strong_baseline_m1.yaml

# global configs
Global:
  checkpoints: null
  pretrained_model: null
  output_dir: "./output/"
  device: "gpu"
  save_interval: 40
  eval_during_train: True
  eval_interval: 10
  epochs: 120
  print_batch_step: 20
  use_visualdl: False
  eval_mode: "retrieval"
  retrieval_feature_from: "features" # 'backbone' or 'features'
  # used for static mode and model export
  image_shape: [3, 256, 128]
  save_inference_dir: "./inference"

# model architecture
Arch:
  name: "RecModel"
  infer_output_key: "features"
  infer_add_softmax: False
  Backbone:
    name: "ResNet50_last_stage_stride1"
    pretrained: True
    stem_act: null
  BackboneStopLayer:
    name: "flatten"
  Neck:
    name: BNNeck
    num_features: &feat_dim 2048
    weight_attr:
      initializer:
        name: Constant
        value: 1.0
    bias_attr:
      initializer:
        name: Constant
        value: 0.0
      learning_rate: 1.0e-20 # NOTE: Temporarily set lr small enough to freeze the bias to zero
  Head:
    name: "FC"
    embedding_size: *feat_dim
    class_num: 751
    weight_attr:
      initializer:
        name: Normal
        std: 0.001
    bias_attr: False

# loss function config for traing/eval process
Loss:
  Train:
    - CELoss:
        weight: 1.0
        epsilon: 0.1
    - TripletLossV2:
        weight: 1.0
        margin: 0.3
        normalize_feature: False
        feature_from: "backbone"
  Eval:
    - CELoss:
        weight: 1.0

Optimizer:
  name: Adam
  lr:
    name: Piecewise
    decay_epochs: [30, 60]
    values: [0.00035, 0.000035, 0.0000035]
    warmup_epoch: 10
    warmup_start_lr: 0.0000035
    by_epoch: True
    last_epoch: 0
  regularizer:
    name: 'L2'
    coeff: 0.0005

# data loader for train and eval
DataLoader:
  Train:
    dataset:
        name: "Market1501"
        image_root: "./dataset/"
        cls_label_path: "bounding_box_train"
        backend: "pil"
        transform_ops:
          - ResizeImage:
              size: [128, 256]
              return_numpy: False
              backend: "pil"
          - RandFlipImage:
              flip_code: 1
          - Pad:
              padding: 10
          - RandCropImageV2:
              size: [128, 256]
          - ToTensor:
          - Normalize:
              mean: [0.485, 0.456, 0.406]
              std: [0.229, 0.224, 0.225]
          - RandomErasing:
              EPSILON: 0.5
              sl: 0.02
              sh: 0.4
              r1: 0.3
              mean: [0.485, 0.456, 0.406]
    sampler:
        name: DistributedRandomIdentitySampler
        batch_size: 64
        num_instances: 4
        drop_last: False
        shuffle: True
    loader:
        num_workers: 4
        use_shared_memory: True
  Eval:
    Query:
      dataset:
        name: "Market1501"
        image_root: "./dataset/"
        cls_label_path: "query"
        backend: "pil"
        transform_ops:
          - ResizeImage:
              size: [128, 256]
              return_numpy: False
              backend: "pil"
          - ToTensor:
          - Normalize:
              mean: [0.485, 0.456, 0.406]
              std: [0.229, 0.224, 0.225]
      sampler:
        name: DistributedBatchSampler
        batch_size: 128
        drop_last: False
        shuffle: False
      loader:
        num_workers: 4
        use_shared_memory: True

    Gallery:
      dataset:
        name: "Market1501"
        image_root: "./dataset/"
        cls_label_path: "bounding_box_test"
        backend: "pil"
        transform_ops:
          - ResizeImage:
              size: [128, 256]
              return_numpy: False
              backend: "pil"
          - ToTensor:
          - Normalize:
              mean: [0.485, 0.456, 0.406]
              std: [0.229, 0.224, 0.225]
      sampler:
        name: DistributedBatchSampler
        batch_size: 128
        drop_last: False
        shuffle: False
      loader:
        num_workers: 4
        use_shared_memory: True

Metric:
  Eval:
    - Recallk:
        topk: [1, 5]
    - mAP: {}