Pointcept/configs/waymo/semseg-pt-v3m1-0-base.py

_base_ = ["../_base_/default_runtime.py"]

# misc custom setting
batch_size = 12  # bs: total bs in all gpus
mix_prob = 0.8
empty_cache = False
enable_amp = True

# model settings
model = dict(
    type="DefaultSegmentorV2",
    num_classes=22,
    backbone_out_channels=64,
    backbone=dict(
        type="PT-v3m1",
        in_channels=4,
        order=["z", "z-trans", "hilbert", "hilbert-trans"],
        stride=(2, 2, 2, 2),
        enc_depths=(2, 2, 2, 6, 2),
        enc_channels=(32, 64, 128, 256, 512),
        enc_num_head=(2, 4, 8, 16, 32),
        enc_patch_size=(1024, 1024, 1024, 1024, 1024),
        dec_depths=(2, 2, 2, 2),
        dec_channels=(64, 64, 128, 256),
        dec_num_head=(4, 4, 8, 16),
        dec_patch_size=(1024, 1024, 1024, 1024),
        mlp_ratio=4,
        qkv_bias=True,
        qk_scale=None,
        attn_drop=0.0,
        proj_drop=0.0,
        drop_path=0.3,
        shuffle_orders=True,
        pre_norm=True,
        enable_rpe=False,
        enable_flash=True,
        upcast_attention=False,
        upcast_softmax=False,
        cls_mode=False,
        pdnorm_bn=False,
        pdnorm_ln=False,
        pdnorm_decouple=True,
        pdnorm_adaptive=False,
        pdnorm_affine=True,
        pdnorm_conditions=("nuScenes", "SemanticKITTI", "Waymo"),
    ),
    criteria=[
        dict(type="CrossEntropyLoss", loss_weight=1.0, ignore_index=-1),
        dict(type="LovaszLoss", mode="multiclass", loss_weight=1.0, ignore_index=-1),
    ],
)

# scheduler settings
epoch = 50
eval_epoch = 50
optimizer = dict(type="AdamW", lr=0.002, weight_decay=0.005)
scheduler = dict(
    type="OneCycleLR",
    max_lr=[0.002, 0.0002],
    pct_start=0.04,
    anneal_strategy="cos",
    div_factor=10.0,
    final_div_factor=100.0,
)
param_dicts = [dict(keyword="block", lr=0.0002)]

# dataset settings
dataset_type = "WaymoDataset"
data_root = "data/waymo"
ignore_index = -1
names = [
    "Car",
    "Truck",
    "Bus",
    # Other small vehicles (e.g. pedicab) and large vehicles (e.g. construction vehicles, RV, limo, tram).
    "Other Vehicle",
    "Motorcyclist",
    "Bicyclist",
    "Pedestrian",
    "Sign",
    "Traffic Light",
    # Lamp post, traffic sign pole etc.
    "Pole",
    # Construction cone/pole.
    "Construction Cone",
    "Bicycle",
    "Motorcycle",
    "Building",
    # Bushes, tree branches, tall grasses, flowers etc.
    "Vegetation",
    "Tree Trunk",
    # Curb on the edge of roads. This does not include road boundaries if there’s no curb.
    "Curb",
    # Surface a vehicle could drive on. This includes the driveway connecting
    # parking lot and road over a section of sidewalk.
    "Road",
    # Marking on the road that’s specifically for defining lanes such as
    # single/double white/yellow lines.
    "Lane Marker",
    # Marking on the road other than lane markers, bumps, cateyes, railtracks etc.
    "Other Ground",
    # Most horizontal surface that’s not drivable, e.g. grassy hill, pedestrian walkway stairs etc.
    "Walkable",
    # Nicely paved walkable surface when pedestrians most likely to walk on.
    "Sidewalk",
]

data = dict(
    num_classes=22,
    ignore_index=ignore_index,
    names=names,
    train=dict(
        type=dataset_type,
        split="training",
        data_root=data_root,
        transform=[
            # dict(type="RandomDropout", dropout_ratio=0.2, dropout_application_ratio=0.2),
            # dict(type="RandomRotateTargetAngle", angle=(1/2, 1, 3/2), center=[0, 0, 0], axis="z", p=0.75),
            dict(type="RandomRotate", angle=[-1, 1], axis="z", center=[0, 0, 0], p=0.5),
            # dict(type="RandomRotate", angle=[-1/6, 1/6], axis="x", p=0.5),
            # dict(type="RandomRotate", angle=[-1/6, 1/6], axis="y", p=0.5),
            dict(type="PointClip", point_cloud_range=(-75.2, -75.2, -4, 75.2, 75.2, 2)),
            dict(type="RandomScale", scale=[0.9, 1.1]),
            # dict(type="RandomShift", shift=[0.2, 0.2, 0.2]),
            dict(type="RandomFlip", p=0.5),
            dict(type="RandomJitter", sigma=0.005, clip=0.02),
            # dict(type="ElasticDistortion", distortion_params=[[0.2, 0.4], [0.8, 1.6]]),
            dict(
                type="GridSample",
                grid_size=0.05,
                hash_type="fnv",
                mode="train",
                keys=("coord", "strength", "segment"),
                return_grid_coord=True,
            ),
            # dict(type="SphereCrop", point_max=1000000, mode="random"),
            # dict(type="CenterShift", apply_z=False),
            dict(type="ToTensor"),
            dict(
                type="Collect",
                keys=("coord", "grid_coord", "segment"),
                feat_keys=("coord", "strength"),
            ),
        ],
        test_mode=False,
        ignore_index=ignore_index,
    ),
    val=dict(
        type=dataset_type,
        split="validation",
        data_root=data_root,
        transform=[
            dict(type="PointClip", point_cloud_range=(-75.2, -75.2, -4, 75.2, 75.2, 2)),
            dict(
                type="GridSample",
                grid_size=0.05,
                hash_type="fnv",
                mode="train",
                keys=("coord", "strength", "segment"),
                return_grid_coord=True,
            ),
            dict(type="ToTensor"),
            dict(
                type="Collect",
                keys=("coord", "grid_coord", "segment"),
                feat_keys=("coord", "strength"),
            ),
        ],
        test_mode=False,
        ignore_index=ignore_index,
    ),
    test=dict(
        type=dataset_type,
        split="validation",
        data_root=data_root,
        transform=[
            dict(type="PointClip", point_cloud_range=(-75.2, -75.2, -4, 75.2, 75.2, 2)),
            dict(type="Copy", keys_dict={"segment": "origin_segment"}),
            dict(
                type="GridSample",
                grid_size=0.025,
                hash_type="fnv",
                mode="train",
                keys=("coord", "strength", "segment"),
                return_inverse=True,
            ),
        ],
        test_mode=True,
        test_cfg=dict(
            voxelize=dict(
                type="GridSample",
                grid_size=0.05,
                hash_type="fnv",
                mode="test",
                return_grid_coord=True,
                keys=("coord", "strength"),
            ),
            crop=None,
            post_transform=[
                dict(type="ToTensor"),
                dict(
                    type="Collect",
                    keys=("coord", "grid_coord", "index"),
                    feat_keys=("coord", "strength"),
                ),
            ],
            aug_transform=[
                [
                    dict(
                        type="RandomRotateTargetAngle",
                        angle=[0],
                        axis="z",
                        center=[0, 0, 0],
                        p=1,
                    )
                ],
                [
                    dict(
                        type="RandomRotateTargetAngle",
                        angle=[1 / 2],
                        axis="z",
                        center=[0, 0, 0],
                        p=1,
                    )
                ],
                [
                    dict(
                        type="RandomRotateTargetAngle",
                        angle=[1],
                        axis="z",
                        center=[0, 0, 0],
                        p=1,
                    )
                ],
                [
                    dict(
                        type="RandomRotateTargetAngle",
                        angle=[3 / 2],
                        axis="z",
                        center=[0, 0, 0],
                        p=1,
                    )
                ],
            ],
        ),
        ignore_index=ignore_index,
    ),
)