panoptic_deeplab_MBNV2_100_os32_cityscapes_bs16_lr_x2.yaml

MODEL:
  META_ARCHITECTURE: "panoptic_deeplab"
  BN_MOMENTUM: 0.01
  BACKBONE:
    META: "mobilenet_v2"
    NAME: "mobilenet_v2"
    DILATION: (False, False, False)
    PRETRAINED: True
  DECODER:
    IN_CHANNELS: 320
    FEATURE_KEY: "layer_18"
    DECODER_CHANNELS: 256
    ATROUS_RATES: (3, 6, 9)
  PANOPTIC_DEEPLAB:
    LOW_LEVEL_CHANNELS: (96, 32, 24)
    LOW_LEVEL_KEY: ["layer_14", "layer_7", "layer_4"]
    LOW_LEVEL_CHANNELS_PROJECT: (128, 64, 32)
    INSTANCE:
      ENABLE: True
      LOW_LEVEL_CHANNELS_PROJECT: (64, 32, 16)
      DECODER_CHANNELS: 128
      HEAD_CHANNELS: 32
      ASPP_CHANNELS: 256
      NUM_CLASSES: (1, 2)
      CLASS_KEY: ["center", "offset"]
DATASET:
  ROOT: "./datasets/cityscapes"
  DATASET: "cityscapes_panoptic"
  NUM_CLASSES: 19
  TRAIN_SPLIT: 'train'
  TEST_SPLIT: 'val'
  CROP_SIZE: (1025, 2049)
  MIRROR: True
  MIN_SCALE: 0.5
  MAX_SCALE: 2.0
  SCALE_STEP_SIZE: 0.1
  MEAN: (0.485, 0.456, 0.406)
  STD: (0.229, 0.224, 0.225)
  SEMANTIC_ONLY: False
  IGNORE_STUFF_IN_OFFSET: True
  SMALL_INSTANCE_AREA: 4096
  SMALL_INSTANCE_WEIGHT: 3
SOLVER:
  BASE_LR: 0.0001
  WEIGHT_DECAY: 0.0
  WEIGHT_DECAY_NORM: 0.0
  BIAS_LR_FACTOR: 1.0
  WEIGHT_DECAY_BIAS: 0.0
  OPTIMIZER: "adam"
  LR_SCHEDULER_NAME: "WarmupPolyLR"
  WARMUP_ITERS: 0
LOSS:
  SEMANTIC:
    NAME: "hard_pixel_mining"
    IGNORE: 255
    TOP_K_PERCENT: 0.2
    WEIGHT: 1.0
  CENTER:
    NAME: "mse"
    WEIGHT: 200.0
  OFFSET:
    NAME: "l1"
    WEIGHT: 0.01
TRAIN:
  IMS_PER_BATCH: 16
  MAX_ITER: 90000
DEBUG:
  DEBUG: True
  DEBUG_FREQ: 100
TEST:
  EVAL_INSTANCE: True
  EVAL_PANOPTIC: True
POST_PROCESSING:
  CENTER_THRESHOLD: 0.1
  NMS_KERNEL: 7
  TOP_K_INSTANCE: 200
  STUFF_AREA: 2048
OUTPUT_DIR: "./output/panoptic_deeplab_MBNV2_100_os32_cityscapes_bs16_lr_x2"
GPUS: (0, 1, 2, 3, 4, 5, 6, 7)
WORKERS: 1