MODEL:
  META_ARCHITECTURE: "TransformerPureDetector"
  MASK_ON: False
  PIXEL_MEAN: [123.675, 116.280, 103.530]
  PIXEL_STD: [58.395, 57.120, 57.375]
  BACKBONE:
    NAME: "build_swin_backbone"
    FREEZE_AT: -1
  SWIN:
    TYPE: 'tiny'
    DROP_PATH_RATE: 0.2
  TRANSFORMER:
    ENABLED: True
    NUM_FEATURE_LEVELS: 4
    TEMPERATURE: 10000
    ENC_LAYERS: 6
    DEC_LAYERS: 6
    DIM_FEEDFORWARD: 1024
    HIDDEN_DIM: 256
    DROPOUT: 0.0
    NHEADS: 8
    NUM_QUERIES: 100
    ENC_N_POINTS: 4
    DEC_N_POINTS: 4
    NUM_POINTS: 25
    INFERENCE_TH_TEST: 0.4
    LOSS:
      BEZIER_SAMPLE_POINTS: 25
      BEZIER_CLASS_WEIGHT: 1.0
      BEZIER_COORD_WEIGHT: 1.0
      POINT_CLASS_WEIGHT: 1.0
      POINT_COORD_WEIGHT: 1.0
      POINT_TEXT_WEIGHT: 0.5
      BOUNDARY_WEIGHT: 0.5

SOLVER:
  WEIGHT_DECAY: 1e-4
  OPTIMIZER: "ADAMW"
  LR_BACKBONE_NAMES: ['backbone.0']
  LR_LINEAR_PROJ_NAMES: ['reference_points', 'sampling_offsets']
  LR_LINEAR_PROJ_MULT: 1.
  CLIP_GRADIENTS:
    ENABLED: True
    CLIP_TYPE: "full_model"
    CLIP_VALUE: 0.1
    NORM_TYPE: 2.0

INPUT:
  HFLIP_TRAIN: False
  MIN_SIZE_TRAIN: (480, 512, 544, 576, 608, 640, 672, 704, 736, 768, 800, 832, 864, 896)
  MAX_SIZE_TRAIN: 1600
  MIN_SIZE_TEST: 1000
  MAX_SIZE_TEST: 1892
  CROP:
    ENABLED: True
    CROP_INSTANCE: False
    SIZE: [0.1, 0.1]
  FORMAT: "RGB"

DATALOADER:
  NUM_WORKERS: 8

VERSION: 2
SEED: 42