2023-03-31 07:55:25,762 INFO **********************Start logging********************** 2023-03-31 07:55:25,763 INFO CUDA_VISIBLE_DEVICES=ALL 2023-03-31 07:55:25,764 INFO total_batch_size: 16 2023-03-31 07:55:25,765 INFO cfg_file cfgs/sunrgbd_models/CAGroup3D.yaml 2023-03-31 07:55:25,765 INFO batch_size 16 2023-03-31 07:55:25,766 INFO workers 4 2023-03-31 07:55:25,766 INFO extra_tag cagroup3d-win10-sunrgbd-eval 2023-03-31 07:55:25,767 INFO ckpt ../output/sunrgbd_models/CAGroup3D/cagroup3d-win10-sunrgbd-train/ckpt/checkpoint_epoch_1.pth 2023-03-31 07:55:25,768 INFO launcher pytorch 2023-03-31 07:55:25,768 INFO tcp_port 18888 2023-03-31 07:55:25,768 INFO set_cfgs None 2023-03-31 07:55:25,770 INFO max_waiting_mins 30 2023-03-31 07:55:25,770 INFO start_epoch 0 2023-03-31 07:55:25,771 INFO eval_tag default 2023-03-31 07:55:25,772 INFO eval_all False 2023-03-31 07:55:25,772 INFO ckpt_dir None 2023-03-31 07:55:25,772 INFO save_to_file False 2023-03-31 07:55:25,773 INFO cfg.ROOT_DIR: C:\PINKAMENA\CITYU\CS5182\proj\CAGroup3D 2023-03-31 07:55:25,774 INFO cfg.LOCAL_RANK: 0 2023-03-31 07:55:25,774 INFO cfg.CLASS_NAMES: ['bed', 'table', 'sofa', 'chair', 'toilet', 'desk', 'dresser', 'night_stand', 'bookshelf', 'bathtub'] 2023-03-31 07:55:25,775 INFO cfg.DATA_CONFIG = edict() 2023-03-31 07:55:25,776 INFO cfg.DATA_CONFIG.DATASET: SunrgbdDataset 2023-03-31 07:55:25,776 INFO cfg.DATA_CONFIG.DATA_PATH: ../data/sunrgbd_data/sunrgbd 2023-03-31 07:55:25,777 INFO cfg.DATA_CONFIG.PROCESSED_DATA_TAG: sunrgbd_processed_data_v0_5_0 2023-03-31 07:55:25,778 INFO cfg.DATA_CONFIG.POINT_CLOUD_RANGE: [-40, -40, -10, 40, 40, 10] 2023-03-31 07:55:25,779 INFO cfg.DATA_CONFIG.DATA_SPLIT = edict() 2023-03-31 07:55:25,781 INFO cfg.DATA_CONFIG.DATA_SPLIT.train: train 2023-03-31 07:55:25,781 INFO cfg.DATA_CONFIG.DATA_SPLIT.test: val 2023-03-31 07:55:25,783 INFO cfg.DATA_CONFIG.REPEAT = edict() 2023-03-31 07:55:25,784 INFO cfg.DATA_CONFIG.REPEAT.train: 4 2023-03-31 07:55:25,784 INFO cfg.DATA_CONFIG.REPEAT.test: 1 2023-03-31 07:55:25,785 INFO cfg.DATA_CONFIG.INFO_PATH = edict() 2023-03-31 07:55:25,786 INFO cfg.DATA_CONFIG.INFO_PATH.train: ['sunrgbd_infos_train.pkl'] 2023-03-31 07:55:25,787 INFO cfg.DATA_CONFIG.INFO_PATH.test: ['sunrgbd_infos_val.pkl'] 2023-03-31 07:55:25,789 INFO cfg.DATA_CONFIG.GET_ITEM_LIST: ['points'] 2023-03-31 07:55:25,789 INFO cfg.DATA_CONFIG.FILTER_EMPTY_BOXES_FOR_TRAIN: True 2023-03-31 07:55:25,790 INFO cfg.DATA_CONFIG.DATA_AUGMENTOR_TRAIN = edict() 2023-03-31 07:55:25,791 INFO cfg.DATA_CONFIG.DATA_AUGMENTOR_TRAIN.DISABLE_AUG_LIST: ['placeholder'] 2023-03-31 07:55:25,792 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}] 2023-03-31 07:55:25,793 INFO cfg.DATA_CONFIG.DATA_AUGMENTOR_TEST = edict() 2023-03-31 07:55:25,796 INFO cfg.DATA_CONFIG.DATA_AUGMENTOR_TEST.DISABLE_AUG_LIST: ['placeholder'] 2023-03-31 07:55:25,797 INFO cfg.DATA_CONFIG.DATA_AUGMENTOR_TEST.AUG_CONFIG_LIST: [{'NAME': 'indoor_point_sample', 'num_points': 100000}] 2023-03-31 07:55:25,798 INFO cfg.DATA_CONFIG.DATA_AUGMENTOR = edict() 2023-03-31 07:55:25,799 INFO cfg.DATA_CONFIG.DATA_AUGMENTOR.DISABLE_AUG_LIST: ['placeholder'] 2023-03-31 07:55:25,800 INFO cfg.DATA_CONFIG.DATA_AUGMENTOR.AUG_CONFIG_LIST: [{'NAME': 'indoor_point_sample', 'num_points': 50000}] 2023-03-31 07:55:25,801 INFO cfg.DATA_CONFIG.POINT_FEATURE_ENCODING = edict() 2023-03-31 07:55:25,801 INFO cfg.DATA_CONFIG.POINT_FEATURE_ENCODING.encoding_type: absolute_coordinates_encoding 2023-03-31 07:55:25,804 INFO cfg.DATA_CONFIG.POINT_FEATURE_ENCODING.used_feature_list: ['x', 'y', 'z', 'r', 'g', 'b'] 2023-03-31 07:55:25,804 INFO cfg.DATA_CONFIG.POINT_FEATURE_ENCODING.src_feature_list: ['x', 'y', 'z', 'r', 'g', 'b'] 2023-03-31 07:55:25,805 INFO cfg.DATA_CONFIG.DATA_PROCESSOR: [{'NAME': 'mask_points_and_boxes_outside_range', 'REMOVE_OUTSIDE_BOXES': False}] 2023-03-31 07:55:25,806 INFO cfg.DATA_CONFIG._BASE_CONFIG_: cfgs/dataset_configs/sunrgbd_dataset.yaml 2023-03-31 07:55:25,807 INFO cfg.VOXEL_SIZE: 0.02 2023-03-31 07:55:25,808 INFO cfg.N_CLASSES: 10 2023-03-31 07:55:25,810 INFO cfg.SEMANTIC_THR: 0.15 2023-03-31 07:55:25,810 INFO cfg.MODEL = edict() 2023-03-31 07:55:25,811 INFO cfg.MODEL.NAME: CAGroup3D 2023-03-31 07:55:25,811 INFO cfg.MODEL.VOXEL_SIZE: 0.02 2023-03-31 07:55:25,812 INFO cfg.MODEL.SEMANTIC_MIN_THR: 0.05 2023-03-31 07:55:25,813 INFO cfg.MODEL.SEMANTIC_ITER_VALUE: 0.02 2023-03-31 07:55:25,814 INFO cfg.MODEL.SEMANTIC_THR: 0.15 2023-03-31 07:55:25,815 INFO cfg.MODEL.BACKBONE_3D = edict() 2023-03-31 07:55:25,816 INFO cfg.MODEL.BACKBONE_3D.NAME: BiResNet 2023-03-31 07:55:25,816 INFO cfg.MODEL.BACKBONE_3D.IN_CHANNELS: 3 2023-03-31 07:55:25,817 INFO cfg.MODEL.BACKBONE_3D.OUT_CHANNELS: 64 2023-03-31 07:55:25,819 INFO cfg.MODEL.DENSE_HEAD = edict() 2023-03-31 07:55:25,820 INFO cfg.MODEL.DENSE_HEAD.NAME: CAGroup3DHead 2023-03-31 07:55:25,821 INFO cfg.MODEL.DENSE_HEAD.IN_CHANNELS: [64, 128, 256, 512] 2023-03-31 07:55:25,821 INFO cfg.MODEL.DENSE_HEAD.OUT_CHANNELS: 64 2023-03-31 07:55:25,822 INFO cfg.MODEL.DENSE_HEAD.SEMANTIC_THR: 0.15 2023-03-31 07:55:25,823 INFO cfg.MODEL.DENSE_HEAD.VOXEL_SIZE: 0.02 2023-03-31 07:55:25,824 INFO cfg.MODEL.DENSE_HEAD.N_CLASSES: 10 2023-03-31 07:55:25,824 INFO cfg.MODEL.DENSE_HEAD.N_REG_OUTS: 8 2023-03-31 07:55:25,825 INFO cfg.MODEL.DENSE_HEAD.CLS_KERNEL: 9 2023-03-31 07:55:25,825 INFO cfg.MODEL.DENSE_HEAD.WITH_YAW: True 2023-03-31 07:55:25,826 INFO cfg.MODEL.DENSE_HEAD.USE_SEM_SCORE: False 2023-03-31 07:55:25,827 INFO cfg.MODEL.DENSE_HEAD.EXPAND_RATIO: 3 2023-03-31 07:55:25,828 INFO cfg.MODEL.DENSE_HEAD.ASSIGNER = edict() 2023-03-31 07:55:25,828 INFO cfg.MODEL.DENSE_HEAD.ASSIGNER.NAME: CAGroup3DAssigner 2023-03-31 07:55:25,830 INFO cfg.MODEL.DENSE_HEAD.ASSIGNER.LIMIT: 27 2023-03-31 07:55:25,831 INFO cfg.MODEL.DENSE_HEAD.ASSIGNER.TOPK: 18 2023-03-31 07:55:25,831 INFO cfg.MODEL.DENSE_HEAD.ASSIGNER.N_SCALES: 4 2023-03-31 07:55:25,833 INFO cfg.MODEL.DENSE_HEAD.LOSS_OFFSET = edict() 2023-03-31 07:55:25,833 INFO cfg.MODEL.DENSE_HEAD.LOSS_OFFSET.NAME: SmoothL1Loss 2023-03-31 07:55:25,834 INFO cfg.MODEL.DENSE_HEAD.LOSS_OFFSET.BETA: 0.04 2023-03-31 07:55:25,835 INFO cfg.MODEL.DENSE_HEAD.LOSS_OFFSET.REDUCTION: sum 2023-03-31 07:55:25,836 INFO cfg.MODEL.DENSE_HEAD.LOSS_OFFSET.LOSS_WEIGHT: 0.2 2023-03-31 07:55:25,837 INFO cfg.MODEL.DENSE_HEAD.LOSS_BBOX = edict() 2023-03-31 07:55:25,838 INFO cfg.MODEL.DENSE_HEAD.LOSS_BBOX.NAME: IoU3DLoss 2023-03-31 07:55:25,839 INFO cfg.MODEL.DENSE_HEAD.LOSS_BBOX.WITH_YAW: True 2023-03-31 07:55:25,840 INFO cfg.MODEL.DENSE_HEAD.LOSS_BBOX.LOSS_WEIGHT: 1.0 2023-03-31 07:55:25,841 INFO cfg.MODEL.DENSE_HEAD.NMS_CONFIG = edict() 2023-03-31 07:55:25,842 INFO cfg.MODEL.DENSE_HEAD.NMS_CONFIG.SCORE_THR: 0.01 2023-03-31 07:55:25,842 INFO cfg.MODEL.DENSE_HEAD.NMS_CONFIG.NMS_PRE: 1000 2023-03-31 07:55:25,843 INFO cfg.MODEL.DENSE_HEAD.NMS_CONFIG.IOU_THR: 0.5 2023-03-31 07:55:25,843 INFO cfg.MODEL.ROI_HEAD = edict() 2023-03-31 07:55:25,845 INFO cfg.MODEL.ROI_HEAD.NAME: CAGroup3DRoIHead 2023-03-31 07:55:25,847 INFO cfg.MODEL.ROI_HEAD.NUM_CLASSES: 10 2023-03-31 07:55:25,848 INFO cfg.MODEL.ROI_HEAD.MIDDLE_FEATURE_SOURCE: [3] 2023-03-31 07:55:25,849 INFO cfg.MODEL.ROI_HEAD.GRID_SIZE: 7 2023-03-31 07:55:25,849 INFO cfg.MODEL.ROI_HEAD.VOXEL_SIZE: 0.02 2023-03-31 07:55:25,850 INFO cfg.MODEL.ROI_HEAD.COORD_KEY: 2 2023-03-31 07:55:25,851 INFO cfg.MODEL.ROI_HEAD.MLPS: [[64, 128, 128]] 2023-03-31 07:55:25,852 INFO cfg.MODEL.ROI_HEAD.CODE_SIZE: 7 2023-03-31 07:55:25,853 INFO cfg.MODEL.ROI_HEAD.ENCODE_SINCOS: True 2023-03-31 07:55:25,853 INFO cfg.MODEL.ROI_HEAD.ROI_PER_IMAGE: 128 2023-03-31 07:55:25,855 INFO cfg.MODEL.ROI_HEAD.ROI_FG_RATIO: 0.9 2023-03-31 07:55:25,857 INFO cfg.MODEL.ROI_HEAD.REG_FG_THRESH: 0.3 2023-03-31 07:55:25,858 INFO cfg.MODEL.ROI_HEAD.ROI_CONV_KERNEL: 5 2023-03-31 07:55:25,858 INFO cfg.MODEL.ROI_HEAD.ENLARGE_RATIO: False 2023-03-31 07:55:25,860 INFO cfg.MODEL.ROI_HEAD.USE_IOU_LOSS: True 2023-03-31 07:55:25,862 INFO cfg.MODEL.ROI_HEAD.USE_GRID_OFFSET: False 2023-03-31 07:55:25,862 INFO cfg.MODEL.ROI_HEAD.USE_SIMPLE_POOLING: True 2023-03-31 07:55:25,863 INFO cfg.MODEL.ROI_HEAD.USE_CENTER_POOLING: True 2023-03-31 07:55:25,864 INFO cfg.MODEL.ROI_HEAD.LOSS_WEIGHTS = edict() 2023-03-31 07:55:25,865 INFO cfg.MODEL.ROI_HEAD.LOSS_WEIGHTS.RCNN_CLS_WEIGHT: 1.0 2023-03-31 07:55:25,865 INFO cfg.MODEL.ROI_HEAD.LOSS_WEIGHTS.RCNN_REG_WEIGHT: 0.5 2023-03-31 07:55:25,865 INFO cfg.MODEL.ROI_HEAD.LOSS_WEIGHTS.RCNN_IOU_WEIGHT: 1.0 2023-03-31 07:55:25,866 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] 2023-03-31 07:55:25,867 INFO cfg.MODEL.POST_PROCESSING = edict() 2023-03-31 07:55:25,868 INFO cfg.MODEL.POST_PROCESSING.RECALL_THRESH_LIST: [0.25, 0.5] 2023-03-31 07:55:25,870 INFO cfg.MODEL.POST_PROCESSING.EVAL_METRIC: scannet 2023-03-31 07:55:25,871 INFO cfg.OPTIMIZATION = edict() 2023-03-31 07:55:25,872 INFO cfg.OPTIMIZATION.BATCH_SIZE_PER_GPU: 16 2023-03-31 07:55:25,873 INFO cfg.OPTIMIZATION.NUM_EPOCHS: 1 2023-03-31 07:55:25,873 INFO cfg.OPTIMIZATION.OPTIMIZER: adamW 2023-03-31 07:55:25,875 INFO cfg.OPTIMIZATION.LR: 0.001 2023-03-31 07:55:25,876 INFO cfg.OPTIMIZATION.WEIGHT_DECAY: 0.0001 2023-03-31 07:55:25,877 INFO cfg.OPTIMIZATION.DECAY_STEP_LIST: [8, 11] 2023-03-31 07:55:25,878 INFO cfg.OPTIMIZATION.LR_DECAY: 0.1 2023-03-31 07:55:25,878 INFO cfg.OPTIMIZATION.GRAD_NORM_CLIP: 10 2023-03-31 07:55:25,879 INFO cfg.OPTIMIZATION.PCT_START: 0.4 2023-03-31 07:55:25,880 INFO cfg.OPTIMIZATION.DIV_FACTOR: 10 2023-03-31 07:55:25,881 INFO cfg.OPTIMIZATION.LR_CLIP: 1e-07 2023-03-31 07:55:25,881 INFO cfg.OPTIMIZATION.LR_WARMUP: False 2023-03-31 07:55:25,882 INFO cfg.OPTIMIZATION.WARMUP_EPOCH: 1 2023-03-31 07:55:25,882 INFO cfg.TAG: CAGroup3D 2023-03-31 07:55:25,883 INFO cfg.EXP_GROUP_PATH: sunrgbd_models 2023-03-31 07:55:25,883 INFO Loading SUNRGBD dataset 2023-03-31 07:55:26,075 INFO Total samples for SUNRGBD dataset: 5050 2023-03-31 07:55:27,373 INFO ==> Loading parameters from checkpoint ../output/sunrgbd_models/CAGroup3D/cagroup3d-win10-sunrgbd-train/ckpt/checkpoint_epoch_1.pth to CPU 2023-03-31 07:55:27,909 INFO ==> Checkpoint trained from version: pcdet+0.5.2+18bc5f5+py60edc0c 2023-03-31 07:55:27,980 INFO ==> Done (loaded 638/638) 2023-03-31 07:55:28,322 INFO *************** EPOCH 1 EVALUATION ***************** 2023-03-31 11:42:57,916 INFO *************** Performance of EPOCH 1 ***************** 2023-03-31 11:42:57,916 INFO Generate label finished(sec_per_example: 2.7025 second). 2023-03-31 11:42:57,917 INFO recall_roi_0.25: 0.000000 2023-03-31 11:42:57,917 INFO recall_rcnn_0.25: 0.000000 2023-03-31 11:42:57,918 INFO recall_roi_0.5: 0.000000 2023-03-31 11:42:57,919 INFO recall_rcnn_0.5: 0.000000 2023-03-31 11:42:57,923 INFO Average predicted number of objects(5050 samples): 55.676 2023-03-31 11:43:27,831 INFO {'bed_AP_0.25': 0.42717495560646057, 'table_AP_0.25': 6.860281428089365e-05, 'sofa_AP_0.25': 0.011460769921541214, 'chair_AP_0.25': 0.0, 'toilet_AP_0.25': 0.0, 'desk_AP_0.25': 4.534261825028807e-05, 'dresser_AP_0.25': 0.0, 'night_stand_AP_0.25': 0.0, 'bookshelf_AP_0.25': 5.094970219943207e-06, 'bathtub_AP_0.25': 0.0, 'mAP_0.25': 0.04387547820806503, 'bed_rec_0.25': 0.6135922330097088, 'table_rec_0.25': 0.008943781942078365, 'sofa_rec_0.25': 0.13875598086124402, 'chair_rec_0.25': 0.0, 'toilet_rec_0.25': 0.0, 'desk_rec_0.25': 0.019128586609989374, 'dresser_rec_0.25': 0.0, 'night_stand_rec_0.25': 0.0, 'bookshelf_rec_0.25': 0.0035460992907801418, 'bathtub_rec_0.25': 0.0, 'mAR_0.25': 0.07839666817138005, 'bed_AP_0.50': 0.07867174595594406, 'table_AP_0.50': 0.0, 'sofa_AP_0.50': 0.0, 'chair_AP_0.50': 0.0, 'toilet_AP_0.50': 0.0, 'desk_AP_0.50': 0.0, 'dresser_AP_0.50': 0.0, 'night_stand_AP_0.50': 0.0, 'bookshelf_AP_0.50': 0.0, 'bathtub_AP_0.50': 0.0, 'mAP_0.50': 0.007867174223065376, 'bed_rec_0.50': 0.2058252427184466, 'table_rec_0.50': 0.0, 'sofa_rec_0.50': 0.0, 'chair_rec_0.50': 0.0, 'toilet_rec_0.50': 0.0, 'desk_rec_0.50': 0.0, 'dresser_rec_0.50': 0.0, 'night_stand_rec_0.50': 0.0, 'bookshelf_rec_0.50': 0.0, 'bathtub_rec_0.50': 0.0, 'mAR_0.50': 0.02058252427184466} 2023-03-31 11:43:27,836 INFO Result is save to C:\PINKAMENA\CITYU\CS5182\proj\CAGroup3D\output\sunrgbd_models\CAGroup3D\cagroup3d-win10-sunrgbd-eval\eval\epoch_1\val\default 2023-03-31 11:43:27,838 INFO ****************Evaluation done.*****************