_base_ = ["../_base_/default_runtime.py"] # misc custom setting batch_size = 24 # bs: total bs in all gpus num_worker = 48 mix_prob = 0.8 empty_cache = False enable_amp = True find_unused_parameters = True # trainer train = dict( type="MultiDatasetTrainer", ) # model settings model = dict( type="PPT-v1m1", backbone=dict( type="SpUNet-v1m3", in_channels=6, num_classes=0, base_channels=32, context_channels=256, channels=(32, 64, 128, 256, 256, 128, 96, 96), layers=(2, 3, 4, 6, 2, 2, 2, 2), cls_mode=False, conditions=("ScanNet", "S3DIS", "Structured3D"), zero_init=False, norm_decouple=True, norm_adaptive=True, norm_affine=True, ), criteria=[dict(type="CrossEntropyLoss", loss_weight=1.0, ignore_index=-1)], backbone_out_channels=96, context_channels=256, conditions=("Structured3D", "ScanNet", "S3DIS"), template="[x]", clip_model="ViT-B/16", class_name=( "wall", "floor", "cabinet", "bed", "chair", "sofa", "table", "door", "window", "bookshelf", "bookcase", "picture", "counter", "desk", "shelves", "curtain", "dresser", "pillow", "mirror", "ceiling", "refrigerator", "television", "shower curtain", "nightstand", "toilet", "sink", "lamp", "bathtub", "garbagebin", "board", "beam", "column", "clutter", "otherstructure", "otherfurniture", "otherprop", ), valid_index=( ( 0, 1, 2, 3, 4, 5, 6, 7, 8, 11, 13, 14, 15, 16, 17, 18, 19, 20, 21, 23, 25, 26, 33, 34, 35, ), (0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 11, 12, 13, 15, 20, 22, 24, 25, 27, 34), (0, 1, 4, 5, 6, 7, 8, 10, 19, 29, 30, 31, 32), ), backbone_mode=False, ) # scheduler settings epoch = 100 optimizer = dict(type="SGD", lr=0.05, momentum=0.9, weight_decay=0.0001, nesterov=True) scheduler = dict( type="OneCycleLR", max_lr=optimizer["lr"], pct_start=0.05, anneal_strategy="cos", div_factor=10.0, final_div_factor=10000.0, ) # param_dicts = [dict(keyword="modulation", lr=0.005)] # dataset settings data = dict( num_classes=13, ignore_index=-1, names=[ "ceiling", "floor", "wall", "beam", "column", "window", "door", "table", "chair", "sofa", "bookcase", "board", "clutter", ], train=dict( type="ConcatDataset", datasets=[ # Structured3D dict( type="Structured3DDataset", split="train", data_root="data/structured3d", transform=[ dict(type="CenterShift", apply_z=True), 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 / 64, 1 / 64], axis="x", p=0.5), dict(type="RandomRotate", angle=[-1 / 64, 1 / 64], axis="y", p=0.5), 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="ChromaticAutoContrast", p=0.2, blend_factor=None), dict(type="ChromaticTranslation", p=0.95, ratio=0.05), dict(type="ChromaticJitter", p=0.95, std=0.05), # dict(type="HueSaturationTranslation", hue_max=0.2, saturation_max=0.2), # dict(type="RandomColorDrop", p=0.2, color_augment=0.0), dict( type="GridSample", grid_size=0.02, hash_type="fnv", mode="train", return_grid_coord=True, ), dict(type="SphereCrop", sample_rate=0.8, mode="random"), dict(type="CenterShift", apply_z=False), dict(type="NormalizeColor"), dict(type="ShufflePoint"), dict(type="Add", keys_dict={"condition": "Structured3D"}), dict(type="ToTensor"), dict( type="Collect", keys=("coord", "grid_coord", "segment", "condition"), feat_keys=("color", "normal"), ), ], test_mode=False, loop=4, # sampling weight ), # ScanNet dict( type="ScanNetDataset", split="train", data_root="data/scannet", transform=[ dict(type="CenterShift", apply_z=True), 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 / 64, 1 / 64], axis="x", p=0.5), dict(type="RandomRotate", angle=[-1 / 64, 1 / 64], axis="y", p=0.5), 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="ChromaticAutoContrast", p=0.2, blend_factor=None), dict(type="ChromaticTranslation", p=0.95, ratio=0.05), dict(type="ChromaticJitter", p=0.95, std=0.05), # dict(type="HueSaturationTranslation", hue_max=0.2, saturation_max=0.2), # dict(type="RandomColorDrop", p=0.2, color_augment=0.0), dict( type="GridSample", grid_size=0.02, hash_type="fnv", mode="train", return_grid_coord=True, ), dict(type="SphereCrop", point_max=100000, mode="random"), dict(type="CenterShift", apply_z=False), dict(type="NormalizeColor"), dict(type="ShufflePoint"), dict(type="Add", keys_dict={"condition": "ScanNet"}), dict(type="ToTensor"), dict( type="Collect", keys=("coord", "grid_coord", "segment", "condition"), feat_keys=("color", "normal"), ), ], test_mode=False, loop=2, # sampling weight ), # S3DIS dict( type="S3DISDataset", split=("Area_1", "Area_2", "Area_3", "Area_4", "Area_6"), data_root="data/s3dis", transform=[ dict(type="CenterShift", apply_z=True), 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 / 64, 1 / 64], axis="x", p=0.5), dict(type="RandomRotate", angle=[-1 / 64, 1 / 64], axis="y", p=0.5), 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="ChromaticAutoContrast", p=0.2, blend_factor=None), dict(type="ChromaticTranslation", p=0.95, ratio=0.05), dict(type="ChromaticJitter", p=0.95, std=0.05), # dict(type="HueSaturationTranslation", hue_max=0.2, saturation_max=0.2), # dict(type="RandomColorDrop", p=0.2, color_augment=0.0), dict( type="GridSample", grid_size=0.02, hash_type="fnv", mode="train", return_grid_coord=True, ), dict(type="SphereCrop", sample_rate=0.6, mode="random"), dict(type="CenterShift", apply_z=False), dict(type="NormalizeColor"), dict(type="ShufflePoint"), dict(type="Add", keys_dict={"condition": "S3DIS"}), dict(type="ToTensor"), dict( type="Collect", keys=("coord", "grid_coord", "segment", "condition"), feat_keys=("color", "normal"), ), ], test_mode=False, loop=1, # sampling weight ), ], ), val=dict( type="S3DISDataset", split="Area_5", data_root="data/s3dis", transform=[ dict(type="CenterShift", apply_z=True), dict( type="GridSample", grid_size=0.02, hash_type="fnv", mode="train", return_grid_coord=True, ), # dict(type="SphereCrop", point_max=1000000, mode="center"), dict(type="CenterShift", apply_z=False), dict(type="NormalizeColor"), dict(type="ToTensor"), dict(type="Add", keys_dict={"condition": "S3DIS"}), dict( type="Collect", keys=("coord", "grid_coord", "segment", "condition"), feat_keys=("color", "normal"), ), ], test_mode=False, ), test=dict( type="S3DISDataset", split="Area_5", data_root="data/s3dis", transform=[ dict(type="CenterShift", apply_z=True), dict(type="NormalizeColor"), ], test_mode=True, test_cfg=dict( voxelize=dict( type="GridSample", grid_size=0.02, hash_type="fnv", mode="test", return_grid_coord=True, keys=("coord", "color", "normal"), ), crop=None, post_transform=[ dict(type="CenterShift", apply_z=False), dict(type="Add", keys_dict={"condition": "S3DIS"}), dict(type="ToTensor"), dict( type="Collect", keys=("coord", "grid_coord", "index", "condition"), feat_keys=("color", "normal"), ), ], 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, ) ], [ dict( type="RandomRotateTargetAngle", angle=[0], axis="z", center=[0, 0, 0], p=1, ), dict(type="RandomScale", scale=[0.95, 0.95]), ], [ dict( type="RandomRotateTargetAngle", angle=[1 / 2], axis="z", center=[0, 0, 0], p=1, ), dict(type="RandomScale", scale=[0.95, 0.95]), ], [ dict( type="RandomRotateTargetAngle", angle=[1], axis="z", center=[0, 0, 0], p=1, ), dict(type="RandomScale", scale=[0.95, 0.95]), ], [ dict( type="RandomRotateTargetAngle", angle=[3 / 2], axis="z", center=[0, 0, 0], p=1, ), dict(type="RandomScale", scale=[0.95, 0.95]), ], [ dict( type="RandomRotateTargetAngle", angle=[0], axis="z", center=[0, 0, 0], p=1, ), dict(type="RandomScale", scale=[1.05, 1.05]), ], [ dict( type="RandomRotateTargetAngle", angle=[1 / 2], axis="z", center=[0, 0, 0], p=1, ), dict(type="RandomScale", scale=[1.05, 1.05]), ], [ dict( type="RandomRotateTargetAngle", angle=[1], axis="z", center=[0, 0, 0], p=1, ), dict(type="RandomScale", scale=[1.05, 1.05]), ], [ dict( type="RandomRotateTargetAngle", angle=[3 / 2], axis="z", center=[0, 0, 0], p=1, ), dict(type="RandomScale", scale=[1.05, 1.05]), ], [dict(type="RandomFlip", p=1)], ], ), ), )