2023-04-01 21:45:09,865 INFO **********************Start logging********************** 2023-04-01 21:45:09,866 INFO CUDA_VISIBLE_DEVICES=ALL 2023-04-01 21:45:09,866 INFO total_batch_size: 16 2023-04-01 21:45:09,867 INFO cfg_file cfgs/scannet_models/CAGroup3D.yaml 2023-04-01 21:45:09,868 INFO batch_size 16 2023-04-01 21:45:09,868 INFO workers 4 2023-04-01 21:45:09,869 INFO extra_tag cagroup3d-win10-scannet-eval 2023-04-01 21:45:09,869 INFO ckpt ../output/scannet_models/CAGroup3D/cagroup3d-win10-scannet-train/ckpt/checkpoint_epoch_8.pth 2023-04-01 21:45:09,870 INFO launcher pytorch 2023-04-01 21:45:09,870 INFO tcp_port 18888 2023-04-01 21:45:09,870 INFO set_cfgs None 2023-04-01 21:45:09,872 INFO max_waiting_mins 30 2023-04-01 21:45:09,872 INFO start_epoch 0 2023-04-01 21:45:09,872 INFO eval_tag default 2023-04-01 21:45:09,872 INFO eval_all False 2023-04-01 21:45:09,873 INFO ckpt_dir None 2023-04-01 21:45:09,873 INFO save_to_file False 2023-04-01 21:45:09,875 INFO cfg.ROOT_DIR: C:\CITYU\CS5182\proj\CAGroup3D 2023-04-01 21:45:09,876 INFO cfg.LOCAL_RANK: 0 2023-04-01 21:45:09,876 INFO cfg.CLASS_NAMES: ['cabinet', 'bed', 'chair', 'sofa', 'table', 'door', 'window', 'bookshelf', 'picture', 'counter', 'desk', 'curtain', 'refrigerator', 'showercurtrain', 'toilet', 'sink', 'bathtub', 'garbagebin'] 2023-04-01 21:45:09,876 INFO cfg.DATA_CONFIG = edict() 2023-04-01 21:45:09,878 INFO cfg.DATA_CONFIG.DATASET: ScannetDataset 2023-04-01 21:45:09,878 INFO cfg.DATA_CONFIG.DATA_PATH: ../data/scannet_data/ScanNetV2 2023-04-01 21:45:09,879 INFO cfg.DATA_CONFIG.PROCESSED_DATA_TAG: scannet_processed_data_v0_5_0 2023-04-01 21:45:09,879 INFO cfg.DATA_CONFIG.POINT_CLOUD_RANGE: [-40, -40, -10, 40, 40, 10] 2023-04-01 21:45:09,880 INFO cfg.DATA_CONFIG.DATA_SPLIT = edict() 2023-04-01 21:45:09,880 INFO cfg.DATA_CONFIG.DATA_SPLIT.train: train 2023-04-01 21:45:09,880 INFO cfg.DATA_CONFIG.DATA_SPLIT.test: val 2023-04-01 21:45:09,882 INFO cfg.DATA_CONFIG.REPEAT = edict() 2023-04-01 21:45:09,883 INFO cfg.DATA_CONFIG.REPEAT.train: 10 2023-04-01 21:45:09,883 INFO cfg.DATA_CONFIG.REPEAT.test: 1 2023-04-01 21:45:09,884 INFO cfg.DATA_CONFIG.INFO_PATH = edict() 2023-04-01 21:45:09,884 INFO cfg.DATA_CONFIG.INFO_PATH.train: ['scannet_infos_train.pkl'] 2023-04-01 21:45:09,885 INFO cfg.DATA_CONFIG.INFO_PATH.test: ['scannet_infos_val.pkl'] 2023-04-01 21:45:09,886 INFO cfg.DATA_CONFIG.GET_ITEM_LIST: ['points', 'instance_mask', 'semantic_mask'] 2023-04-01 21:45:09,886 INFO cfg.DATA_CONFIG.FILTER_EMPTY_BOXES_FOR_TRAIN: True 2023-04-01 21:45:09,887 INFO cfg.DATA_CONFIG.DATA_AUGMENTOR_TRAIN = edict() 2023-04-01 21:45:09,887 INFO cfg.DATA_CONFIG.DATA_AUGMENTOR_TRAIN.DISABLE_AUG_LIST: ['placeholder'] 2023-04-01 21:45:09,888 INFO cfg.DATA_CONFIG.DATA_AUGMENTOR_TRAIN.AUG_CONFIG_LIST: [{'NAME': 'global_alignment', 'rotation_axis': 2}, {'NAME': 'point_seg_class_mapping', 'valid_cat_ids': [3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 14, 16, 24, 28, 33, 34, 36, 39], 'max_cat_id': 40}, {'NAME': 'random_world_flip', 'ALONG_AXIS_LIST': ['x', 'y']}, {'NAME': 'random_world_rotation', 'WORLD_ROT_ANGLE': [-0.087266, 0.087266]}, {'NAME': 'random_world_scaling', 'WORLD_SCALE_RANGE': [0.9, 1.1]}, {'NAME': 'random_world_translation', 'ALONG_AXIS_LIST': ['x', 'y', 'z'], 'NOISE_TRANSLATE_STD': 0.1}] 2023-04-01 21:45:09,890 INFO cfg.DATA_CONFIG.DATA_AUGMENTOR_TEST = edict() 2023-04-01 21:45:09,890 INFO cfg.DATA_CONFIG.DATA_AUGMENTOR_TEST.DISABLE_AUG_LIST: ['placeholder'] 2023-04-01 21:45:09,890 INFO cfg.DATA_CONFIG.DATA_AUGMENTOR_TEST.AUG_CONFIG_LIST: [{'NAME': 'global_alignment', 'rotation_axis': 2}] 2023-04-01 21:45:09,890 INFO cfg.DATA_CONFIG.DATA_AUGMENTOR = edict() 2023-04-01 21:45:09,892 INFO cfg.DATA_CONFIG.DATA_AUGMENTOR.DISABLE_AUG_LIST: ['placeholder'] 2023-04-01 21:45:09,893 INFO cfg.DATA_CONFIG.DATA_AUGMENTOR.AUG_CONFIG_LIST: [{'NAME': 'global_alignment', 'rotation_axis': 2}] 2023-04-01 21:45:09,893 INFO cfg.DATA_CONFIG.POINT_FEATURE_ENCODING = edict() 2023-04-01 21:45:09,894 INFO cfg.DATA_CONFIG.POINT_FEATURE_ENCODING.encoding_type: absolute_coordinates_encoding 2023-04-01 21:45:09,895 INFO cfg.DATA_CONFIG.POINT_FEATURE_ENCODING.used_feature_list: ['x', 'y', 'z', 'r', 'g', 'b'] 2023-04-01 21:45:09,895 INFO cfg.DATA_CONFIG.POINT_FEATURE_ENCODING.src_feature_list: ['x', 'y', 'z', 'r', 'g', 'b'] 2023-04-01 21:45:09,896 INFO cfg.DATA_CONFIG.DATA_PROCESSOR: [{'NAME': 'mask_points_and_boxes_outside_range', 'REMOVE_OUTSIDE_BOXES': True}] 2023-04-01 21:45:09,896 INFO cfg.DATA_CONFIG._BASE_CONFIG_: cfgs/dataset_configs/scannet_dataset.yaml 2023-04-01 21:45:09,896 INFO cfg.VOXEL_SIZE: 0.02 2023-04-01 21:45:09,896 INFO cfg.N_CLASSES: 18 2023-04-01 21:45:09,898 INFO cfg.SEMANTIC_THR: 0.15 2023-04-01 21:45:09,898 INFO cfg.MODEL = edict() 2023-04-01 21:45:09,899 INFO cfg.MODEL.NAME: CAGroup3D 2023-04-01 21:45:09,899 INFO cfg.MODEL.VOXEL_SIZE: 0.02 2023-04-01 21:45:09,900 INFO cfg.MODEL.SEMANTIC_MIN_THR: 0.05 2023-04-01 21:45:09,900 INFO cfg.MODEL.SEMANTIC_ITER_VALUE: 0.02 2023-04-01 21:45:09,900 INFO cfg.MODEL.SEMANTIC_THR: 0.15 2023-04-01 21:45:09,901 INFO cfg.MODEL.BACKBONE_3D = edict() 2023-04-01 21:45:09,901 INFO cfg.MODEL.BACKBONE_3D.NAME: BiResNet 2023-04-01 21:45:09,902 INFO cfg.MODEL.BACKBONE_3D.IN_CHANNELS: 3 2023-04-01 21:45:09,902 INFO cfg.MODEL.BACKBONE_3D.OUT_CHANNELS: 64 2023-04-01 21:45:09,902 INFO cfg.MODEL.DENSE_HEAD = edict() 2023-04-01 21:45:09,903 INFO cfg.MODEL.DENSE_HEAD.NAME: CAGroup3DHead 2023-04-01 21:45:09,903 INFO cfg.MODEL.DENSE_HEAD.IN_CHANNELS: [64, 128, 256, 512] 2023-04-01 21:45:09,904 INFO cfg.MODEL.DENSE_HEAD.OUT_CHANNELS: 64 2023-04-01 21:45:09,904 INFO cfg.MODEL.DENSE_HEAD.SEMANTIC_THR: 0.15 2023-04-01 21:45:09,905 INFO cfg.MODEL.DENSE_HEAD.VOXEL_SIZE: 0.02 2023-04-01 21:45:09,905 INFO cfg.MODEL.DENSE_HEAD.N_CLASSES: 18 2023-04-01 21:45:09,906 INFO cfg.MODEL.DENSE_HEAD.N_REG_OUTS: 6 2023-04-01 21:45:09,906 INFO cfg.MODEL.DENSE_HEAD.CLS_KERNEL: 9 2023-04-01 21:45:09,906 INFO cfg.MODEL.DENSE_HEAD.WITH_YAW: False 2023-04-01 21:45:09,907 INFO cfg.MODEL.DENSE_HEAD.USE_SEM_SCORE: False 2023-04-01 21:45:09,907 INFO cfg.MODEL.DENSE_HEAD.EXPAND_RATIO: 3 2023-04-01 21:45:09,908 INFO cfg.MODEL.DENSE_HEAD.ASSIGNER = edict() 2023-04-01 21:45:09,908 INFO cfg.MODEL.DENSE_HEAD.ASSIGNER.NAME: CAGroup3DAssigner 2023-04-01 21:45:09,909 INFO cfg.MODEL.DENSE_HEAD.ASSIGNER.LIMIT: 27 2023-04-01 21:45:09,909 INFO cfg.MODEL.DENSE_HEAD.ASSIGNER.TOPK: 18 2023-04-01 21:45:09,909 INFO cfg.MODEL.DENSE_HEAD.ASSIGNER.N_SCALES: 4 2023-04-01 21:45:09,910 INFO cfg.MODEL.DENSE_HEAD.LOSS_OFFSET = edict() 2023-04-01 21:45:09,910 INFO cfg.MODEL.DENSE_HEAD.LOSS_OFFSET.NAME: SmoothL1Loss 2023-04-01 21:45:09,911 INFO cfg.MODEL.DENSE_HEAD.LOSS_OFFSET.BETA: 0.04 2023-04-01 21:45:09,911 INFO cfg.MODEL.DENSE_HEAD.LOSS_OFFSET.REDUCTION: sum 2023-04-01 21:45:09,912 INFO cfg.MODEL.DENSE_HEAD.LOSS_OFFSET.LOSS_WEIGHT: 1.0 2023-04-01 21:45:09,912 INFO cfg.MODEL.DENSE_HEAD.LOSS_BBOX = edict() 2023-04-01 21:45:09,913 INFO cfg.MODEL.DENSE_HEAD.LOSS_BBOX.NAME: IoU3DLoss 2023-04-01 21:45:09,913 INFO cfg.MODEL.DENSE_HEAD.LOSS_BBOX.WITH_YAW: False 2023-04-01 21:45:09,914 INFO cfg.MODEL.DENSE_HEAD.LOSS_BBOX.LOSS_WEIGHT: 1.0 2023-04-01 21:45:09,914 INFO cfg.MODEL.DENSE_HEAD.NMS_CONFIG = edict() 2023-04-01 21:45:09,915 INFO cfg.MODEL.DENSE_HEAD.NMS_CONFIG.SCORE_THR: 0.01 2023-04-01 21:45:09,915 INFO cfg.MODEL.DENSE_HEAD.NMS_CONFIG.NMS_PRE: 1000 2023-04-01 21:45:09,916 INFO cfg.MODEL.DENSE_HEAD.NMS_CONFIG.IOU_THR: 0.5 2023-04-01 21:45:09,916 INFO cfg.MODEL.ROI_HEAD = edict() 2023-04-01 21:45:09,917 INFO cfg.MODEL.ROI_HEAD.NAME: CAGroup3DRoIHead 2023-04-01 21:45:09,917 INFO cfg.MODEL.ROI_HEAD.NUM_CLASSES: 18 2023-04-01 21:45:09,918 INFO cfg.MODEL.ROI_HEAD.MIDDLE_FEATURE_SOURCE: [3] 2023-04-01 21:45:09,918 INFO cfg.MODEL.ROI_HEAD.GRID_SIZE: 7 2023-04-01 21:45:09,919 INFO cfg.MODEL.ROI_HEAD.VOXEL_SIZE: 0.02 2023-04-01 21:45:09,919 INFO cfg.MODEL.ROI_HEAD.COORD_KEY: 2 2023-04-01 21:45:09,919 INFO cfg.MODEL.ROI_HEAD.MLPS: [[64, 128, 128]] 2023-04-01 21:45:09,920 INFO cfg.MODEL.ROI_HEAD.CODE_SIZE: 6 2023-04-01 21:45:09,920 INFO cfg.MODEL.ROI_HEAD.ENCODE_SINCOS: False 2023-04-01 21:45:09,921 INFO cfg.MODEL.ROI_HEAD.ROI_PER_IMAGE: 128 2023-04-01 21:45:09,921 INFO cfg.MODEL.ROI_HEAD.ROI_FG_RATIO: 0.9 2023-04-01 21:45:09,921 INFO cfg.MODEL.ROI_HEAD.REG_FG_THRESH: 0.3 2023-04-01 21:45:09,922 INFO cfg.MODEL.ROI_HEAD.ROI_CONV_KERNEL: 5 2023-04-01 21:45:09,922 INFO cfg.MODEL.ROI_HEAD.ENLARGE_RATIO: False 2023-04-01 21:45:09,923 INFO cfg.MODEL.ROI_HEAD.USE_IOU_LOSS: False 2023-04-01 21:45:09,923 INFO cfg.MODEL.ROI_HEAD.USE_GRID_OFFSET: False 2023-04-01 21:45:09,923 INFO cfg.MODEL.ROI_HEAD.USE_SIMPLE_POOLING: True 2023-04-01 21:45:09,923 INFO cfg.MODEL.ROI_HEAD.USE_CENTER_POOLING: True 2023-04-01 21:45:09,924 INFO cfg.MODEL.ROI_HEAD.LOSS_WEIGHTS = edict() 2023-04-01 21:45:09,925 INFO cfg.MODEL.ROI_HEAD.LOSS_WEIGHTS.RCNN_CLS_WEIGHT: 1.0 2023-04-01 21:45:09,927 INFO cfg.MODEL.ROI_HEAD.LOSS_WEIGHTS.RCNN_REG_WEIGHT: 1.0 2023-04-01 21:45:09,927 INFO cfg.MODEL.ROI_HEAD.LOSS_WEIGHTS.RCNN_IOU_WEIGHT: 1.0 2023-04-01 21:45:09,928 INFO cfg.MODEL.ROI_HEAD.LOSS_WEIGHTS.CODE_WEIGHT: [1.0, 1.0, 1.0, 1.0, 1.0, 1.0] 2023-04-01 21:45:09,928 INFO cfg.MODEL.POST_PROCESSING = edict() 2023-04-01 21:45:09,928 INFO cfg.MODEL.POST_PROCESSING.RECALL_THRESH_LIST: [0.25, 0.5] 2023-04-01 21:45:09,929 INFO cfg.MODEL.POST_PROCESSING.EVAL_METRIC: scannet 2023-04-01 21:45:09,929 INFO cfg.OPTIMIZATION = edict() 2023-04-01 21:45:09,931 INFO cfg.OPTIMIZATION.BATCH_SIZE_PER_GPU: 16 2023-04-01 21:45:09,931 INFO cfg.OPTIMIZATION.NUM_EPOCHS: 1 2023-04-01 21:45:09,931 INFO cfg.OPTIMIZATION.OPTIMIZER: adamW 2023-04-01 21:45:09,932 INFO cfg.OPTIMIZATION.LR: 0.001 2023-04-01 21:45:09,932 INFO cfg.OPTIMIZATION.WEIGHT_DECAY: 0.0001 2023-04-01 21:45:09,933 INFO cfg.OPTIMIZATION.DECAY_STEP_LIST: [7, 9] 2023-04-01 21:45:09,933 INFO cfg.OPTIMIZATION.LR_DECAY: 0.1 2023-04-01 21:45:09,933 INFO cfg.OPTIMIZATION.GRAD_NORM_CLIP: 10 2023-04-01 21:45:09,934 INFO cfg.OPTIMIZATION.PCT_START: 0.4 2023-04-01 21:45:09,934 INFO cfg.OPTIMIZATION.DIV_FACTOR: 10 2023-04-01 21:45:09,935 INFO cfg.OPTIMIZATION.LR_CLIP: 1e-07 2023-04-01 21:45:09,935 INFO cfg.OPTIMIZATION.LR_WARMUP: False 2023-04-01 21:45:09,936 INFO cfg.OPTIMIZATION.WARMUP_EPOCH: 1 2023-04-01 21:45:09,936 INFO cfg.TAG: CAGroup3D 2023-04-01 21:45:09,937 INFO cfg.EXP_GROUP_PATH: scannet_models 2023-04-01 21:45:09,937 INFO Loading SCANNET dataset 2023-04-01 21:45:09,972 INFO Total samples for SCANNET dataset: 312 2023-04-01 21:45:12,816 INFO ==> Loading parameters from checkpoint ../output/scannet_models/CAGroup3D/cagroup3d-win10-scannet-train/ckpt/checkpoint_epoch_8.pth to CPU 2023-04-01 21:45:13,846 INFO ==> Checkpoint trained from version: pcdet+0.5.2+0000000 2023-04-01 21:45:13,984 INFO ==> Done (loaded 838/838) 2023-04-01 21:45:14,393 INFO *************** EPOCH 8 EVALUATION ***************** 2023-04-02 00:01:12,334 INFO *************** Performance of EPOCH 8 ***************** 2023-04-02 00:01:12,334 INFO Generate label finished(sec_per_example: 26.1406 second). 2023-04-02 00:01:12,335 INFO recall_roi_0.25: 0.000000 2023-04-02 00:01:12,336 INFO recall_rcnn_0.25: 0.000000 2023-04-02 00:01:12,336 INFO recall_roi_0.5: 0.000000 2023-04-02 00:01:12,337 INFO recall_rcnn_0.5: 0.000000 2023-04-02 00:01:12,337 INFO Average predicted number of objects(312 samples): 490.029 2023-04-02 00:01:30,129 INFO {'cabinet_AP_0.25': 0.5813855528831482, 'bed_AP_0.25': 0.8816422820091248, 'chair_AP_0.25': 0.9543977379798889, 'sofa_AP_0.25': 0.8942864537239075, 'table_AP_0.25': 0.6969487071037292, 'door_AP_0.25': 0.682966947555542, 'window_AP_0.25': 0.6282612085342407, 'bookshelf_AP_0.25': 0.7065274715423584, 'picture_AP_0.25': 0.3937835395336151, 'counter_AP_0.25': 0.7749427556991577, 'desk_AP_0.25': 0.84075927734375, 'curtain_AP_0.25': 0.7117773294448853, 'refrigerator_AP_0.25': 0.5583232641220093, 'showercurtrain_AP_0.25': 0.7335410118103027, 'toilet_AP_0.25': 1.0, 'sink_AP_0.25': 0.7548432350158691, 'bathtub_AP_0.25': 0.87208491563797, 'garbagebin_AP_0.25': 0.6607850790023804, 'mAP_0.25': 0.7404030561447144, 'cabinet_rec_0.25': 0.9112903225806451, 'bed_rec_0.25': 0.9259259259259259, 'chair_rec_0.25': 0.9714912280701754, 'sofa_rec_0.25': 0.9690721649484536, 'table_rec_0.25': 0.8485714285714285, 'door_rec_0.25': 0.9036402569593148, 'window_rec_0.25': 0.875886524822695, 'bookshelf_rec_0.25': 0.9090909090909091, 'picture_rec_0.25': 0.6441441441441441, 'counter_rec_0.25': 0.9038461538461539, 'desk_rec_0.25': 0.968503937007874, 'curtain_rec_0.25': 0.8656716417910447, 'refrigerator_rec_0.25': 0.8421052631578947, 'showercurtrain_rec_0.25': 0.9642857142857143, 'toilet_rec_0.25': 1.0, 'sink_rec_0.25': 0.8469387755102041, 'bathtub_rec_0.25': 0.9032258064516129, 'garbagebin_rec_0.25': 0.8849056603773585, 'mAR_0.25': 0.8965886587523083, 'cabinet_AP_0.50': 0.4341818690299988, 'bed_AP_0.50': 0.8374742865562439, 'chair_AP_0.50': 0.9089729189872742, 'sofa_AP_0.50': 0.8061330914497375, 'table_AP_0.50': 0.6511608958244324, 'door_AP_0.50': 0.5602805614471436, 'window_AP_0.50': 0.3997173607349396, 'bookshelf_AP_0.50': 0.5989829897880554, 'picture_AP_0.50': 0.28941845893859863, 'counter_AP_0.50': 0.4558248519897461, 'desk_AP_0.50': 0.6606391072273254, 'curtain_AP_0.50': 0.5577377676963806, 'refrigerator_AP_0.50': 0.5126804113388062, 'showercurtrain_AP_0.50': 0.48787981271743774, 'toilet_AP_0.50': 0.958685576915741, 'sink_AP_0.50': 0.5293706655502319, 'bathtub_AP_0.50': 0.7828920483589172, 'garbagebin_AP_0.50': 0.5928370952606201, 'mAP_0.50': 0.6124927997589111, 'cabinet_rec_0.50': 0.7446236559139785, 'bed_rec_0.50': 0.8765432098765432, 'chair_rec_0.50': 0.9305555555555556, 'sofa_rec_0.50': 0.9072164948453608, 'table_rec_0.50': 0.7742857142857142, 'door_rec_0.50': 0.7601713062098501, 'window_rec_0.50': 0.648936170212766, 'bookshelf_rec_0.50': 0.7792207792207793, 'picture_rec_0.50': 0.481981981981982, 'counter_rec_0.50': 0.6153846153846154, 'desk_rec_0.50': 0.84251968503937, 'curtain_rec_0.50': 0.7313432835820896, 'refrigerator_rec_0.50': 0.7719298245614035, 'showercurtrain_rec_0.50': 0.6428571428571429, 'toilet_rec_0.50': 0.9655172413793104, 'sink_rec_0.50': 0.6224489795918368, 'bathtub_rec_0.50': 0.8387096774193549, 'garbagebin_rec_0.50': 0.7754716981132076, 'mAR_0.50': 0.7616509453350477} 2023-04-02 00:01:30,138 INFO Result is save to C:\CITYU\CS5182\proj\CAGroup3D\output\scannet_models\CAGroup3D\cagroup3d-win10-scannet-eval\eval\epoch_8\val\default 2023-04-02 00:01:30,139 INFO ****************Evaluation done.*****************