MobileViTv3_x1_0.yaml 3.7 KB
Newer Older
Y
update  
Yang Nie 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56
# global configs
Global:
  checkpoints: null
  pretrained_model: null
  output_dir: ./output/
  device: gpu
  save_interval: 1
  eval_during_train: True
  eval_interval: 1
  epochs: 300
  print_batch_step: 10
  use_visualdl: False
  # used for static mode and model export
  image_shape: [3, 256, 256]
  save_inference_dir: ./inference
  use_dali: False

# mixed precision training
AMP:
  scale_loss: 65536
  use_dynamic_loss_scaling: True
  # O1: mixed fp16
  level: O1

# model ema
EMA:
  decay: 0.9995

# model architecture
Arch:
  name: MobileViTv3_x1_0
  class_num: 1000
  classifier_dropout: 0.

# loss function config for traing/eval process
Loss:
  Train:
    - CELoss:
        weight: 1.0
        epsilon: 0.1
  Eval:
    - CELoss:
        weight: 1.0

Optimizer:
  name: AdamW
  beta1: 0.9
  beta2: 0.999
  epsilon: 1e-8
  weight_decay: 0.05
  one_dim_param_no_weight_decay: True
  lr:
    # for 8 cards
    name: Cosine
    learning_rate: 0.002
    eta_min: 0.0002
Y
Yang Nie 已提交
57
    warmup_epoch: 16  # 20000 iterations
Y
update  
Yang Nie 已提交
58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80
    warmup_start_lr: 1e-6
    # by_epoch: True
  clip_norm: 10

# 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:
        - DecodeImage:
            to_rgb: True
            channel_first: False
            backend: pil
        - RandCropImage:
            size: 256
            interpolation: bicubic
            backend: pil
            use_log_aspect: True
        - RandFlipImage:
            flip_code: 1
Y
Yang Nie 已提交
81 82
        - RandAugmentV3:
            num_layers: 2
Y
update  
Yang Nie 已提交
83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108
            interpolation: bicubic
        - NormalizeImage:
            scale: 1.0/255.0
            mean: [0.0, 0.0, 0.0]
            std: [1.0, 1.0, 1.0]
            order: ''
        - RandomErasing:
            EPSILON: 0.25
            sl: 0.02
            sh: 1.0/3.0
            r1: 0.3
            attempt: 10
            use_log_aspect: True
            mode: pixel
      batch_transform_ops:
        - OpSampler:
            MixupOperator:
              alpha: 0.2
              prob: 0.5
            CutmixOperator:
              alpha: 1.0
              prob: 0.5
    sampler:
      name: DistributedBatchSampler
      batch_size: 128
      drop_last: False
Y
Yang Nie 已提交
109
      shuffle: True
Y
update  
Yang Nie 已提交
110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174
    loader:
      num_workers: 4
      use_shared_memory: True
  Eval:
    dataset: 
      name: ImageNetDataset
      image_root: ./dataset/ILSVRC2012/
      cls_label_path: ./dataset/ILSVRC2012/val_list.txt
      transform_ops:
        - DecodeImage:
            to_np: False
            channel_first: False
            backend: pil
        - ResizeImage:
            resize_short: 288
            interpolation: bicubic
            backend: pil
        - CropImage:
            size: 256
        - NormalizeImage:
            scale: 1.0/255.0
            mean: [0.0, 0.0, 0.0]
            std: [1.0, 1.0, 1.0]
            order: ''
    sampler:
      name: DistributedBatchSampler
      batch_size: 128
      drop_last: False
      shuffle: False
    loader:
      num_workers: 4
      use_shared_memory: True

Infer:
  infer_imgs: docs/images/inference_deployment/whl_demo.jpg
  batch_size: 10
  transforms:
    - DecodeImage:
        to_np: False
        channel_first: False
        backend: pil
    - ResizeImage:
        resize_short: 288
        interpolation: bicubic
        backend: pil
    - CropImage:
        size: 256
    - NormalizeImage:
        scale: 1.0/255.0
        mean: [0.0, 0.0, 0.0]
        std: [1.0, 1.0, 1.0]
        order: ''
    - ToCHWImage:
  PostProcess:
    name: Topk
    topk: 5
    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt

Metric:
  Train:
    - TopkAcc:
        topk: [1, 5]
  Eval:
    - TopkAcc:
        topk: [1, 5]