler"
+      name: DistributedBatchSampler
       batch_size: 64
       drop_last: False
       shuffle: True
     loader:
-      num_workers: 6
+      num_workers: 4
       use_shared_memory: True
 
   Eval:
     # TOTO: modify to the latest trainer
     dataset: 
-      name: "ImageNetDataset"
-      image_root: "./dataset/ILSVRC2012/"
-      cls_label_path: "./dataset/ILSVRC2012/val_list.txt"
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
       transform_ops:
         - ResizeImage:
             resize_short: 256
@@ -88,16 +88,16 @@ DataLoader:
             std: [0.229, 0.224, 0.225]
             order: ''
     sampler:
-      name: "DistributedBatchSampler"
+      name: DistributedBatchSampler
       batch_size: 64
       drop_last: False
       shuffle: False
     loader:
-      num_workers: 6
+      num_workers: 4
       use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  infer_imgs: docs/images/whl/demo.jpg
   batch_size: 10
   transforms:
     - DecodeImage:
@@ -114,9 +114,9 @@ Infer:
         order: ''
     - ToCHWImage:
   PostProcess:
-    name: "Topk"
+    name: Topk
     topk: 5
-    class_id_map_file: "ppcls/utils/imagenet1k_label_list.txt"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
   Train:

ppcls/configs/ImageNet/ResNet/ResNet152.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "ResNet152"
+  name: ResNet152
  
 # loss function config for traing/eval process
 Loss:
@@ -30,10 +30,10 @@ Loss:
 
 
 Optimizer:
-  name: "Momentum"
+  name: Momentum
   momentum: 0.9
   lr:
-    name: "Piecewise"
+    name: Piecewise
     learning_rate: 0.1
     decay_epochs: [30, 60, 90]
     values: [0.1, 0.01, 0.001, 0.0001]
@@ -46,9 +46,9 @@ Optimizer:
 DataLoader:
   Train:
     dataset:
-      name: "ImageNetDataset"
-      image_root: "./dataset/ILSVRC2012/"
-      cls_label_path: "./dataset/ILSVRC2012/train_list.txt"
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/train_list.txt
       transform_ops:
         - RandCropImage:
             size: 224
@@ -61,20 +61,20 @@ DataLoader:
             order: ''
 
     sampler:
-      name: "DistributedBatchSampler"
+      name: DistributedBatchSampler
       batch_size: 64
       drop_last: False
       shuffle: True
     loader:
-      num_workers: 6
+      num_workers: 4
       use_shared_memory: True
 
   Eval:
     # TOTO: modify to the latest trainer
     dataset: 
-      name: "ImageNetDataset"
-      image_root: "./dataset/ILSVRC2012/"
-      cls_label_path: "./dataset/ILSVRC2012/val_list.txt"
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
       transform_ops:
         - ResizeImage:
             resize_short: 256
@@ -86,16 +86,16 @@ DataLoader:
             std: [0.229, 0.224, 0.225]
             order: ''
     sampler:
-      name: "DistributedBatchSampler"
+      name: DistributedBatchSampler
       batch_size: 64
       drop_last: False
       shuffle: False
     loader:
-      num_workers: 6
+      num_workers: 4
       use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  infer_imgs: docs/images/whl/demo.jpg
   batch_size: 10
   transforms:
     - DecodeImage:
@@ -112,9 +112,9 @@ Infer:
         order: ''
     - ToCHWImage:
   PostProcess:
-    name: "Topk"
+    name: Topk
     topk: 5
-    class_id_map_file: "ppcls/utils/imagenet1k_label_list.txt"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
   Train:

ppcls/configs/ImageNet/ResNet/ResNet152_vd.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "ResNet152_vd"
+  name: ResNet152_vd
  
 # loss function config for traing/eval process
 Loss:
@@ -31,10 +31,10 @@ Loss:
 
 
 Optimizer:
-  name: "Momentum"
+  name: Momentum
   momentum: 0.9
   lr:
-    name: "Cosine"
+    name: Cosine
     learning_rate: 0.1
   regularizer:
     name: 'L2'
@@ -45,9 +45,9 @@ Optimizer:
 DataLoader:
   Train:
     dataset:
-      name: "ImageNetDataset"
-      image_root: "./dataset/ILSVRC2012/"
-      cls_label_path: "./dataset/ILSVRC2012/train_list.txt"
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/train_list.txt
       transform_ops:
         - RandCropImage:
             size: 224
@@ -63,20 +63,20 @@ DataLoader:
             alpha: 0.2
 
     sampler:
-      name: "DistributedBatchSampler"
+      name: DistributedBatchSampler
       batch_size: 64
       drop_last: False
       shuffle: True
     loader:
-      num_workers: 6
+      num_workers: 4
       use_shared_memory: True
 
   Eval:
     # TOTO: modify to the latest trainer
     dataset: 
-      name: "ImageNetDataset"
-      image_root: "./dataset/ILSVRC2012/"
-      cls_label_path: "./dataset/ILSVRC2012/val_list.txt"
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
       transform_ops:
         - ResizeImage:
             resize_short: 256
@@ -88,16 +88,16 @@ DataLoader:
             std: [0.229, 0.224, 0.225]
             order: ''
     sampler:
-      name: "DistributedBatchSampler"
+      name: DistributedBatchSampler
       batch_size: 64
       drop_last: False
       shuffle: False
     loader:
-      num_workers: 6
+      num_workers: 4
       use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  infer_imgs: docs/images/whl/demo.jpg
   batch_size: 10
   transforms:
     - DecodeImage:
@@ -114,9 +114,9 @@ Infer:
         order: ''
     - ToCHWImage:
   PostProcess:
-    name: "Topk"
+    name: Topk
     topk: 5
-    class_id_map_file: "ppcls/utils/imagenet1k_label_list.txt"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
   Train:

ppcls/configs/ImageNet/ResNet/ResNet18.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "ResNet18"
+  name: ResNet18
  
 # loss function config for traing/eval process
 Loss:
@@ -30,10 +30,10 @@ Loss:
 
 
 Optimizer:
-  name: "Momentum"
+  name: Momentum
   momentum: 0.9
   lr:
-    name: "Piecewise"
+    name: Piecewise
     learning_rate: 0.1
     decay_epochs: [30, 60, 90]
     values: [0.1, 0.01, 0.001, 0.0001]
@@ -46,9 +46,9 @@ Optimizer:
 DataLoader:
   Train:
     dataset:
-      name: "ImageNetDataset"
-      image_root: "./dataset/ILSVRC2012/"
-      cls_label_path: "./dataset/ILSVRC2012/train_list.txt"
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/train_list.txt
       transform_ops:
         - RandCropImage:
             size: 224
@@ -61,20 +61,20 @@ DataLoader:
             order: ''
 
     sampler:
-      name: "DistributedBatchSampler"
+      name: DistributedBatchSampler
       batch_size: 64
       drop_last: False
       shuffle: True
     loader:
-      num_workers: 6
+      num_workers: 4
       use_shared_memory: True
 
   Eval:
     # TOTO: modify to the latest trainer
     dataset: 
-      name: "ImageNetDataset"
-      image_root: "./dataset/ILSVRC2012/"
-      cls_label_path: "./dataset/ILSVRC2012/val_list.txt"
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
       transform_ops:
         - ResizeImage:
             resize_short: 256
@@ -86,16 +86,16 @@ DataLoader:
             std: [0.229, 0.224, 0.225]
             order: ''
     sampler:
-      name: "DistributedBatchSampler"
+      name: DistributedBatchSampler
       batch_size: 64
       drop_last: False
       shuffle: False
     loader:
-      num_workers: 6
+      num_workers: 4
       use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  infer_imgs: docs/images/whl/demo.jpg
   batch_size: 10
   transforms:
     - DecodeImage:
@@ -112,9 +112,9 @@ Infer:
         order: ''
     - ToCHWImage:
   PostProcess:
-    name: "Topk"
+    name: Topk
     topk: 5
-    class_id_map_file: "ppcls/utils/imagenet1k_label_list.txt"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
   Train:

ppcls/configs/ImageNet/ResNet/ResNet18_vd.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "ResNet18_vd"
+  name: ResNet18_vd
  
 # loss function config for traing/eval process
 Loss:
@@ -31,10 +31,10 @@ Loss:
 
 
 Optimizer:
-  name: "Momentum"
+  name: Momentum
   momentum: 0.9
   lr:
-    name: "Cosine"
+    name: Cosine
     learning_rate: 0.1
   regularizer:
     name: 'L2'
@@ -45,9 +45,9 @@ Optimizer:
 DataLoader:
   Train:
     dataset:
-      name: "ImageNetDataset"
-      image_root: "./dataset/ILSVRC2012/"
-      cls_label_path: "./dataset/ILSVRC2012/train_list.txt"
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/train_list.txt
       transform_ops:
         - RandCropImage:
             size: 224
@@ -63,20 +63,20 @@ DataLoader:
             alpha: 0.2
 
     sampler:
-      name: "DistributedBatchSampler"
+      name: DistributedBatchSampler
       batch_size: 64
       drop_last: False
       shuffle: True
     loader:
-      num_workers: 6
+      num_workers: 4
       use_shared_memory: True
 
   Eval:
     # TOTO: modify to the latest trainer
     dataset: 
-      name: "ImageNetDataset"
-      image_root: "./dataset/ILSVRC2012/"
-      cls_label_path: "./dataset/ILSVRC2012/val_list.txt"
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
       transform_ops:
         - ResizeImage:
             resize_short: 256
@@ -88,16 +88,16 @@ DataLoader:
             std: [0.229, 0.224, 0.225]
             order: ''
     sampler:
-      name: "DistributedBatchSampler"
+      name: DistributedBatchSampler
       batch_size: 64
       drop_last: False
       shuffle: False
     loader:
-      num_workers: 6
+      num_workers: 4
       use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  infer_imgs: docs/images/whl/demo.jpg
   batch_size: 10
   transforms:
     - DecodeImage:
@@ -114,9 +114,9 @@ Infer:
         order: ''
     - ToCHWImage:
   PostProcess:
-    name: "Topk"
+    name: Topk
     topk: 5
-    class_id_map_file: "ppcls/utils/imagenet1k_label_list.txt"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
   Train:

ppcls/configs/ImageNet/ResNet/ResNet200_vd.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "ResNet200_vd"
+  name: ResNet200_vd
  
 # loss function config for traing/eval process
 Loss:
@@ -31,10 +31,10 @@ Loss:
 
 
 Optimizer:
-  name: "Momentum"
+  name: Momentum
   momentum: 0.9
   lr:
-    name: "Cosine"
+    name: Cosine
     learning_rate: 0.1
   regularizer:
     name: 'L2'
@@ -45,9 +45,9 @@ Optimizer:
 DataLoader:
   Train:
     dataset:
-      name: "ImageNetDataset"
-      image_root: "./dataset/ILSVRC2012/"
-      cls_label_path: "./dataset/ILSVRC2012/train_list.txt"
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/train_list.txt
       transform_ops:
         - RandCropImage:
             size: 224
@@ -63,20 +63,20 @@ DataLoader:
             alpha: 0.2
 
     sampler:
-      name: "DistributedBatchSampler"
+      name: DistributedBatchSampler
       batch_size: 64
       drop_last: False
       shuffle: True
     loader:
-      num_workers: 6
+      num_workers: 4
       use_shared_memory: True
 
   Eval:
     # TOTO: modify to the latest trainer
     dataset: 
-      name: "ImageNetDataset"
-      image_root: "./dataset/ILSVRC2012/"
-      cls_label_path: "./dataset/ILSVRC2012/val_list.txt"
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
       transform_ops:
         - ResizeImage:
             resize_short: 256
@@ -88,16 +88,16 @@ DataLoader:
             std: [0.229, 0.224, 0.225]
             order: ''
     sampler:
-      name: "DistributedBatchSampler"
+      name: DistributedBatchSampler
       batch_size: 64
       drop_last: False
       shuffle: False
     loader:
-      num_workers: 6
+      num_workers: 4
       use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  infer_imgs: docs/images/whl/demo.jpg
   batch_size: 10
   transforms:
     - DecodeImage:
@@ -114,9 +114,9 @@ Infer:
         order: ''
     - ToCHWImage:
   PostProcess:
-    name: "Topk"
+    name: Topk
     topk: 5
-    class_id_map_file: "ppcls/utils/imagenet1k_label_list.txt"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
   Train:

ppcls/configs/ImageNet/ResNet/ResNet34.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "ResNet34"
+  name: ResNet34
  
 # loss function config for traing/eval process
 Loss:
@@ -30,10 +30,10 @@ Loss:
 
 
 Optimizer:
-  name: "Momentum"
+  name: Momentum
   momentum: 0.9
   lr:
-    name: "Piecewise"
+    name: Piecewise
     learning_rate: 0.1
     decay_epochs: [30, 60, 90]
     values: [0.1, 0.01, 0.001, 0.0001]
@@ -46,9 +46,9 @@ Optimizer:
 DataLoader:
   Train:
     dataset:
-      name: "ImageNetDataset"
-      image_root: "./dataset/ILSVRC2012/"
-      cls_label_path: "./dataset/ILSVRC2012/train_list.txt"
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/train_list.txt
       transform_ops:
         - RandCropImage:
             size: 224
@@ -61,20 +61,20 @@ DataLoader:
             order: ''
 
     sampler:
-      name: "DistributedBatchSampler"
+      name: DistributedBatchSampler
       batch_size: 64
       drop_last: False
       shuffle: True
     loader:
-      num_workers: 6
+      num_workers: 4
       use_shared_memory: True
 
   Eval:
     # TOTO: modify to the latest trainer
     dataset: 
-      name: "ImageNetDataset"
-      image_root: "./dataset/ILSVRC2012/"
-      cls_label_path: "./dataset/ILSVRC2012/val_list.txt"
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
       transform_ops:
         - ResizeImage:
             resize_short: 256
@@ -86,16 +86,16 @@ DataLoader:
             std: [0.229, 0.224, 0.225]
             order: ''
     sampler:
-      name: "DistributedBatchSampler"
+      name: DistributedBatchSampler
       batch_size: 64
       drop_last: False
       shuffle: False
     loader:
-      num_workers: 6
+      num_workers: 4
       use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  infer_imgs: docs/images/whl/demo.jpg
   batch_size: 10
   transforms:
     - DecodeImage:
@@ -112,9 +112,9 @@ Infer:
         order: ''
     - ToCHWImage:
   PostProcess:
-    name: "Topk"
+    name: Topk
     topk: 5
-    class_id_map_file: "ppcls/utils/imagenet1k_label_list.txt"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
   Train:

ppcls/configs/ImageNet/ResNet/ResNet34_vd.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "ResNet34_vd"
+  name: ResNet34_vd
  
 # loss function config for traing/eval process
 Loss:
@@ -31,10 +31,10 @@ Loss:
 
 
 Optimizer:
-  name: "Momentum"
+  name: Momentum
   momentum: 0.9
   lr:
-    name: "Cosine"
+    name: Cosine
     learning_rate: 0.1
   regularizer:
     name: 'L2'
@@ -45,9 +45,9 @@ Optimizer:
 DataLoader:
   Train:
     dataset:
-      name: "ImageNetDataset"
-      image_root: "./dataset/ILSVRC2012/"
-      cls_label_path: "./dataset/ILSVRC2012/train_list.txt"
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/train_list.txt
       transform_ops:
         - RandCropImage:
             size: 224
@@ -63,20 +63,20 @@ DataLoader:
             alpha: 0.2
 
     sampler:
-      name: "DistributedBatchSampler"
+      name: DistributedBatchSampler
       batch_size: 64
       drop_last: False
       shuffle: True
     loader:
-      num_workers: 6
+      num_workers: 4
       use_shared_memory: True
 
   Eval:
     # TOTO: modify to the latest trainer
     dataset: 
-      name: "ImageNetDataset"
-      image_root: "./dataset/ILSVRC2012/"
-      cls_label_path: "./dataset/ILSVRC2012/val_list.txt"
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
       transform_ops:
         - ResizeImage:
             resize_short: 256
@@ -88,16 +88,16 @@ DataLoader:
             std: [0.229, 0.224, 0.225]
             order: ''
     sampler:
-      name: "DistributedBatchSampler"
+      name: DistributedBatchSampler
       batch_size: 64
       drop_last: False
       shuffle: False
     loader:
-      num_workers: 6
+      num_workers: 4
       use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  infer_imgs: docs/images/whl/demo.jpg
   batch_size: 10
   transforms:
     - DecodeImage:
@@ -114,9 +114,9 @@ Infer:
         order: ''
     - ToCHWImage:
   PostProcess:
-    name: "Topk"
+    name: Topk
     topk: 5
-    class_id_map_file: "ppcls/utils/imagenet1k_label_list.txt"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
   Train:

ppcls/configs/ImageNet/ResNet/ResNet50.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "ResNet50"
+  name: ResNet50
  
 # loss function config for traing/eval process
 Loss:
@@ -30,10 +30,10 @@ Loss:
 
 
 Optimizer:
-  name: "Momentum"
+  name: Momentum
   momentum: 0.9
   lr:
-    name: "Piecewise"
+    name: Piecewise
     learning_rate: 0.1
     decay_epochs: [30, 60, 90]
     values: [0.1, 0.01, 0.001, 0.0001]
@@ -46,9 +46,9 @@ Optimizer:
 DataLoader:
   Train:
     dataset:
-      name: "ImageNetDataset"
-      image_root: "./dataset/ILSVRC2012/"
-      cls_label_path: "./dataset/ILSVRC2012/train_list.txt"
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/train_list.txt
       transform_ops:
         - RandCropImage:
             size: 224
@@ -61,20 +61,20 @@ DataLoader:
             order: ''
 
     sampler:
-      name: "DistributedBatchSampler"
+      name: DistributedBatchSampler
       batch_size: 64
       drop_last: False
       shuffle: True
     loader:
-      num_workers: 6
+      num_workers: 4
       use_shared_memory: True
 
   Eval:
     # TOTO: modify to the latest trainer
     dataset: 
-      name: "ImageNetDataset"
-      image_root: "./dataset/ILSVRC2012/"
-      cls_label_path: "./dataset/ILSVRC2012/val_list.txt"
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
       transform_ops:
         - ResizeImage:
             resize_short: 256
@@ -86,16 +86,16 @@ DataLoader:
             std: [0.229, 0.224, 0.225]
             order: ''
     sampler:
-      name: "DistributedBatchSampler"
+      name: DistributedBatchSampler
       batch_size: 64
       drop_last: False
       shuffle: False
     loader:
-      num_workers: 6
+      num_workers: 4
       use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  infer_imgs: docs/images/whl/demo.jpg
   batch_size: 10
   transforms:
     - DecodeImage:
@@ -112,9 +112,9 @@ Infer:
         order: ''
     - ToCHWImage:
   PostProcess:
-    name: "Topk"
+    name: Topk
     topk: 5
-    class_id_map_file: "ppcls/utils/imagenet1k_label_list.txt"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
   Train:

ppcls/configs/ImageNet/ResNet/ResNet50_vd.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "ResNet50_vd"
+  name: ResNet50_vd
  
 # loss function config for traing/eval process
 Loss:
@@ -31,10 +31,10 @@ Loss:
 
 
 Optimizer:
-  name: "Momentum"
+  name: Momentum
   momentum: 0.9
   lr:
-    name: "Cosine"
+    name: Cosine
     learning_rate: 0.1
   regularizer:
     name: 'L2'
@@ -45,9 +45,9 @@ Optimizer:
 DataLoader:
   Train:
     dataset:
-      name: "ImageNetDataset"
-      image_root: "./dataset/ILSVRC2012/"
-      cls_label_path: "./dataset/ILSVRC2012/train_list.txt"
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/train_list.txt
       transform_ops:
         - RandCropImage:
             size: 224
@@ -63,20 +63,20 @@ DataLoader:
             alpha: 0.2
 
     sampler:
-      name: "DistributedBatchSampler"
+      name: DistributedBatchSampler
       batch_size: 64
       drop_last: False
       shuffle: True
     loader:
-      num_workers: 6
+      num_workers: 4
       use_shared_memory: True
 
   Eval:
     # TOTO: modify to the latest trainer
     dataset: 
-      name: "ImageNetDataset"
-      image_root: "./dataset/ILSVRC2012/"
-      cls_label_path: "./dataset/ILSVRC2012/val_list.txt"
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
       transform_ops:
         - ResizeImage:
             resize_short: 256
@@ -88,16 +88,16 @@ DataLoader:
             std: [0.229, 0.224, 0.225]
             order: ''
     sampler:
-      name: "DistributedBatchSampler"
+      name: DistributedBatchSampler
       batch_size: 64
       drop_last: False
       shuffle: False
     loader:
-      num_workers: 6
+      num_workers: 4
       use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  infer_imgs: docs/images/whl/demo.jpg
   batch_size: 10
   transforms:
     - DecodeImage:
@@ -114,9 +114,9 @@ Infer:
         order: ''
     - ToCHWImage:
   PostProcess:
-    name: "Topk"
+    name: Topk
     topk: 5
-    class_id_map_file: "ppcls/utils/imagenet1k_label_list.txt"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
   Train:

ppcls/configs/ImageNet/VGG/VGG11.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "VGG11"
+  name: VGG11
  
 # loss function config for traing/eval process
 Loss:
@@ -44,80 +44,80 @@ Optimizer:
 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: ''
+      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: 1.0/255.0
+            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
+      name: DistributedBatchSampler
+      batch_size: 128
+      drop_last: False
+      shuffle: True
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
   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:
-              resize_short: 256
-          - CropImage:
-              size: 224
-          - NormalizeImage:
-              scale: 0.00392157
-              mean: [0.485, 0.456, 0.406]
-              std: [0.229, 0.224, 0.225]
-              order: ''
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
+      transform_ops:
+        - 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: ''
     sampler:
-        name: DistributedBatchSampler
-        batch_size: 64
-        drop_last: False
-        shuffle: False
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: False
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  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:
+    - 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"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
-    Train:
+  Train:
     - TopkAcc:
         topk: [1, 5]
-    Eval:
+  Eval:
     - TopkAcc:
         topk: [1, 5]

ppcls/configs/ImageNet/VGG/VGG13.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "VGG13"
+  name: VGG13
  
 # loss function config for traing/eval process
 Loss:
@@ -44,80 +44,80 @@ Optimizer:
 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: ''
+      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: 1.0/255.0
+            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
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: True
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
   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:
-              resize_short: 256
-          - CropImage:
-              size: 224
-          - NormalizeImage:
-              scale: 0.00392157
-              mean: [0.485, 0.456, 0.406]
-              std: [0.229, 0.224, 0.225]
-              order: ''
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
+      transform_ops:
+        - 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: ''
     sampler:
-        name: DistributedBatchSampler
-        batch_size: 64
-        drop_last: False
-        shuffle: False
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: False
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  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:
+    - 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"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
-    Train:
+  Train:
     - TopkAcc:
         topk: [1, 5]
-    Eval:
+  Eval:
     - TopkAcc:
         topk: [1, 5]

ppcls/configs/ImageNet/VGG/VGG16.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "VGG16"
+  name: VGG16
  
 # loss function config for traing/eval process
 Loss:
@@ -44,80 +44,80 @@ Optimizer:
 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: ''
+      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: 1.0/255.0
+            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
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: True
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
   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:
-              resize_short: 256
-          - CropImage:
-              size: 224
-          - NormalizeImage:
-              scale: 0.00392157
-              mean: [0.485, 0.456, 0.406]
-              std: [0.229, 0.224, 0.225]
-              order: ''
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
+      transform_ops:
+        - 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: ''
     sampler:
-        name: DistributedBatchSampler
-        batch_size: 64
-        drop_last: False
-        shuffle: False
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: False
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  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:
+    - 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"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
-    Train:
+  Train:
     - TopkAcc:
         topk: [1, 5]
-    Eval:
+  Eval:
     - TopkAcc:
         topk: [1, 5]

ppcls/configs/ImageNet/VGG/VGG19.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 1000
   save_interval: 1
   eval_during_train: True
@@ -13,11 +13,11 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
+  save_inference_dir: ./inference
 
 # model architecture
 Arch:
-  name: "VGG19"
+  name: VGG19
  
 # loss function config for traing/eval process
 Loss:
@@ -44,80 +44,80 @@ Optimizer:
 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: ''
+      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: 1.0/255.0
+            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
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: True
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
   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:
-              resize_short: 256
-          - CropImage:
-              size: 224
-          - NormalizeImage:
-              scale: 0.00392157
-              mean: [0.485, 0.456, 0.406]
-              std: [0.229, 0.224, 0.225]
-              order: ''
+      name: ImageNetDataset
+      image_root: ./dataset/ILSVRC2012/
+      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
+      transform_ops:
+        - 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: ''
     sampler:
-        name: DistributedBatchSampler
-        batch_size: 64
-        drop_last: False
-        shuffle: False
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: False
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  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:
+    - 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"
+    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt
 
 Metric:
-    Train:
+  Train:
     - TopkAcc:
         topk: [1, 5]
-    Eval:
+  Eval:
     - TopkAcc:
         topk: [1, 5]

ppcls/configs/Products/ResNet50_vd_Aliproduct.yaml

@@ -2,8 +2,8 @@
 Global:
   checkpoints: null
   pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+  output_dir: ./output/
+  device: gpu
   class_num: 50030
   save_interval: 1
   eval_during_train: True
@@ -13,23 +13,23 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
-  eval_mode: "classification"
+  save_inference_dir: ./inference
+  eval_mode: classification
 
 # model architecture
 Arch:
-  name: "RecModel"
+  name: RecModel
   Backbone: 
-    name: "ResNet50_vd"
-    pretrained: False
+    name: ResNet50_vd
+    pretrained: True
   BackboneStopLayer:
-    name: "flatten_0"
+    name: flatten_0
   Neck:
-    name: "FC"
+    name: FC
     embedding_size: 2048
     class_num: 512
   Head:
-    name: "FC"  
+    name: FC  
     embedding_size: 512
     class_num: 50030
  
@@ -56,52 +56,52 @@ Optimizer:
 DataLoader:
   Train:
     dataset:
-        name: "ImageNetDataset"
-        image_root: "./dataset/Aliproduct/"
-        cls_label_path: "./dataset/Aliproduct/train_list.txt"
-        transform_ops:
-          - ResizeImage:
-              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: ''
+      name: ImageNetDataset
+      image_root: ./dataset/Aliproduct/
+      cls_label_path: ./dataset/Aliproduct/train_list.txt
+      transform_ops:
+        - ResizeImage:
+            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
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: True
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
   Eval:
     # TOTO: modify to the latest trainer
     dataset: 
-        name: "ImageNetDataset"
-        image_root: "./dataset/Aliproduct/"
-        cls_label_path: "./dataset/Aliproduct/val_list.txt"
-        transform_ops:
-          - ResizeImage:
-              resize_short: 256
-          - CropImage:
-              size: 224
-          - NormalizeImage:
-              scale: 0.00392157
-              mean: [0.485, 0.456, 0.406]
-              std: [0.229, 0.224, 0.225]
-              order: ''
+      name: ImageNetDataset
+      image_root: ./dataset/Aliproduct/
+      cls_label_path: ./dataset/Aliproduct/val_list.txt
+      transform_ops:
+        - ResizeImage:
+            resize_short: 256
+        - CropImage:
+            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
+      name: DistributedBatchSampler
+      batch_size: 64
+      drop_last: False
+      shuffle: False
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
 Metric:
   Train:
     - TopkAcc:
@@ -111,17 +111,17 @@ Metric:
         topk: [1, 5]
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  infer_imgs: docs/images/whl/demo.jpg
   batch_size: 10
   transforms:
-      - DecodeImage:
-          to_rgb: True
-          channel_first: False
-      - ResizeImage:
-          resize_short: 224
-      - NormalizeImage:
-          scale: 1.0/255.0
-          mean: [0.485, 0.456, 0.406]
-          std: [0.229, 0.224, 0.225]
-          order: ''
-      - ToCHWImage:
+    - DecodeImage:
+        to_rgb: True
+        channel_first: False
+    - ResizeImage:
+        resize_short: 224
+    - NormalizeImage:
+        scale: 1.0/255.0
+        mean: [0.485, 0.456, 0.406]
+        std: [0.229, 0.224, 0.225]
+        order: ''
+    - ToCHWImage:

ppcls/configs/Products/ResNet50_vd_Inshop.yaml

 # global configs
 Global:
   checkpoints: null
-  pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+# please download pretrained model via this link:
+# https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/rec/models/pretrain/product_ResNet50_vd_Aliproduct_v1.0_pretrained.pdparams
+  pretrained_model: product_ResNet50_vd_Aliproduct_v1.0_pretrained
+  output_dir: ./output/
+  device: gpu
   class_num: 3997
   save_interval: 10
   eval_during_train: True
@@ -13,29 +15,30 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
-  eval_mode: "retrieval"
+  save_inference_dir: ./inference
+  eval_mode: retrieval
 
 # model architecture
 Arch:
-  name: "RecModel"
+  name: RecModel
+  infer_output_key: features
+  infer_add_softmax: False
+  
   Backbone: 
-    name: "ResNet50_vd"
+    name: ResNet50_vd
     pretrained: False
   BackboneStopLayer:
-    name: "flatten_0"
+    name: flatten_0
   Neck:
-    name: "FC"
+    name: FC
     embedding_size: 2048
     class_num: 512
   Head:
-    name: "ArcMargin"  
+    name: ArcMargin  
     embedding_size: 512
     class_num: 3997
     margin: 0.15
     scale: 30
-  infer_output_key: "features"
-  infer_add_softmax: False
    
 # loss function config for traing/eval process
 Loss:
@@ -67,42 +70,41 @@ Optimizer:
 DataLoader:
   Train:
     dataset:
-        name: "ImageNetDataset"
-        image_root: "./dataset/Inshop/"
-        cls_label_path: "./dataset/Inshop/train_list.txt"
-        transform_ops:
-          - ResizeImage:
-              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: ''
-          - RandomErasing:
-              EPSILON: 0.5
-              sl: 0.02
-              sh: 0.4
-              r1: 0.3
-              mean: [0., 0., 0.]
-
+      name: ImageNetDataset
+      image_root: ./dataset/Inshop/
+      cls_label_path: ./dataset/Inshop/train_list.txt
+      transform_ops:
+        - ResizeImage:
+            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: ''
+        - RandomErasing:
+            EPSILON: 0.5
+            sl: 0.02
+            sh: 0.4
+            r1: 0.3
+            mean: [0., 0., 0.]
     sampler:
-        name: DistributedRandomIdentitySampler
-        batch_size: 64
-        num_instances: 2
-        drop_last: False
-        shuffle: True
+      name: DistributedRandomIdentitySampler
+      batch_size: 64
+      num_instances: 2
+      drop_last: False
+      shuffle: True
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 4
+      use_shared_memory: True
+      
   Eval:
     Query:
-    # TOTO: modify to the latest trainer
       dataset: 
-        name: "ImageNetDataset"
-        image_root: "./dataset/Inshop/"
-        cls_label_path: "./dataset/Inshop/query_list.txt"
+        name: ImageNetDataset
+        image_root: ./dataset/Inshop/
+        cls_label_path: ./dataset/Inshop/query_list.txt
         transform_ops:
           - ResizeImage:
               size: 224
@@ -117,20 +119,19 @@ DataLoader:
         drop_last: False
         shuffle: False
       loader:
-        num_workers: 6
+        num_workers: 4
         use_shared_memory: True
 
     Gallery:
-    # TOTO: modify to the latest trainer
       dataset: 
-        name: "ImageNetDataset"
-        image_root: "./dataset/Inshop/"
-        cls_label_path: "./dataset/Inshop/gallery_list.txt"
+        name: ImageNetDataset
+        image_root: ./dataset/Inshop/
+        cls_label_path: ./dataset/Inshop/gallery_list.txt
         transform_ops:
           - ResizeImage:
               size: 224
           - NormalizeImage:
-              scale: 0.00392157
+              scale: 1.0/255.0
               mean: [0.485, 0.456, 0.406]
               std: [0.229, 0.224, 0.225]
               order: ''
@@ -140,7 +141,7 @@ DataLoader:
         drop_last: False
         shuffle: False
       loader:
-        num_workers: 6
+        num_workers: 4
         use_shared_memory: True
 
 Metric:
@@ -149,17 +150,17 @@ Metric:
         topk: [1, 5]
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  infer_imgs: docs/images/whl/demo.jpg
   batch_size: 10
   transforms:
-      - DecodeImage:
-          to_rgb: True
-          channel_first: False
-      - ResizeImage:
-          resize_short: 224
-      - NormalizeImage:
-          scale: 1.0/255.0
-          mean: [0.485, 0.456, 0.406]
-          std: [0.229, 0.224, 0.225]
-          order: ''
-      - ToCHWImage:
+    - DecodeImage:
+        to_rgb: True
+        channel_first: False
+    - ResizeImage:
+        resize_short: 224
+    - NormalizeImage:
+        scale: 1.0/255.0
+        mean: [0.485, 0.456, 0.406]
+        std: [0.229, 0.224, 0.225]
+        order: ''
+    - ToCHWImage:

ppcls/configs/Products/ResNet50_vd_SOP.yaml

 # global configs
 Global:
   checkpoints: null
-  pretrained_model: null
-  output_dir: "./output/"
-  device: "gpu"
+# please download pretrained model via this link:
+# https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/rec/models/pretrain/product_ResNet50_vd_Aliproduct_v1.0_pretrained.pdparams
+  pretrained_model: product_ResNet50_vd_Aliproduct_v1.0_pretrained
+  output_dir: ./output/
+  device: gpu
   class_num: 11319
   save_interval: 10
   eval_during_train: True
@@ -13,28 +15,28 @@ Global:
   use_visualdl: False
   # used for static mode and model export
   image_shape: [3, 224, 224]
-  save_inference_dir: "./inference"
-  eval_mode: "retrieval"
+  save_inference_dir: ./inference
+  eval_mode: retrieval
 
 # model architecture
 Arch:
-  name: "RecModel"
+  name: RecModel
   Backbone: 
-    name: "ResNet50_vd"
+    name: ResNet50_vd
     pretrained: False
   BackboneStopLayer:
-    name: "flatten_0"
+    name: flatten_0
   Neck:
-    name: "FC"
+    name: FC
     embedding_size: 2048
     class_num: 512
   Head:
-    name: "ArcMargin"  
+    name: ArcMargin  
     embedding_size: 512
     class_num: 11319
     margin: 0.15
     scale: 30
-  infer_output_key: "features"
+  infer_output_key: features
   infer_add_softmax: False
 
 # loss function config for traing/eval process
@@ -67,42 +69,41 @@ Optimizer:
 DataLoader:
   Train:
     dataset:
-        name: "ImageNetDataset"
-        image_root: "./dataset/Stanford_Online_Products/"
-        cls_label_path: "./dataset/Stanford_Online_Products/train_list.txt"
-        transform_ops:
-          - ResizeImage:
-              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: ''
-          - RandomErasing:
-              EPSILON: 0.5
-              sl: 0.02
-              sh: 0.4
-              r1: 0.3
-              mean: [0., 0., 0.]
+      name: ImageNetDataset
+      image_root: ./dataset/Stanford_Online_Products/
+      cls_label_path: ./dataset/Stanford_Online_Products/train_list.txt
+      transform_ops:
+        - ResizeImage:
+            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: ''
+        - RandomErasing:
+            EPSILON: 0.5
+            sl: 0.02
+            sh: 0.4
+            r1: 0.3
+            mean: [0., 0., 0.]
 
     sampler:
-        name: DistributedRandomIdentitySampler
-        batch_size: 64
-        num_instances: 2
-        drop_last: False
-        shuffle: True
+      name: DistributedRandomIdentitySampler
+      batch_size: 64
+      num_instances: 2
+      drop_last: False
+      shuffle: True
     loader:
-        num_workers: 6
-        use_shared_memory: True
+      num_workers: 6
+      use_shared_memory: True
   Eval:
     Query:
-    # TOTO: modify to the latest trainer
       dataset: 
-        name: "ImageNetDataset"
-        image_root: "./dataset/Stanford_Online_Products/"
-        cls_label_path: "./dataset/Stanford_Online_Products/test_list.txt"
+        name: ImageNetDataset
+        image_root: ./dataset/Stanford_Online_Products/
+        cls_label_path: ./dataset/Stanford_Online_Products/test_list.txt
         transform_ops:
           - ResizeImage:
               size: 224
@@ -117,15 +118,14 @@ DataLoader:
         drop_last: False
         shuffle: False
       loader:
-        num_workers: 6
+        num_workers: 4
         use_shared_memory: True
 
     Gallery:
-    # TOTO: modify to the latest trainer
       dataset: 
-        name: "ImageNetDataset"
-        image_root: "./dataset/Stanford_Online_Products/"
-        cls_label_path: "./dataset/Stanford_Online_Products/test_list.txt"
+        name: ImageNetDataset
+        image_root: ./dataset/Stanford_Online_Products/
+        cls_label_path: ./dataset/Stanford_Online_Products/test_list.txt
         transform_ops:
           - ResizeImage:
               size: 224
@@ -140,7 +140,7 @@ DataLoader:
         drop_last: False
         shuffle: False
       loader:
-        num_workers: 6
+        num_workers: 4
         use_shared_memory: True
 
 Metric:
@@ -149,17 +149,17 @@ Metric:
         topk: [1, 5]
 
 Infer:
-  infer_imgs: "docs/images/whl/demo.jpg"
+  infer_imgs: docs/images/whl/demo.jpg
   batch_size: 10
   transforms:
-      - DecodeImage:
-          to_rgb: True
-          channel_first: False
-      - ResizeImage:
-          resize_short: 224
-      - NormalizeImage:
-          scale: 1.0/255.0
-          mean: [0.485, 0.456, 0.406]
-          std: [0.229, 0.224, 0.225]
-          order: ''
-      - ToCHWImage:
+    - DecodeImage:
+        to_rgb: True
+        channel_first: False
+    - ResizeImage:
+        resize_short: 224
+    - NormalizeImage:
+        scale: 1.0/255.0
+        mean: [0.485, 0.456, 0.406]
+        std: [0.229, 0.224, 0.225]
+        order: ''
+    - ToCHWImage: