Merge pull request #800 from cuicheng01/develop_reg

Add products configs

ppcls/arch/gears/fc.py

@@ -30,6 +30,6 @@ class FC(nn.Layer):
         self.fc = paddle.nn.Linear(
             self.embedding_size, self.class_num, weight_attr=weight_attr)
 
-    def forward(self, input):
+    def forward(self, input, label=None):
         out = self.fc(input)
         return out

ppcls/configs/Products/ResNet50_vd_Aliproduct.yaml

+# global configs
+Global:
+  checkpoints: null
+  pretrained_model: null
+  output_dir: "./output/"
+  device: "gpu"
+  class_num: 50030
+  save_interval: 10
+  eval_during_train: False
+  eval_interval: 1
+  epochs: 120
+  print_batch_step: 10
+  use_visualdl: False
+  # used for static mode and model export
+  image_shape: [3, 224, 224]
+  save_inference_dir: "./inference"
+  eval_mode: "classification"
+
+# model architecture
+Arch:
+  name: "RecModel"
+  Backbone: 
+    name: "ResNet50_vd"
+    pretrained: False
+  BackboneStopLayer:
+    name: "flatten_0"
+  Neck:
+    name: "FC"
+    embedding_size: 2048
+    class_num: 512
+  Head:
+    name: "FC"  
+    embedding_size: 512
+    class_num: 50030
+ 
+# loss function config for traing/eval process
+Loss:
+  Train:
+    - CELoss:
+        weight: 1.0
+  Eval:
+    - CELoss:
+        weight: 1.0
+
+Optimizer:
+  name: Momentum
+  momentum: 0.9
+  lr:
+    name: Cosine
+    learning_rate: 0.05
+  regularizer:
+    name: 'L2'
+    coeff: 0.00007
+
+# data loader for train and eval
+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: ''
+
+    sampler:
+        name: DistributedBatchSampler
+        batch_size: 64
+        drop_last: False
+        shuffle: True
+    loader:
+        num_workers: 6
+        use_shared_memory: False
+  Eval:
+    # TOTO: modify to the latest trainer
+    dataset: 
+        name: "ImageNetDataset"
+        image_root: "./dataset/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
+    loader:
+        num_workers: 6
+        use_shared_memory: False
+Metric:
+  Train:
+    - TopkAcc:
+        topk: [1, 5]
+  Eval:
+    - TopkAcc:
+        topk: [1, 5]
+
+Infer:
+  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:

ppcls/configs/Products/ResNet50_vd_Inshop.yaml

+# global configs
+Global:
+  checkpoints: null
+  pretrained_model: null
+  output_dir: "./output/"
+  device: "gpu"
+  class_num: 3997
+  save_interval: 10
+  eval_during_train: False
+  eval_interval: 1
+  epochs: 120
+  print_batch_step: 10
+  use_visualdl: False
+  # used for static mode and model export
+  image_shape: [3, 224, 224]
+  save_inference_dir: "./inference"
+  eval_mode: "retrieval"
+
+# model architecture
+Arch:
+  name: "RecModel"
+  Backbone: 
+    name: "ResNet50_vd"
+    pretrained: False
+  BackboneStopLayer:
+    name: "flatten_0"
+  Neck:
+    name: "FC"
+    embedding_size: 2048
+    class_num: 512
+  Head:
+    name: "ArcMargin"  
+    embedding_size: 512
+    class_num: 3997
+    margin: 0.15
+    scale: 30
+ 
+# loss function config for traing/eval process
+Loss:
+  Train:
+    - CELoss:
+        weight: 1.0
+    - TripletLossV2:
+        weight: 1.0
+        margin: 0.5
+  Eval:
+    - CELoss:
+        weight: 1.0
+
+Optimizer:
+  name: Momentum
+  momentum: 0.9
+  lr:
+    name: MultiStepDecay
+    learning_rate: 0.01
+    milestones: [30, 60, 70, 80, 90, 100]
+    gamma: 0.5
+    verbose: False
+    last_epoch: -1
+  regularizer:
+    name: 'L2'
+    coeff: 0.0005
+
+# data loader for train and eval
+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.]
+
+    sampler:
+        name: DistributedRandomIdentitySampler
+        batch_size: 64
+        num_instances: 2
+        drop_last: False
+        shuffle: True
+    loader:
+        num_workers: 6
+        use_shared_memory: False
+  Eval:
+    Query:
+    # TOTO: modify to the latest trainer
+      dataset: 
+        name: "ImageNetDataset"
+        image_root: "./dataset/Inshop/"
+        cls_label_path: "./dataset/Inshop/query_list.txt"
+        transform_ops:
+          - ResizeImage:
+              size: 224
+          - NormalizeImage:
+              scale: 0.00392157
+              mean: [0.485, 0.456, 0.406]
+              std: [0.229, 0.224, 0.225]
+              order: ''
+      sampler:
+        name: DistributedBatchSampler
+        batch_size: 64
+        drop_last: False
+        shuffle: False
+      loader:
+        num_workers: 6
+        use_shared_memory: False
+
+    Gallery:
+    # TOTO: modify to the latest trainer
+      dataset: 
+        name: "ImageNetDataset"
+        image_root: "./dataset/Inshop/"
+        cls_label_path: "./dataset/Inshop/gallery_list.txt"
+        transform_ops:
+          - ResizeImage:
+              size: 224
+          - NormalizeImage:
+              scale: 0.00392157
+              mean: [0.485, 0.456, 0.406]
+              std: [0.229, 0.224, 0.225]
+              order: ''
+      sampler:
+        name: DistributedBatchSampler
+        batch_size: 64
+        drop_last: False
+        shuffle: False
+      loader:
+        num_workers: 6
+        use_shared_memory: False
+
+Metric:
+  Eval:
+    - Recallk:
+        topk: [1, 5]
+
+Infer:
+  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:

ppcls/configs/Products/ResNet50_vd_SOP.yaml

+# global configs
+Global:
+  checkpoints: null
+  pretrained_model: null
+  output_dir: "./output/"
+  device: "gpu"
+  class_num: 11319
+  save_interval: 10
+  eval_during_train: False
+  eval_interval: 1
+  epochs: 120
+  print_batch_step: 10
+  use_visualdl: False
+  # used for static mode and model export
+  image_shape: [3, 224, 224]
+  save_inference_dir: "./inference"
+  eval_mode: "retrieval"
+
+# model architecture
+Arch:
+  name: "RecModel"
+  Backbone: 
+    name: "ResNet50_vd"
+    pretrained: False
+  BackboneStopLayer:
+    name: "flatten_0"
+  Neck:
+    name: "FC"
+    embedding_size: 2048
+    class_num: 512
+  Head:
+    name: "ArcMargin"  
+    embedding_size: 512
+    class_num: 11319
+    margin: 0.15
+    scale: 30
+ 
+# loss function config for traing/eval process
+Loss:
+  Train:
+    - CELoss:
+        weight: 1.0
+    - TripletLossV2:
+        weight: 1.0
+        margin: 0.5
+  Eval:
+    - CELoss:
+        weight: 1.0
+
+Optimizer:
+  name: Momentum
+  momentum: 0.9
+  lr:
+    name: MultiStepDecay
+    learning_rate: 0.01
+    milestones: [30, 60, 70, 80, 90, 100]
+    gamma: 0.5
+    verbose: False
+    last_epoch: -1
+  regularizer:
+    name: 'L2'
+    coeff: 0.0005
+
+# data loader for train and eval
+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.]
+
+    sampler:
+        name: DistributedRandomIdentitySampler
+        batch_size: 64
+        num_instances: 2
+        drop_last: False
+        shuffle: True
+    loader:
+        num_workers: 6
+        use_shared_memory: False
+  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"
+        transform_ops:
+          - ResizeImage:
+              size: 224
+          - NormalizeImage:
+              scale: 0.00392157
+              mean: [0.485, 0.456, 0.406]
+              std: [0.229, 0.224, 0.225]
+              order: ''
+      sampler:
+        name: DistributedBatchSampler
+        batch_size: 32
+        drop_last: False
+        shuffle: False
+      loader:
+        num_workers: 6
+        use_shared_memory: False
+
+    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"
+        transform_ops:
+          - ResizeImage:
+              size: 224
+          - NormalizeImage:
+              scale: 0.00392157
+              mean: [0.485, 0.456, 0.406]
+              std: [0.229, 0.224, 0.225]
+              order: ''
+      sampler:
+        name: DistributedBatchSampler
+        batch_size: 32
+        drop_last: False
+        shuffle: False
+      loader:
+        num_workers: 6
+        use_shared_memory: False
+
+Metric:
+  Eval:
+    - Recallk:
+        topk: [1, 5]
+
+Infer:
+  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: