swh_kitti.yaml

CLASS_NAMES: ['Car', 'Pedestrian', 'Cyclist']

DATA_CONFIG:
    _BASE_CONFIG_: cfgs/dataset_configs/kitti_dataset.yaml
    DATA_AUGMENTOR:
        DISABLE_AUG_LIST: ['placeholder']
        AUG_CONFIG_LIST:
            - NAME: gt_sampling
              USE_ROAD_PLANE: False  # TODO: swith this to True
              DB_INFO_PATH:
                  - kitti_dbinfos_train.pkl
              PREPARE: {
                 filter_by_min_points: ['Car:5', 'Pedestrian:5', 'Cyclist:5'],
                 filter_by_difficulty: [-1],
              }

              SAMPLE_GROUPS: ['Car:15','Pedestrian:10', 'Cyclist:10']
              NUM_POINT_FEATURES: 4
              DATABASE_WITH_FAKELIDAR: False
              REMOVE_EXTRA_WIDTH: [0.0, 0.0, 0.0]
              LIMIT_WHOLE_SCENE: False

            - NAME: random_world_flip
              ALONG_AXIS_LIST: ['x']

            - NAME: random_world_rotation
              WORLD_ROT_ANGLE: [-0.78539816, 0.78539816]

            - NAME: random_world_scaling
              WORLD_SCALE_RANGE: [0.95, 1.05]

    DATA_SPLIT: {
        'train': train,
        'test': test
    }

    INFO_PATH: {
        'train': [ kitti_infos_train.pkl ],
        'test': [ kitti_infos_test.pkl ],
    }

MODEL:
    NAME: SECONDWithHead

    VFE:
        NAME: MeanVFE

    BACKBONE_3D:
        NAME: VoxelBackBone8x
        NORM_TYPE: batch

    MAP_TO_BEV:
        NAME: HeightCompression
        NUM_BEV_FEATURES: 256

    BACKBONE_2D:
        NAME: BaseBEVBackbone

        LAYER_NUMS: [5, 5]
        LAYER_STRIDES: [1, 2]
        NUM_FILTERS: [128, 256]
        UPSAMPLE_STRIDES: [1, 2]
        NUM_UPSAMPLE_FILTERS: [256, 256]

    DENSE_HEAD:
        NAME: AnchorHeadSingle
        CLASS_AGNOSTIC: False

        USE_DIRECTION_CLASSIFIER: True
        DIR_OFFSET: 0.78539
        DIR_LIMIT_OFFSET: 0.0
        NUM_DIR_BINS: 2

        ANCHOR_GENERATOR_CONFIG: [
            {
                'class_name': 'Car',
                'anchor_sizes': [[3.9, 1.6, 1.56]],
                'anchor_rotations': [0, 1.57],
                'anchor_bottom_heights': [-1.78],
                'align_center': False,
                'feature_map_stride': 8,
                'matched_threshold': 0.6,
                'unmatched_threshold': 0.45
            },
            {
                'class_name': 'Pedestrian',
                'anchor_sizes': [[0.8, 0.6, 1.73]],
                'anchor_rotations': [0, 1.57],
                'anchor_bottom_heights': [-0.6],
                'align_center': False,
                'feature_map_stride': 8,
                'matched_threshold': 0.5,
                'unmatched_threshold': 0.35
            },
            {
                'class_name': 'Cyclist',
                'anchor_sizes': [[1.76, 0.6, 1.73]],
                'anchor_rotations': [0, 1.57],
                'anchor_bottom_heights': [-0.6],
                'align_center': False,
                'feature_map_stride': 8,
                'matched_threshold': 0.5,
                'unmatched_threshold': 0.35
            }
        ]

        TARGET_ASSIGNER_CONFIG:
            NAME: AxisAlignedTargetAssigner
            POS_FRACTION: -1.0
            SAMPLE_SIZE: 512
            NORM_BY_NUM_EXAMPLES: False
            MATCH_HEIGHT: False
            BOX_CODER: ResidualCoder

        LOSS_CONFIG:
            LOSS_WEIGHTS: {
                'cls_weight': 1.0,
                'loc_weight': 2.0,
                'dir_weight': 0.2,
                'code_weights': [1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0]
            }

    POINT_HEAD:
        ENABLE: True
        POINT_TASKS: ['cls', 'part', 'reg']  # 'cls', 'part', 'reg'
        NAME: SECONDPointHead
        CLASS_AGNOSTIC: True

        POINT_SRC: [ 'x_conv3', 'x_conv4' ]  # ['x_conv??'] or ['unet_out']
        POINT_SRC_DIM: [ 64, 64 ]
        POINT_SRC_UNIFIED_DIM: 64
        POINT_REGRESS_TO: center  # or box, none

        POINT_CLOUD_RANGE: [ 0, -40, -3, 70.4, 40, 1 ]
        VOXEL_SIZE: [ 0.05, 0.05, 0.1 ]

        CLS_FC: [ 128 ]
        PART_FC: [ 128 ]
        REG_FC: [ 128 ]

        TARGET_CONFIG:
            GT_EXTRA_WIDTH: [ 0.2, 0.2, 0.2 ]
            BOX_CODER: PointResidualCoder
            BOX_CODER_CONFIG: {
                'use_mean_size': True,
                'mean_size': [
                    [ 3.9, 1.6, 1.56 ],
                    [ 0.8, 0.6, 1.73 ],
                    [ 1.76, 0.6, 1.73 ]
                ]
            }

        LOSS_CONFIG:
            LOSS_REG: WeightedSmoothL1Loss
            LOSS_WEIGHTS: {
                'point_cls_weight': 1.0,
                'point_part_weight': 1.0,
                'point_box_weight': 1.0,
                'code_weights': [1.0, 1.0, 1.0]  # regress to center only need 3
            }

    ROI_HEAD:
        NAME: SECONDAttnHead
        CLASS_AGNOSTIC: True

        POINT_CLOUD_RANGE: [ 0, -40, -3, 70.4, 40, 1 ]
        VOXEL_SIZE: [ 0.05, 0.05, 0.1 ]
        N_CLASS: 3

        SHARED_FC: [ 256, 256 ]
        CLS_FC: [ 256, 256 ]
        REG_FC: [ 256, 256 ]
        DP_RATIO: 0.3

        ROI_POINT_POOL:
            POOL_SRC: [ 'x_conv4', 'x_conv3', 'x_conv1' ]
            N_POINTS_PER_SRC: [ 64, 128, 256 ]
            POOL_SRC_DIM: [ 64, 64, 16 ]
            POOL_EXTRA_WIDTH: [ 0.5, 0.5, 0.5 ]
            MAP_TO_ROI_CANONICAL: True

        ATTENTION:
            NUM_TRANSFORMERS: 4  # applicable if not use sequential attn
            DIM: 128
            POSITION_ENCODING_MODE: center_and_corners_diff  # Options: center_diff, center_and_corners_diff
            # the following is only applicable if using sequential attention
            SEQUENTIAL: True
            N_REPETITION: 3

        NMS_CONFIG:
            TRAIN:
                NMS_TYPE: nms_gpu
                MULTI_CLASSES_NMS: False
                NMS_PRE_MAXSIZE: 9000
                NMS_POST_MAXSIZE: 512
                NMS_THRESH: 0.8
            TEST:
                NMS_TYPE: nms_gpu
                MULTI_CLASSES_NMS: False
                NMS_PRE_MAXSIZE: 1024
                NMS_POST_MAXSIZE: 100
                NMS_THRESH: 0.7

        TARGET_CONFIG:
            BOX_CODER: ResidualCoder
            ROI_PER_IMAGE: 128  # TODO: to increase
            FG_RATIO: 0.5

            SAMPLE_ROI_BY_EACH_CLASS: True
            CLS_SCORE_TYPE: roi_iou

            CLS_FG_THRESH: 0.75
            CLS_BG_THRESH: 0.25
            CLS_BG_THRESH_LO: 0.1
            HARD_BG_RATIO: 0.8

            REG_FG_THRESH: 0.55

        LOSS_CONFIG:
            CLS_LOSS: BinaryCrossEntropy
            REG_LOSS: smooth-l1
            CORNER_LOSS_REGULARIZATION: True
            LOSS_WEIGHTS: {
                'rcnn_cls_weight': 1.0,
                'rcnn_reg_weight': 1.0,
                'rcnn_corner_weight': 1.0,
                'code_weights': [ 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0 ],
            }

    POST_PROCESSING:
        RECALL_THRESH_LIST: [ 0.3, 0.5, 0.7 ]
        SCORE_THRESH: 0.1
        OUTPUT_RAW_SCORE: False

        EVAL_METRIC: kitti

        NMS_CONFIG:
            MULTI_CLASSES_NMS: False
            NMS_TYPE: nms_gpu
            NMS_THRESH: 0.1
            NMS_PRE_MAXSIZE: 4096
            NMS_POST_MAXSIZE: 500

OPTIMIZATION:
    BATCH_SIZE_PER_GPU: 2
    NUM_EPOCHS: 100

    OPTIMIZER: adam_onecycle
    LR: 0.01
    WEIGHT_DECAY: 0.01
    MOMENTUM: 0.9

    MOMS: [0.95, 0.85]
    PCT_START: 0.4
    DIV_FACTOR: 10
    DECAY_STEP_LIST: [35, 45]
    LR_DECAY: 0.1
    LR_CLIP: 0.0000001

    LR_WARMUP: False
    WARMUP_EPOCH: 1

    GRAD_NORM_CLIP: 10