upernet_cae_base_ade20k_512x512_160k.yml

_base_: '../_base_/ade20k.yml'

batch_size: 4 # total is 16
iters: 160000

model:
  type: UPerNetCAE
  backbone:
    type: CAE
    img_size: 512
    patch_size: 16
    embed_dim: 768
    depth: 12
    num_heads: 12
    mlp_ratio: 4
    qkv_bias: True
    drop_rate: 0.0
    drop_path_rate: 0.1
    init_values: 0.1
    use_rel_pos_bias: True
    pretrained: https://bj.bcebos.com/paddleseg/dygraph/ade20k/upernet_caebase_ade20k_512x512_160k/pretrained/model.pdparams
  backbone_indices: [3, 5, 7, 11]
  channels: 768
  fpn_channels: 768
  head_channels: 768
  channels_fpn: [768, 768, 768, 768]

train_dataset:
  type: ADE20K
  dataset_root: data/ADEChallengeData2016/
  transforms:
    - type: ResizeStepScaling
      min_scale_factor: 0.5
      max_scale_factor: 2.0
      scale_step_size: 0.0
    - type: RandomPaddingCrop
      crop_size: [512, 512]
    - type: RandomHorizontalFlip
    - type: RandomDistort
    - type: Normalize
      mean: [0.485, 0.456, 0.406]
      std: [0.229, 0.224, 0.225]
  mode: train

val_dataset:
  type: ADE20K
  dataset_root: data/ADEChallengeData2016/
  transforms:
    - type: ResizeByShort
      short_size: 512
    - type: Normalize
      mean: [0.485, 0.456, 0.406]
      std: [0.229, 0.224, 0.225]

test_config:
  is_slide: True
  crop_size: [512, 512]
  stride: [341, 341]

loss:
  types:
    - type: CrossEntropyLoss
  coef: [1, 0.4]

optimizer:
  _inherited_: False
  type: AdamWDL_CAE
  beta1: 0.9
  beta2: 0.999
  weight_decay: 0.0
  layerwise_decay: 0.65

lr_scheduler:
  type: PolynomialDecay
  warmup_iters: 1500
  warmup_start_lr: 0.0
  learning_rate: 0.0002
  end_lr: 0.0
  power: 1.0