2023-03-26 11:13:34,846 INFO **********************Start logging********************** 2023-03-26 11:13:34,846 INFO CUDA_VISIBLE_DEVICES=ALL 2023-03-26 11:13:34,847 INFO total_batch_size: 16 2023-03-26 11:13:34,847 INFO cfg_file cfgs/sunrgbd_models/CAGroup3D.yaml 2023-03-26 11:13:34,848 INFO batch_size 16 2023-03-26 11:13:34,849 INFO epochs 1 2023-03-26 11:13:34,850 INFO workers 4 2023-03-26 11:13:34,851 INFO extra_tag cagroup3d-win10-sunrgbd 2023-03-26 11:13:34,851 INFO ckpt None 2023-03-26 11:13:34,852 INFO pretrained_model None 2023-03-26 11:13:34,853 INFO launcher pytorch 2023-03-26 11:13:34,854 INFO tcp_port 18888 2023-03-26 11:13:34,854 INFO sync_bn False 2023-03-26 11:13:34,855 INFO fix_random_seed True 2023-03-26 11:13:34,856 INFO ckpt_save_interval 1 2023-03-26 11:13:34,856 INFO max_ckpt_save_num 30 2023-03-26 11:13:34,857 INFO merge_all_iters_to_one_epoch False 2023-03-26 11:13:34,858 INFO set_cfgs None 2023-03-26 11:13:34,859 INFO max_waiting_mins 0 2023-03-26 11:13:34,859 INFO start_epoch 0 2023-03-26 11:13:34,860 INFO num_epochs_to_eval 0 2023-03-26 11:13:34,860 INFO save_to_file False 2023-03-26 11:13:34,861 INFO cfg.ROOT_DIR: C:\PINKAMENA\CITYU\CS5182\proj\CAGroup3D 2023-03-26 11:13:34,862 INFO cfg.LOCAL_RANK: 0 2023-03-26 11:13:34,863 INFO cfg.CLASS_NAMES: ['bed', 'table', 'sofa', 'chair', 'toilet', 'desk', 'dresser', 'night_stand', 'bookshelf', 'bathtub'] 2023-03-26 11:13:34,864 INFO cfg.DATA_CONFIG = edict() 2023-03-26 11:13:34,865 INFO cfg.DATA_CONFIG.DATASET: SunrgbdDataset 2023-03-26 11:13:34,866 INFO cfg.DATA_CONFIG.DATA_PATH: ../data/sunrgbd_data/sunrgbd 2023-03-26 11:13:34,866 INFO cfg.DATA_CONFIG.PROCESSED_DATA_TAG: sunrgbd_processed_data_v0_5_0 2023-03-26 11:13:34,868 INFO cfg.DATA_CONFIG.POINT_CLOUD_RANGE: [-40, -40, -10, 40, 40, 10] 2023-03-26 11:13:34,869 INFO cfg.DATA_CONFIG.DATA_SPLIT = edict() 2023-03-26 11:13:34,869 INFO cfg.DATA_CONFIG.DATA_SPLIT.train: train 2023-03-26 11:13:34,870 INFO cfg.DATA_CONFIG.DATA_SPLIT.test: val 2023-03-26 11:13:34,870 INFO cfg.DATA_CONFIG.REPEAT = edict() 2023-03-26 11:13:34,871 INFO cfg.DATA_CONFIG.REPEAT.train: 4 2023-03-26 11:13:34,872 INFO cfg.DATA_CONFIG.REPEAT.test: 1 2023-03-26 11:13:34,873 INFO cfg.DATA_CONFIG.INFO_PATH = edict() 2023-03-26 11:13:34,874 INFO cfg.DATA_CONFIG.INFO_PATH.train: ['sunrgbd_infos_train.pkl'] 2023-03-26 11:13:34,875 INFO cfg.DATA_CONFIG.INFO_PATH.test: ['sunrgbd_infos_val.pkl'] 2023-03-26 11:13:34,876 INFO cfg.DATA_CONFIG.GET_ITEM_LIST: ['points'] 2023-03-26 11:13:34,877 INFO cfg.DATA_CONFIG.FILTER_EMPTY_BOXES_FOR_TRAIN: True 2023-03-26 11:13:34,877 INFO cfg.DATA_CONFIG.DATA_AUGMENTOR_TRAIN = edict() 2023-03-26 11:13:34,878 INFO cfg.DATA_CONFIG.DATA_AUGMENTOR_TRAIN.DISABLE_AUG_LIST: ['placeholder'] 2023-03-26 11:13:34,879 INFO cfg.DATA_CONFIG.DATA_AUGMENTOR_TRAIN.AUG_CONFIG_LIST: [{'NAME': 'indoor_point_sample', 'num_points': 100000}, {'NAME': 'random_world_flip', 'ALONG_AXIS_LIST': ['y']}, {'NAME': 'random_world_rotation_mmdet3d', 'WORLD_ROT_ANGLE': [-0.523599, 0.523599]}, {'NAME': 'random_world_scaling', 'WORLD_SCALE_RANGE': [0.85, 1.15]}, {'NAME': 'random_world_translation', 'ALONG_AXIS_LIST': ['x', 'y', 'z'], 'NOISE_TRANSLATE_STD': 0.1}] 2023-03-26 11:13:34,881 INFO cfg.DATA_CONFIG.DATA_AUGMENTOR_TEST = edict() 2023-03-26 11:13:34,882 INFO cfg.DATA_CONFIG.DATA_AUGMENTOR_TEST.DISABLE_AUG_LIST: ['placeholder'] 2023-03-26 11:13:34,884 INFO cfg.DATA_CONFIG.DATA_AUGMENTOR_TEST.AUG_CONFIG_LIST: [{'NAME': 'indoor_point_sample', 'num_points': 100000}] 2023-03-26 11:13:34,884 INFO cfg.DATA_CONFIG.DATA_AUGMENTOR = edict() 2023-03-26 11:13:34,885 INFO cfg.DATA_CONFIG.DATA_AUGMENTOR.DISABLE_AUG_LIST: ['placeholder'] 2023-03-26 11:13:34,886 INFO cfg.DATA_CONFIG.DATA_AUGMENTOR.AUG_CONFIG_LIST: [{'NAME': 'indoor_point_sample', 'num_points': 50000}] 2023-03-26 11:13:34,887 INFO cfg.DATA_CONFIG.POINT_FEATURE_ENCODING = edict() 2023-03-26 11:13:34,888 INFO cfg.DATA_CONFIG.POINT_FEATURE_ENCODING.encoding_type: absolute_coordinates_encoding 2023-03-26 11:13:34,889 INFO cfg.DATA_CONFIG.POINT_FEATURE_ENCODING.used_feature_list: ['x', 'y', 'z', 'r', 'g', 'b'] 2023-03-26 11:13:34,890 INFO cfg.DATA_CONFIG.POINT_FEATURE_ENCODING.src_feature_list: ['x', 'y', 'z', 'r', 'g', 'b'] 2023-03-26 11:13:34,892 INFO cfg.DATA_CONFIG.DATA_PROCESSOR: [{'NAME': 'mask_points_and_boxes_outside_range', 'REMOVE_OUTSIDE_BOXES': False}] 2023-03-26 11:13:34,893 INFO cfg.DATA_CONFIG._BASE_CONFIG_: cfgs/dataset_configs/sunrgbd_dataset.yaml 2023-03-26 11:13:34,894 INFO cfg.VOXEL_SIZE: 0.02 2023-03-26 11:13:34,894 INFO cfg.N_CLASSES: 10 2023-03-26 11:13:34,894 INFO cfg.SEMANTIC_THR: 0.15 2023-03-26 11:13:34,895 INFO cfg.MODEL = edict() 2023-03-26 11:13:34,896 INFO cfg.MODEL.NAME: CAGroup3D 2023-03-26 11:13:34,896 INFO cfg.MODEL.VOXEL_SIZE: 0.02 2023-03-26 11:13:34,897 INFO cfg.MODEL.SEMANTIC_MIN_THR: 0.05 2023-03-26 11:13:34,898 INFO cfg.MODEL.SEMANTIC_ITER_VALUE: 0.02 2023-03-26 11:13:34,899 INFO cfg.MODEL.SEMANTIC_THR: 0.15 2023-03-26 11:13:34,899 INFO cfg.MODEL.BACKBONE_3D = edict() 2023-03-26 11:13:34,900 INFO cfg.MODEL.BACKBONE_3D.NAME: BiResNet 2023-03-26 11:13:34,900 INFO cfg.MODEL.BACKBONE_3D.IN_CHANNELS: 3 2023-03-26 11:13:34,900 INFO cfg.MODEL.BACKBONE_3D.OUT_CHANNELS: 64 2023-03-26 11:13:34,901 INFO cfg.MODEL.DENSE_HEAD = edict() 2023-03-26 11:13:34,902 INFO cfg.MODEL.DENSE_HEAD.NAME: CAGroup3DHead 2023-03-26 11:13:34,902 INFO cfg.MODEL.DENSE_HEAD.IN_CHANNELS: [64, 128, 256, 512] 2023-03-26 11:13:34,902 INFO cfg.MODEL.DENSE_HEAD.OUT_CHANNELS: 64 2023-03-26 11:13:34,903 INFO cfg.MODEL.DENSE_HEAD.SEMANTIC_THR: 0.15 2023-03-26 11:13:34,903 INFO cfg.MODEL.DENSE_HEAD.VOXEL_SIZE: 0.02 2023-03-26 11:13:34,904 INFO cfg.MODEL.DENSE_HEAD.N_CLASSES: 10 2023-03-26 11:13:34,904 INFO cfg.MODEL.DENSE_HEAD.N_REG_OUTS: 8 2023-03-26 11:13:34,905 INFO cfg.MODEL.DENSE_HEAD.CLS_KERNEL: 9 2023-03-26 11:13:34,906 INFO cfg.MODEL.DENSE_HEAD.WITH_YAW: True 2023-03-26 11:13:34,907 INFO cfg.MODEL.DENSE_HEAD.USE_SEM_SCORE: False 2023-03-26 11:13:34,907 INFO cfg.MODEL.DENSE_HEAD.EXPAND_RATIO: 3 2023-03-26 11:13:34,908 INFO cfg.MODEL.DENSE_HEAD.ASSIGNER = edict() 2023-03-26 11:13:34,909 INFO cfg.MODEL.DENSE_HEAD.ASSIGNER.NAME: CAGroup3DAssigner 2023-03-26 11:13:34,910 INFO cfg.MODEL.DENSE_HEAD.ASSIGNER.LIMIT: 27 2023-03-26 11:13:34,910 INFO cfg.MODEL.DENSE_HEAD.ASSIGNER.TOPK: 18 2023-03-26 11:13:34,911 INFO cfg.MODEL.DENSE_HEAD.ASSIGNER.N_SCALES: 4 2023-03-26 11:13:34,911 INFO cfg.MODEL.DENSE_HEAD.LOSS_OFFSET = edict() 2023-03-26 11:13:34,912 INFO cfg.MODEL.DENSE_HEAD.LOSS_OFFSET.NAME: SmoothL1Loss 2023-03-26 11:13:34,912 INFO cfg.MODEL.DENSE_HEAD.LOSS_OFFSET.BETA: 0.04 2023-03-26 11:13:34,913 INFO cfg.MODEL.DENSE_HEAD.LOSS_OFFSET.REDUCTION: sum 2023-03-26 11:13:34,913 INFO cfg.MODEL.DENSE_HEAD.LOSS_OFFSET.LOSS_WEIGHT: 0.2 2023-03-26 11:13:34,914 INFO cfg.MODEL.DENSE_HEAD.LOSS_BBOX = edict() 2023-03-26 11:13:34,914 INFO cfg.MODEL.DENSE_HEAD.LOSS_BBOX.NAME: IoU3DLoss 2023-03-26 11:13:34,915 INFO cfg.MODEL.DENSE_HEAD.LOSS_BBOX.WITH_YAW: True 2023-03-26 11:13:34,916 INFO cfg.MODEL.DENSE_HEAD.LOSS_BBOX.LOSS_WEIGHT: 1.0 2023-03-26 11:13:34,916 INFO cfg.MODEL.DENSE_HEAD.NMS_CONFIG = edict() 2023-03-26 11:13:34,917 INFO cfg.MODEL.DENSE_HEAD.NMS_CONFIG.SCORE_THR: 0.01 2023-03-26 11:13:34,917 INFO cfg.MODEL.DENSE_HEAD.NMS_CONFIG.NMS_PRE: 1000 2023-03-26 11:13:34,917 INFO cfg.MODEL.DENSE_HEAD.NMS_CONFIG.IOU_THR: 0.5 2023-03-26 11:13:34,918 INFO cfg.MODEL.ROI_HEAD = edict() 2023-03-26 11:13:34,918 INFO cfg.MODEL.ROI_HEAD.NAME: CAGroup3DRoIHead 2023-03-26 11:13:34,919 INFO cfg.MODEL.ROI_HEAD.NUM_CLASSES: 10 2023-03-26 11:13:34,919 INFO cfg.MODEL.ROI_HEAD.MIDDLE_FEATURE_SOURCE: [3] 2023-03-26 11:13:34,920 INFO cfg.MODEL.ROI_HEAD.GRID_SIZE: 7 2023-03-26 11:13:34,921 INFO cfg.MODEL.ROI_HEAD.VOXEL_SIZE: 0.02 2023-03-26 11:13:34,921 INFO cfg.MODEL.ROI_HEAD.COORD_KEY: 2 2023-03-26 11:13:34,922 INFO cfg.MODEL.ROI_HEAD.MLPS: [[64, 128, 128]] 2023-03-26 11:13:34,923 INFO cfg.MODEL.ROI_HEAD.CODE_SIZE: 7 2023-03-26 11:13:34,924 INFO cfg.MODEL.ROI_HEAD.ENCODE_SINCOS: True 2023-03-26 11:13:34,925 INFO cfg.MODEL.ROI_HEAD.ROI_PER_IMAGE: 128 2023-03-26 11:13:34,926 INFO cfg.MODEL.ROI_HEAD.ROI_FG_RATIO: 0.9 2023-03-26 11:13:34,926 INFO cfg.MODEL.ROI_HEAD.REG_FG_THRESH: 0.3 2023-03-26 11:13:34,926 INFO cfg.MODEL.ROI_HEAD.ROI_CONV_KERNEL: 5 2023-03-26 11:13:34,927 INFO cfg.MODEL.ROI_HEAD.ENLARGE_RATIO: False 2023-03-26 11:13:34,927 INFO cfg.MODEL.ROI_HEAD.USE_IOU_LOSS: True 2023-03-26 11:13:34,928 INFO cfg.MODEL.ROI_HEAD.USE_GRID_OFFSET: False 2023-03-26 11:13:34,928 INFO cfg.MODEL.ROI_HEAD.USE_SIMPLE_POOLING: True 2023-03-26 11:13:34,928 INFO cfg.MODEL.ROI_HEAD.USE_CENTER_POOLING: True 2023-03-26 11:13:34,929 INFO cfg.MODEL.ROI_HEAD.LOSS_WEIGHTS = edict() 2023-03-26 11:13:34,929 INFO cfg.MODEL.ROI_HEAD.LOSS_WEIGHTS.RCNN_CLS_WEIGHT: 1.0 2023-03-26 11:13:34,930 INFO cfg.MODEL.ROI_HEAD.LOSS_WEIGHTS.RCNN_REG_WEIGHT: 0.5 2023-03-26 11:13:34,930 INFO cfg.MODEL.ROI_HEAD.LOSS_WEIGHTS.RCNN_IOU_WEIGHT: 1.0 2023-03-26 11:13:34,931 INFO cfg.MODEL.ROI_HEAD.LOSS_WEIGHTS.CODE_WEIGHT: [1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0] 2023-03-26 11:13:34,932 INFO cfg.MODEL.POST_PROCESSING = edict() 2023-03-26 11:13:34,933 INFO cfg.MODEL.POST_PROCESSING.RECALL_THRESH_LIST: [0.25, 0.5] 2023-03-26 11:13:34,934 INFO cfg.MODEL.POST_PROCESSING.EVAL_METRIC: scannet 2023-03-26 11:13:34,934 INFO cfg.OPTIMIZATION = edict() 2023-03-26 11:13:34,935 INFO cfg.OPTIMIZATION.BATCH_SIZE_PER_GPU: 16 2023-03-26 11:13:34,935 INFO cfg.OPTIMIZATION.NUM_EPOCHS: 1 2023-03-26 11:13:34,936 INFO cfg.OPTIMIZATION.OPTIMIZER: adamW 2023-03-26 11:13:34,936 INFO cfg.OPTIMIZATION.LR: 0.001 2023-03-26 11:13:34,937 INFO cfg.OPTIMIZATION.WEIGHT_DECAY: 0.0001 2023-03-26 11:13:34,937 INFO cfg.OPTIMIZATION.DECAY_STEP_LIST: [8, 11] 2023-03-26 11:13:34,938 INFO cfg.OPTIMIZATION.LR_DECAY: 0.1 2023-03-26 11:13:34,938 INFO cfg.OPTIMIZATION.GRAD_NORM_CLIP: 10 2023-03-26 11:13:34,939 INFO cfg.OPTIMIZATION.PCT_START: 0.4 2023-03-26 11:13:34,940 INFO cfg.OPTIMIZATION.DIV_FACTOR: 10 2023-03-26 11:13:34,941 INFO cfg.OPTIMIZATION.LR_CLIP: 1e-07 2023-03-26 11:13:34,941 INFO cfg.OPTIMIZATION.LR_WARMUP: False 2023-03-26 11:13:34,942 INFO cfg.OPTIMIZATION.WARMUP_EPOCH: 1 2023-03-26 11:13:34,942 INFO cfg.TAG: CAGroup3D 2023-03-26 11:13:34,942 INFO cfg.EXP_GROUP_PATH: sunrgbd_models 2023-03-26 11:13:35,054 INFO Loading SUNRGBD dataset 2023-03-26 11:13:35,212 INFO Total samples for SUNRGBD dataset: 5285 2023-03-26 11:13:36,687 INFO DistributedDataParallel( (module): CAGroup3D( (vfe): None (backbone_3d): BiResNet( (conv1): Sequential( (0): MinkowskiConvolution(in=3, out=64, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiReLU() (3): MinkowskiConvolution(in=64, out=64, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1]) (4): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (5): MinkowskiReLU() ) (relu): MinkowskiReLU() (layer1): Sequential( (0): BasicBlock( (conv1): MinkowskiConvolution(in=64, out=64, kernel_size=[3, 3, 3], stride=[2, 2, 2], dilation=[1, 1, 1]) (norm1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): MinkowskiConvolution(in=64, out=64, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1]) (norm2): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): MinkowskiReLU() (downsample): Sequential( (0): MinkowskiConvolution(in=64, out=64, kernel_size=[1, 1, 1], stride=[2, 2, 2], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) (1): BasicBlock( (conv1): MinkowskiConvolution(in=64, out=64, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1]) (norm1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): MinkowskiConvolution(in=64, out=64, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1]) (norm2): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): MinkowskiReLU() ) ) (layer2): Sequential( (0): BasicBlock( (conv1): MinkowskiConvolution(in=64, out=128, kernel_size=[3, 3, 3], stride=[2, 2, 2], dilation=[1, 1, 1]) (norm1): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): MinkowskiConvolution(in=128, out=128, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1]) (norm2): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): MinkowskiReLU() (downsample): Sequential( (0): MinkowskiConvolution(in=64, out=128, kernel_size=[1, 1, 1], stride=[2, 2, 2], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) (1): BasicBlock( (conv1): MinkowskiConvolution(in=128, out=128, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1]) (norm1): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): MinkowskiConvolution(in=128, out=128, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1]) (norm2): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): MinkowskiReLU() ) ) (layer3): Sequential( (0): BasicBlock( (conv1): MinkowskiConvolution(in=128, out=256, kernel_size=[3, 3, 3], stride=[2, 2, 2], dilation=[1, 1, 1]) (norm1): MinkowskiBatchNorm(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): MinkowskiConvolution(in=256, out=256, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1]) (norm2): MinkowskiBatchNorm(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): MinkowskiReLU() (downsample): Sequential( (0): MinkowskiConvolution(in=128, out=256, kernel_size=[1, 1, 1], stride=[2, 2, 2], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) (1): BasicBlock( (conv1): MinkowskiConvolution(in=256, out=256, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1]) (norm1): MinkowskiBatchNorm(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): MinkowskiConvolution(in=256, out=256, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1]) (norm2): MinkowskiBatchNorm(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): MinkowskiReLU() ) ) (layer4): Sequential( (0): BasicBlock( (conv1): MinkowskiConvolution(in=256, out=512, kernel_size=[3, 3, 3], stride=[2, 2, 2], dilation=[1, 1, 1]) (norm1): MinkowskiBatchNorm(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): MinkowskiConvolution(in=512, out=512, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1]) (norm2): MinkowskiBatchNorm(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): MinkowskiReLU() (downsample): Sequential( (0): MinkowskiConvolution(in=256, out=512, kernel_size=[1, 1, 1], stride=[2, 2, 2], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) (1): BasicBlock( (conv1): MinkowskiConvolution(in=512, out=512, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1]) (norm1): MinkowskiBatchNorm(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): MinkowskiConvolution(in=512, out=512, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1]) (norm2): MinkowskiBatchNorm(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): MinkowskiReLU() ) ) (compression3): Sequential( (0): MinkowskiConvolution(in=256, out=128, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (compression4): Sequential( (0): MinkowskiConvolution(in=512, out=128, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (down3): Sequential( (0): MinkowskiConvolution(in=128, out=256, kernel_size=[3, 3, 3], stride=[2, 2, 2], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (down4): Sequential( (0): MinkowskiConvolution(in=128, out=256, kernel_size=[3, 3, 3], stride=[2, 2, 2], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiReLU() (3): MinkowskiConvolution(in=256, out=512, kernel_size=[3, 3, 3], stride=[2, 2, 2], dilation=[1, 1, 1]) (4): MinkowskiBatchNorm(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (layer3_): Sequential( (0): BasicBlock( (conv1): MinkowskiConvolution(in=128, out=128, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1]) (norm1): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): MinkowskiConvolution(in=128, out=128, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1]) (norm2): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): MinkowskiReLU() ) (1): BasicBlock( (conv1): MinkowskiConvolution(in=128, out=128, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1]) (norm1): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): MinkowskiConvolution(in=128, out=128, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1]) (norm2): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): MinkowskiReLU() ) ) (layer4_): Sequential( (0): BasicBlock( (conv1): MinkowskiConvolution(in=128, out=128, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1]) (norm1): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): MinkowskiConvolution(in=128, out=128, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1]) (norm2): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): MinkowskiReLU() ) (1): BasicBlock( (conv1): MinkowskiConvolution(in=128, out=128, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1]) (norm1): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): MinkowskiConvolution(in=128, out=128, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1]) (norm2): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): MinkowskiReLU() ) ) (layer5_): Sequential( (0): Bottleneck( (conv1): MinkowskiConvolution(in=128, out=128, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1]) (norm1): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): MinkowskiConvolution(in=128, out=128, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1]) (norm2): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv3): MinkowskiConvolution(in=128, out=256, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1]) (norm3): MinkowskiBatchNorm(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): MinkowskiReLU() (downsample): Sequential( (0): MinkowskiConvolution(in=128, out=256, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) (layer5): Sequential( (0): Bottleneck( (conv1): MinkowskiConvolution(in=512, out=512, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1]) (norm1): MinkowskiBatchNorm(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): MinkowskiConvolution(in=512, out=512, kernel_size=[3, 3, 3], stride=[2, 2, 2], dilation=[1, 1, 1]) (norm2): MinkowskiBatchNorm(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv3): MinkowskiConvolution(in=512, out=1024, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1]) (norm3): MinkowskiBatchNorm(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): MinkowskiReLU() (downsample): Sequential( (0): MinkowskiConvolution(in=512, out=1024, kernel_size=[1, 1, 1], stride=[2, 2, 2], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) (spp): DAPPM( (scale1): Sequential( (0): MinkowskiAvgPooling(kernel_size=[5, 5, 5], stride=[2, 2, 2], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiReLU() (3): MinkowskiConvolution(in=1024, out=128, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1]) ) (scale2): Sequential( (0): MinkowskiAvgPooling(kernel_size=[9, 9, 9], stride=[4, 4, 4], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiReLU() (3): MinkowskiConvolution(in=1024, out=128, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1]) ) (scale3): Sequential( (0): MinkowskiAvgPooling(kernel_size=[17, 17, 17], stride=[8, 8, 8], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiReLU() (3): MinkowskiConvolution(in=1024, out=128, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1]) ) (scale4): Sequential( (0): MinkowskiAvgPooling(kernel_size=[33, 33, 33], stride=[16, 16, 16], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiReLU() (3): MinkowskiConvolution(in=1024, out=128, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1]) ) (scale0): Sequential( (0): MinkowskiBatchNorm(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (1): MinkowskiReLU() (2): MinkowskiConvolution(in=1024, out=128, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1]) ) (process1): Sequential( (0): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (1): MinkowskiReLU() (2): MinkowskiConvolution(in=128, out=128, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1]) ) (process2): Sequential( (0): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (1): MinkowskiReLU() (2): MinkowskiConvolution(in=128, out=128, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1]) ) (process3): Sequential( (0): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (1): MinkowskiReLU() (2): MinkowskiConvolution(in=128, out=128, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1]) ) (process4): Sequential( (0): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (1): MinkowskiReLU() (2): MinkowskiConvolution(in=128, out=128, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1]) ) (compression): Sequential( (0): MinkowskiBatchNorm(640, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (1): MinkowskiReLU() (2): MinkowskiConvolution(in=640, out=256, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1]) ) (shortcut): Sequential( (0): MinkowskiBatchNorm(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (1): MinkowskiReLU() (2): MinkowskiConvolution(in=1024, out=256, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1]) ) ) (out): Sequential( (0): MinkowskiConvolutionTranspose(in=256, out=256, kernel_size=[2, 2, 2], stride=[2, 2, 2], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiReLU() (3): MinkowskiConvolution(in=256, out=64, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1]) (4): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (5): MinkowskiReLU() ) ) (map_to_bev_module): None (pfe): None (backbone_2d): None (dense_head): CAGroup3DHead( (loss_centerness): CrossEntropy() (loss_bbox): IoU3DLoss() (loss_cls): FocalLoss() (loss_sem): FocalLoss() (loss_offset): SmoothL1Loss() (offset_block): Sequential( (0): MinkowskiConvolution(in=64, out=64, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiELU() (3): MinkowskiConvolution(in=64, out=64, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1]) (4): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (5): MinkowskiELU() (6): MinkowskiConvolution(in=64, out=9, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1]) ) (feature_offset): Sequential( (0): MinkowskiConvolution(in=64, out=192, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiELU() ) (semantic_conv): MinkowskiConvolution(in=64, out=10, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1]) (centerness_conv): MinkowskiConvolution(in=64, out=1, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1]) (reg_conv): MinkowskiConvolution(in=64, out=8, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1]) (cls_conv): MinkowskiConvolution(in=64, out=10, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1]) (scales): ModuleList( (0): Scale() (1): Scale() (2): Scale() (3): Scale() (4): Scale() (5): Scale() (6): Scale() (7): Scale() (8): Scale() (9): Scale() ) (cls_individual_out): ModuleList( (0): Sequential( (0): MinkowskiConvolution(in=64, out=64, kernel_size=[9, 9, 9], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiELU() ) (1): Sequential( (0): MinkowskiConvolution(in=64, out=64, kernel_size=[9, 9, 9], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiELU() ) (2): Sequential( (0): MinkowskiConvolution(in=64, out=64, kernel_size=[9, 9, 9], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiELU() ) (3): Sequential( (0): MinkowskiConvolution(in=64, out=64, kernel_size=[9, 9, 9], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiELU() ) (4): Sequential( (0): MinkowskiConvolution(in=64, out=64, kernel_size=[9, 9, 9], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiELU() ) (5): Sequential( (0): MinkowskiConvolution(in=64, out=64, kernel_size=[9, 9, 9], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiELU() ) (6): Sequential( (0): MinkowskiConvolution(in=64, out=64, kernel_size=[9, 9, 9], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiELU() ) (7): Sequential( (0): MinkowskiConvolution(in=64, out=64, kernel_size=[9, 9, 9], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiELU() ) (8): Sequential( (0): MinkowskiConvolution(in=64, out=64, kernel_size=[9, 9, 9], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiELU() ) (9): Sequential( (0): MinkowskiConvolution(in=64, out=64, kernel_size=[9, 9, 9], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiELU() ) ) (cls_individual_up): ModuleList( (0): ModuleList( (0): MinkowskiGenerativeConvolutionTranspose(in=64, out=64, kernel_size=[3, 3, 3], stride=[3, 3, 3], dilation=[1, 1, 1]) (1): Sequential( (0): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (1): MinkowskiELU() ) ) (1): ModuleList( (0): MinkowskiGenerativeConvolutionTranspose(in=64, out=64, kernel_size=[3, 3, 3], stride=[3, 3, 3], dilation=[1, 1, 1]) (1): Sequential( (0): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (1): MinkowskiELU() ) ) (2): ModuleList( (0): MinkowskiGenerativeConvolutionTranspose(in=64, out=64, kernel_size=[3, 3, 3], stride=[3, 3, 3], dilation=[1, 1, 1]) (1): Sequential( (0): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (1): MinkowskiELU() ) ) (3): ModuleList( (0): MinkowskiGenerativeConvolutionTranspose(in=64, out=64, kernel_size=[3, 3, 3], stride=[3, 3, 3], dilation=[1, 1, 1]) (1): Sequential( (0): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (1): MinkowskiELU() ) ) (4): ModuleList( (0): MinkowskiGenerativeConvolutionTranspose(in=64, out=64, kernel_size=[3, 3, 3], stride=[3, 3, 3], dilation=[1, 1, 1]) (1): Sequential( (0): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (1): MinkowskiELU() ) ) (5): ModuleList( (0): MinkowskiGenerativeConvolutionTranspose(in=64, out=64, kernel_size=[3, 3, 3], stride=[3, 3, 3], dilation=[1, 1, 1]) (1): Sequential( (0): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (1): MinkowskiELU() ) ) (6): ModuleList( (0): MinkowskiGenerativeConvolutionTranspose(in=64, out=64, kernel_size=[3, 3, 3], stride=[3, 3, 3], dilation=[1, 1, 1]) (1): Sequential( (0): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (1): MinkowskiELU() ) ) (7): ModuleList( (0): MinkowskiGenerativeConvolutionTranspose(in=64, out=64, kernel_size=[3, 3, 3], stride=[3, 3, 3], dilation=[1, 1, 1]) (1): Sequential( (0): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (1): MinkowskiELU() ) ) (8): ModuleList( (0): MinkowskiGenerativeConvolutionTranspose(in=64, out=64, kernel_size=[3, 3, 3], stride=[3, 3, 3], dilation=[1, 1, 1]) (1): Sequential( (0): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (1): MinkowskiELU() ) ) (9): ModuleList( (0): MinkowskiGenerativeConvolutionTranspose(in=64, out=64, kernel_size=[3, 3, 3], stride=[3, 3, 3], dilation=[1, 1, 1]) (1): Sequential( (0): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (1): MinkowskiELU() ) ) ) (cls_individual_fuse): ModuleList( (0): Sequential( (0): MinkowskiConvolution(in=128, out=64, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiELU() ) (1): Sequential( (0): MinkowskiConvolution(in=128, out=64, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiELU() ) (2): Sequential( (0): MinkowskiConvolution(in=128, out=64, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiELU() ) (3): Sequential( (0): MinkowskiConvolution(in=128, out=64, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiELU() ) (4): Sequential( (0): MinkowskiConvolution(in=128, out=64, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiELU() ) (5): Sequential( (0): MinkowskiConvolution(in=128, out=64, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiELU() ) (6): Sequential( (0): MinkowskiConvolution(in=128, out=64, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiELU() ) (7): Sequential( (0): MinkowskiConvolution(in=128, out=64, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiELU() ) (8): Sequential( (0): MinkowskiConvolution(in=128, out=64, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiELU() ) (9): Sequential( (0): MinkowskiConvolution(in=128, out=64, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiELU() ) ) (cls_individual_expand_out): ModuleList( (0): Sequential( (0): MinkowskiConvolution(in=64, out=64, kernel_size=[5, 5, 5], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiELU() ) (1): Sequential( (0): MinkowskiConvolution(in=64, out=64, kernel_size=[5, 5, 5], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiELU() ) (2): Sequential( (0): MinkowskiConvolution(in=64, out=64, kernel_size=[5, 5, 5], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiELU() ) (3): Sequential( (0): MinkowskiConvolution(in=64, out=64, kernel_size=[5, 5, 5], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiELU() ) (4): Sequential( (0): MinkowskiConvolution(in=64, out=64, kernel_size=[5, 5, 5], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiELU() ) (5): Sequential( (0): MinkowskiConvolution(in=64, out=64, kernel_size=[5, 5, 5], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiELU() ) (6): Sequential( (0): MinkowskiConvolution(in=64, out=64, kernel_size=[5, 5, 5], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiELU() ) (7): Sequential( (0): MinkowskiConvolution(in=64, out=64, kernel_size=[5, 5, 5], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiELU() ) (8): Sequential( (0): MinkowskiConvolution(in=64, out=64, kernel_size=[5, 5, 5], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiELU() ) (9): Sequential( (0): MinkowskiConvolution(in=64, out=64, kernel_size=[5, 5, 5], stride=[1, 1, 1], dilation=[1, 1, 1]) (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): MinkowskiELU() ) ) ) (point_head): None (roi_head): CAGroup3DRoIHead( (iou_loss_computer): IoU3DLoss() (proposal_target_layer): ProposalTargetLayer() (reg_loss_func): WeightedSmoothL1Loss() (roi_grid_pool_layers): ModuleList( (0): SimplePoolingLayer( (grid_conv): MinkowskiConvolution(in=64, out=128, kernel_size=[5, 5, 5], stride=[1, 1, 1], dilation=[1, 1, 1]) (grid_bn): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (grid_relu): MinkowskiELU() (pooling_conv): MinkowskiConvolution(in=128, out=128, kernel_size=[7, 7, 7], stride=[1, 1, 1], dilation=[1, 1, 1]) (pooling_bn): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) (reg_fc_layers): Sequential( (0): Linear(in_features=128, out_features=256, bias=False) (1): BatchNorm1d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): ReLU() (3): Dropout(p=0.3, inplace=False) (4): Linear(in_features=256, out_features=256, bias=False) (5): BatchNorm1d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (6): ReLU() ) (reg_pred_layer): Linear(in_features=256, out_features=8, bias=True) ) ) ) 2023-03-26 11:13:36,794 INFO **********************Start training sunrgbd_models/CAGroup3D(cagroup3d-win10-sunrgbd)********************** 2023-03-26 12:29:52,942 INFO Epoch [ 1][ 50]/[1322] : lr: 1.000e-03, sem_thr: 0.15, loss_centerness: 0.20667796790599824, loss_bbox: 0.2990968986600637, loss_cls: 0.3071140018105507, loss_sem: 1.0371560204029082, loss_vote: 0.4092518210411072, one_stage_loss: 2.259296736717224, rcnn_loss_reg: 0.06894166469573974, rcnn_loss_iou: 0.03426224589347839, loss_two_stage: 0.10320390939712525, 2023-03-26 13:52:01,310 INFO Epoch [ 1][ 100]/[1322] : lr: 1.000e-03, sem_thr: 0.15, loss_centerness: 0.5310189422965049, loss_bbox: 0.7474155166745186, loss_cls: 0.6665185099840164, loss_sem: 0.6469615936279297, loss_vote: 0.3759205311536789, one_stage_loss: 2.967835102081299, rcnn_loss_reg: 0.4916525638103485, rcnn_loss_iou: 0.31042004823684693, loss_two_stage: 0.8020726108551025, 2023-03-26 15:16:34,527 INFO Epoch [ 1][ 150]/[1322] : lr: 1.000e-03, sem_thr: 0.15, loss_centerness: 0.47814485549926755, loss_bbox: 0.6697727379202842, loss_cls: 0.5062539026141166, loss_sem: 0.5720264983177185, loss_vote: 0.3539592534303665, one_stage_loss: 2.580157253742218, rcnn_loss_reg: 0.4772424042224884, rcnn_loss_iou: 0.2981147265434265, loss_two_stage: 0.7753571343421936, 2023-03-26 16:42:56,533 INFO Epoch [ 1][ 200]/[1322] : lr: 1.000e-03, sem_thr: 0.15, loss_centerness: 0.5514369648694992, loss_bbox: 0.7656048792600632, loss_cls: 0.5199696916341782, loss_sem: 0.5628203338384629, loss_vote: 0.3609760856628418, one_stage_loss: 2.7608079433441164, rcnn_loss_reg: 0.5857562351226807, rcnn_loss_iou: 0.35664750695228575, loss_two_stage: 0.9424037384986877, 2023-03-26 18:04:56,578 INFO Epoch [ 1][ 250]/[1322] : lr: 1.000e-03, sem_thr: 0.15, loss_centerness: 0.47652084708213804, loss_bbox: 0.6971965312957764, loss_cls: 0.4520271807909012, loss_sem: 0.6041016298532486, loss_vote: 0.35814492881298066, one_stage_loss: 2.587991120815277, rcnn_loss_reg: 0.29248207330703735, rcnn_loss_iou: 0.1785850465297699, loss_two_stage: 0.47106712102890014, 2023-03-26 19:28:39,683 INFO Epoch [ 1][ 300]/[1322] : lr: 1.000e-03, sem_thr: 0.15, loss_centerness: 0.5167306911945343, loss_bbox: 0.7328562247753143, loss_cls: 0.45524279356002806, loss_sem: 0.515706689953804, loss_vote: 0.32769578754901885, one_stage_loss: 2.5482321834564208, rcnn_loss_reg: 0.35428098678588865, rcnn_loss_iou: 0.21939268827438355, loss_two_stage: 0.5736736750602722, 2023-03-26 20:53:09,330 INFO Epoch [ 1][ 350]/[1322] : lr: 1.000e-03, sem_thr: 0.15, loss_centerness: 0.604567551612854, loss_bbox: 0.8178253293037414, loss_cls: 0.5145705115795135, loss_sem: 0.4814878588914871, loss_vote: 0.30948863834142687, one_stage_loss: 2.727939887046814, rcnn_loss_reg: 0.8254446721076966, rcnn_loss_iou: 0.5003487575054169, loss_two_stage: 1.3257934308052064, 2023-03-26 22:14:45,021 INFO Epoch [ 1][ 400]/[1322] : lr: 1.000e-03, sem_thr: 0.15, loss_centerness: 0.5232829931378364, loss_bbox: 0.729054206609726, loss_cls: 0.4304437929391861, loss_sem: 0.6052549344301223, loss_vote: 0.32101795822381973, one_stage_loss: 2.609053874015808, rcnn_loss_reg: 0.47852428913116457, rcnn_loss_iou: 0.29645866751670835, loss_two_stage: 0.7749829506874084, 2023-03-26 23:39:29,069 INFO Epoch [ 1][ 450]/[1322] : lr: 1.000e-03, sem_thr: 0.15, loss_centerness: 0.4240903949737549, loss_bbox: 0.6023871430754661, loss_cls: 0.43437127619981764, loss_sem: 0.58591732442379, loss_vote: 0.31612290561199186, one_stage_loss: 2.362889051437378, rcnn_loss_reg: 0.3720583057403564, rcnn_loss_iou: 0.2311265003681183, loss_two_stage: 0.6031848061084747, 2023-03-27 01:03:24,345 INFO Epoch [ 1][ 500]/[1322] : lr: 1.000e-03, sem_thr: 0.15, loss_centerness: 0.3820648092031479, loss_bbox: 0.5443513408303261, loss_cls: 0.36938557237386704, loss_sem: 0.6360691225528717, loss_vote: 0.31682781159877776, one_stage_loss: 2.2486986470222474, rcnn_loss_reg: 0.36717917561531066, rcnn_loss_iou: 0.23230359673500062, loss_two_stage: 0.5994827747344971, 2023-03-27 02:25:21,026 INFO Epoch [ 1][ 550]/[1322] : lr: 1.000e-03, sem_thr: 0.15, loss_centerness: 0.5708097213506699, loss_bbox: 0.7884046626091004, loss_cls: 0.5089094871282578, loss_sem: 0.5691659837961197, loss_vote: 0.3056716549396515, one_stage_loss: 2.742961506843567, rcnn_loss_reg: 0.6101349353790283, rcnn_loss_iou: 0.3731300795078278, loss_two_stage: 0.9832650113105774, 2023-03-27 03:44:41,116 INFO Epoch [ 1][ 600]/[1322] : lr: 1.000e-03, sem_thr: 0.15, loss_centerness: 0.5802704590559006, loss_bbox: 0.7960068082809448, loss_cls: 0.459269055724144, loss_sem: 0.5500217407941819, loss_vote: 0.29440259009599684, one_stage_loss: 2.6799706506729124, rcnn_loss_reg: 0.6698205232620239, rcnn_loss_iou: 0.40001817524433136, loss_two_stage: 1.069838695526123, 2023-03-27 05:05:05,293 INFO Epoch [ 1][ 650]/[1322] : lr: 1.000e-03, sem_thr: 0.15, loss_centerness: 0.5250428706407547, loss_bbox: 0.7363606292009354, loss_cls: 0.4330296140909195, loss_sem: 0.6324895197153091, loss_vote: 0.3019771608710289, one_stage_loss: 2.6288998174667357, rcnn_loss_reg: 0.4743223488330841, rcnn_loss_iou: 0.290767160654068, loss_two_stage: 0.7650895142555236, 2023-03-27 06:23:27,892 INFO Epoch [ 1][ 700]/[1322] : lr: 1.000e-03, sem_thr: 0.15, loss_centerness: 0.5444362837076188, loss_bbox: 0.7522995507717133, loss_cls: 0.4244315469264984, loss_sem: 0.45511561155319213, loss_vote: 0.3005108740925789, one_stage_loss: 2.476793870925903, rcnn_loss_reg: 0.5306920790672303, rcnn_loss_iou: 0.31801644802093504, loss_two_stage: 0.8487085247039795, 2023-03-27 07:47:41,250 INFO Epoch [ 1][ 750]/[1322] : lr: 1.000e-03, sem_thr: 0.15, loss_centerness: 0.6098733115196228, loss_bbox: 0.8267546522617341, loss_cls: 0.45813657343387604, loss_sem: 0.47525602877140044, loss_vote: 0.2934702724218369, one_stage_loss: 2.6634908294677735, rcnn_loss_reg: 0.7940043830871581, rcnn_loss_iou: 0.4807016372680664, loss_two_stage: 1.2747060203552245, 2023-03-27 09:12:50,838 INFO Epoch [ 1][ 800]/[1322] : lr: 1.000e-03, sem_thr: 0.15, loss_centerness: 0.5564212232828141, loss_bbox: 0.7674001061916351, loss_cls: 0.4770412653684616, loss_sem: 0.5462373447418213, loss_vote: 0.28620937079191205, one_stage_loss: 2.6333092975616457, rcnn_loss_reg: 0.633576123714447, rcnn_loss_iou: 0.390158035159111, loss_two_stage: 1.0237341594696046, 2023-03-27 10:33:02,697 INFO Epoch [ 1][ 850]/[1322] : lr: 1.000e-03, sem_thr: 0.15, loss_centerness: 0.6046576589345932, loss_bbox: 0.8187049305438996, loss_cls: 0.44491772472858426, loss_sem: 0.5175370579957962, loss_vote: 0.2872957721352577, one_stage_loss: 2.6731131219863893, rcnn_loss_reg: 0.7730563342571258, rcnn_loss_iou: 0.4753589242696762, loss_two_stage: 1.2484152507781983, 2023-03-27 11:57:22,870 INFO Epoch [ 1][ 900]/[1322] : lr: 1.000e-03, sem_thr: 0.15, loss_centerness: 0.6090051412582398, loss_bbox: 0.8095787620544433, loss_cls: 0.4284948009252548, loss_sem: 0.544955484867096, loss_vote: 0.289944207072258, one_stage_loss: 2.6819783878326415, rcnn_loss_reg: 0.8724133563041687, rcnn_loss_iou: 0.5291033411026, loss_two_stage: 1.4015166926383973, 2023-03-27 13:17:06,861 INFO Epoch [ 1][ 950]/[1322] : lr: 1.000e-03, sem_thr: 0.15, loss_centerness: 0.5685943412780762, loss_bbox: 0.789885265827179, loss_cls: 0.4120765841007233, loss_sem: 0.44937529861927034, loss_vote: 0.2983333826065063, one_stage_loss: 2.5182648515701294, rcnn_loss_reg: 0.532546899318695, rcnn_loss_iou: 0.33089754700660706, loss_two_stage: 0.863444447517395, 2023-03-27 14:41:55,361 INFO Epoch [ 1][1000]/[1322] : lr: 1.000e-03, sem_thr: 0.15, loss_centerness: 0.6334057796001434, loss_bbox: 0.8424523377418518, loss_cls: 0.4098082458972931, loss_sem: 0.40657821238040925, loss_vote: 0.2823134985566139, one_stage_loss: 2.5745581150054933, rcnn_loss_reg: 0.8423162472248077, rcnn_loss_iou: 0.514250785112381, loss_two_stage: 1.3565670371055603, 2023-03-27 16:00:37,355 INFO Epoch [ 1][1050]/[1322] : lr: 1.000e-03, sem_thr: 0.15, loss_centerness: 0.5406019860506057, loss_bbox: 0.7440398615598679, loss_cls: 0.39454961895942686, loss_sem: 0.5495830583572388, loss_vote: 0.2843384724855423, one_stage_loss: 2.5131130170822145, rcnn_loss_reg: 0.6645504760742188, rcnn_loss_iou: 0.4026562488079071, loss_two_stage: 1.0672067260742188, 2023-03-27 17:20:17,667 INFO Epoch [ 1][1100]/[1322] : lr: 1.000e-03, sem_thr: 0.15, loss_centerness: 0.5215873223543167, loss_bbox: 0.7336350619792938, loss_cls: 0.41735528588294984, loss_sem: 0.4811014491319656, loss_vote: 0.29286098569631575, one_stage_loss: 2.4465401220321654, rcnn_loss_reg: 0.40519655585289, rcnn_loss_iou: 0.2512069880962372, loss_two_stage: 0.656403546333313, 2023-03-27 18:39:11,247 INFO Epoch [ 1][1150]/[1322] : lr: 1.000e-03, sem_thr: 0.15, loss_centerness: 0.5330367308855056, loss_bbox: 0.7439761543273926, loss_cls: 0.40116438329219817, loss_sem: 0.5646754199266434, loss_vote: 0.29715222597122193, one_stage_loss: 2.5400049018859865, rcnn_loss_reg: 0.6408816885948181, rcnn_loss_iou: 0.39092864990234377, loss_two_stage: 1.031810338497162, 2023-03-27 20:02:50,511 INFO Epoch [ 1][1200]/[1322] : lr: 1.000e-03, sem_thr: 0.15, loss_centerness: 0.6315625596046448, loss_bbox: 0.8405828297138214, loss_cls: 0.4183447903394699, loss_sem: 0.3946588611602783, loss_vote: 0.28028561860322954, one_stage_loss: 2.565434675216675, rcnn_loss_reg: 0.8700305473804474, rcnn_loss_iou: 0.5194281542301178, loss_two_stage: 1.3894586968421936, 2023-03-27 21:25:10,486 INFO Epoch [ 1][1250]/[1322] : lr: 1.000e-03, sem_thr: 0.15, loss_centerness: 0.6114491987228393, loss_bbox: 0.8168105685710907, loss_cls: 0.4101775807142258, loss_sem: 0.4728459024429321, loss_vote: 0.27880846709012985, one_stage_loss: 2.590091710090637, rcnn_loss_reg: 0.8118275630474091, rcnn_loss_iou: 0.4983943772315979, loss_two_stage: 1.3102219486236573, 2023-03-27 22:47:39,311 INFO Epoch [ 1][1300]/[1322] : lr: 1.000e-03, sem_thr: 0.15, loss_centerness: 0.6145700442790986, loss_bbox: 0.81527019739151, loss_cls: 0.3963031381368637, loss_sem: 0.4105939191579819, loss_vote: 0.27149350941181183, one_stage_loss: 2.5082308149337766, rcnn_loss_reg: 0.8241646671295166, rcnn_loss_iou: 0.4978309834003449, loss_two_stage: 1.3219956469535827, 2023-03-27 23:26:17,307 INFO **********************End training sunrgbd_models/CAGroup3D(cagroup3d-win10-sunrgbd)********************** 2023-03-27 23:26:17,309 INFO **********************Start evaluation sunrgbd_models/CAGroup3D(cagroup3d-win10-sunrgbd)********************** 2023-03-27 23:26:17,310 INFO Loading SUNRGBD dataset 2023-03-27 23:26:17,476 INFO Total samples for SUNRGBD dataset: 5050 2023-03-27 23:26:17,484 INFO ==> Loading parameters from checkpoint C:\PINKAMENA\CITYU\CS5182\proj\CAGroup3D\output\sunrgbd_models\CAGroup3D\cagroup3d-win10-sunrgbd\ckpt\checkpoint_epoch_1.pth to CPU 2023-03-27 23:26:18,098 INFO ==> Checkpoint trained from version: pcdet+0.5.2+18bc5f5+py60edc0c 2023-03-27 23:26:18,162 INFO ==> Done (loaded 638/638) 2023-03-27 23:26:19,088 INFO *************** EPOCH 1 EVALUATION *****************