cityscapes_mask_rcnn_R_50_FPN_coco.yaml

MODEL:
  META_ARCHITECTURE: "GeneralizedRCNN"
  WEIGHTS: "detectron2://COCO-InstanceSegmentation/mask_rcnn_R_50_FPN_3x/137849600/model_final_f10217.pkl"
  BACKBONE:
    NAME: "build_resnet_fpn_backbone"
  RESNETS:
    DEPTH: 50
    OUT_FEATURES: ["res2", "res3", "res4", "res5"]
  FPN:
    IN_FEATURES: ["res2", "res3", "res4", "res5"]
  ANCHOR_GENERATOR:
    SIZES: [[32], [64], [128], [256], [512]]  # One size for each in feature map
    ASPECT_RATIOS: [[0.5, 1.0, 2.0]]  # Three aspect ratios (same for all in feature maps)
  RPN:
    IN_FEATURES: ["p2", "p3", "p4", "p5", "p6"]
    PRE_NMS_TOPK_TRAIN: 2000  # Per FPN level
    PRE_NMS_TOPK_TEST: 1000  # Per FPN level
    # Detectron1 uses 2000 proposals per-batch,
    # (See "modeling/rpn/rpn_outputs.py" for details of this legacy issue)
    # which is approximately 1000 proposals per-image since the default batch size for FPN is 2.
    POST_NMS_TOPK_TRAIN: 1000
    POST_NMS_TOPK_TEST: 1000
  ROI_HEADS:
    NAME: "StandardROIHeads"
    NUM_CLASSES: 8
    IN_FEATURES: ["p2", "p3", "p4", "p5"]
  ROI_BOX_HEAD:
    NAME: "FastRCNNConvFCHead"
    NUM_FC: 2
    POOLER_RESOLUTION: 7
  ROI_MASK_HEAD:
    NAME: "MaskRCNNConvUpsampleHead"
    NUM_CONV: 4
    POOLER_RESOLUTION: 14
    CONTRAST_LOSS:
      LOSS_WEIGHT: 1.6
  MASK_ON: True
DATASETS:
  TRAIN: ("cityscapes_fine_instance_seg_train",)
  TEST: ("cityscapes_fine_instance_seg_val",)
SOLVER:
  IMS_PER_BATCH: 16
  BASE_LR: 0.01
  STEPS: (18000,)
  MAX_ITER: 24000
  IMS_PER_BATCH: 8
TEST:
  EVAL_PERIOD: 8000
INPUT:
  MIN_SIZE_TRAIN: (800, 832, 864, 896, 928, 960, 992, 1024)
  MIN_SIZE_TRAIN_SAMPLING: "choice"
  MIN_SIZE_TEST: 1024
  MAX_SIZE_TRAIN: 2048
  MAX_SIZE_TEST: 2048
VERSION: 2