Hugging Face's logo

6DammK9
/
cagroup3d-win10-scannet

TensorBoard
English
pcdet
Model card Files Metrics Community
6DammK9 commited on
Commit
d8b057b
•
1 Parent(s): 4c8765e

train with eval

Browse files
Files changed (6) hide show
  1. ckpt/checkpoint_epoch_9.pth +3 -0
  2. eval/eval_with_train/epoch_9/val/result.pkl +3 -0
  3. eval/eval_with_train/eval_list_val.txt +1 -0
  4. eval/eval_with_train/tensorboard_val/events.out.tfevents.1680871213.DESKTOP-3FL13RB +3 -0
  5. log_train_20230402-191900.txt +930 -0
  6. tensorboard/events.out.tfevents.1680434340.DESKTOP-3FL13RB +3 -0
ckpt/checkpoint_epoch_9.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d6b910b73d996c405b80830274d59cbf60e36e5b9025e9e80cfcdedf40e4be43
3
+ size 1519000587
eval/eval_with_train/epoch_9/val/result.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ef5a5dcee9b8887db1ec4b43999a16591ec9cbc897798d0d49efde6a11cbeeef
3
+ size 30081537
eval/eval_with_train/eval_list_val.txt CHANGED
@@ -0,0 +1 @@
 
 
1
+ 9
eval/eval_with_train/tensorboard_val/events.out.tfevents.1680871213.DESKTOP-3FL13RB ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f477f591c353c9ca0f9547bbc4f4ced5c8c620edaa67f831de7fd0818a24eea1
3
+ size 4268
log_train_20230402-191900.txt ADDED
@@ -0,0 +1,930 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-04-02 19:19:00,193 INFO **********************Start logging**********************
2
+ 2023-04-02 19:19:00,193 INFO CUDA_VISIBLE_DEVICES=ALL
3
+ 2023-04-02 19:19:00,194 INFO total_batch_size: 16
4
+ 2023-04-02 19:19:00,194 INFO cfg_file cfgs/scannet_models/CAGroup3D.yaml
5
+ 2023-04-02 19:19:00,195 INFO batch_size 16
6
+ 2023-04-02 19:19:00,196 INFO epochs 9
7
+ 2023-04-02 19:19:00,196 INFO workers 4
8
+ 2023-04-02 19:19:00,197 INFO extra_tag cagroup3d-win10-scannet-train
9
+ 2023-04-02 19:19:00,197 INFO ckpt ../output/scannet_models/CAGroup3D/cagroup3d-win10-scannet-train-good/ckpt/checkpoint_epoch_8.pth
10
+ 2023-04-02 19:19:00,198 INFO pretrained_model ../output/scannet_models/CAGroup3D/cagroup3d-win10-scannet-train-good/ckpt/checkpoint_epoch_8.pth
11
+ 2023-04-02 19:19:00,199 INFO launcher pytorch
12
+ 2023-04-02 19:19:00,199 INFO tcp_port 18888
13
+ 2023-04-02 19:19:00,200 INFO sync_bn False
14
+ 2023-04-02 19:19:00,200 INFO fix_random_seed True
15
+ 2023-04-02 19:19:00,201 INFO ckpt_save_interval 1
16
+ 2023-04-02 19:19:00,201 INFO max_ckpt_save_num 30
17
+ 2023-04-02 19:19:00,202 INFO merge_all_iters_to_one_epoch False
18
+ 2023-04-02 19:19:00,202 INFO set_cfgs None
19
+ 2023-04-02 19:19:00,203 INFO max_waiting_mins 0
20
+ 2023-04-02 19:19:00,203 INFO start_epoch 0
21
+ 2023-04-02 19:19:00,204 INFO num_epochs_to_eval 0
22
+ 2023-04-02 19:19:00,204 INFO save_to_file False
23
+ 2023-04-02 19:19:00,205 INFO cfg.ROOT_DIR: C:\CITYU\CS5182\proj\CAGroup3D
24
+ 2023-04-02 19:19:00,205 INFO cfg.LOCAL_RANK: 0
25
+ 2023-04-02 19:19:00,206 INFO cfg.CLASS_NAMES: ['cabinet', 'bed', 'chair', 'sofa', 'table', 'door', 'window', 'bookshelf', 'picture', 'counter', 'desk', 'curtain', 'refrigerator', 'showercurtrain', 'toilet', 'sink', 'bathtub', 'garbagebin']
26
+ 2023-04-02 19:19:00,207 INFO
27
+ cfg.DATA_CONFIG = edict()
28
+ 2023-04-02 19:19:00,207 INFO cfg.DATA_CONFIG.DATASET: ScannetDataset
29
+ 2023-04-02 19:19:00,208 INFO cfg.DATA_CONFIG.DATA_PATH: ../data/scannet_data/ScanNetV2
30
+ 2023-04-02 19:19:00,208 INFO cfg.DATA_CONFIG.PROCESSED_DATA_TAG: scannet_processed_data_v0_5_0
31
+ 2023-04-02 19:19:00,209 INFO cfg.DATA_CONFIG.POINT_CLOUD_RANGE: [-40, -40, -10, 40, 40, 10]
32
+ 2023-04-02 19:19:00,209 INFO
33
+ cfg.DATA_CONFIG.DATA_SPLIT = edict()
34
+ 2023-04-02 19:19:00,210 INFO cfg.DATA_CONFIG.DATA_SPLIT.train: train
35
+ 2023-04-02 19:19:00,210 INFO cfg.DATA_CONFIG.DATA_SPLIT.test: val
36
+ 2023-04-02 19:19:00,211 INFO
37
+ cfg.DATA_CONFIG.REPEAT = edict()
38
+ 2023-04-02 19:19:00,211 INFO cfg.DATA_CONFIG.REPEAT.train: 10
39
+ 2023-04-02 19:19:00,212 INFO cfg.DATA_CONFIG.REPEAT.test: 1
40
+ 2023-04-02 19:19:00,213 INFO
41
+ cfg.DATA_CONFIG.INFO_PATH = edict()
42
+ 2023-04-02 19:19:00,213 INFO cfg.DATA_CONFIG.INFO_PATH.train: ['scannet_infos_train.pkl']
43
+ 2023-04-02 19:19:00,214 INFO cfg.DATA_CONFIG.INFO_PATH.test: ['scannet_infos_val.pkl']
44
+ 2023-04-02 19:19:00,214 INFO cfg.DATA_CONFIG.GET_ITEM_LIST: ['points', 'instance_mask', 'semantic_mask']
45
+ 2023-04-02 19:19:00,215 INFO cfg.DATA_CONFIG.FILTER_EMPTY_BOXES_FOR_TRAIN: True
46
+ 2023-04-02 19:19:00,215 INFO
47
+ cfg.DATA_CONFIG.DATA_AUGMENTOR_TRAIN = edict()
48
+ 2023-04-02 19:19:00,216 INFO cfg.DATA_CONFIG.DATA_AUGMENTOR_TRAIN.DISABLE_AUG_LIST: ['placeholder']
49
+ 2023-04-02 19:19:00,216 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}]
50
+ 2023-04-02 19:19:00,217 INFO
51
+ cfg.DATA_CONFIG.DATA_AUGMENTOR_TEST = edict()
52
+ 2023-04-02 19:19:00,218 INFO cfg.DATA_CONFIG.DATA_AUGMENTOR_TEST.DISABLE_AUG_LIST: ['placeholder']
53
+ 2023-04-02 19:19:00,218 INFO cfg.DATA_CONFIG.DATA_AUGMENTOR_TEST.AUG_CONFIG_LIST: [{'NAME': 'global_alignment', 'rotation_axis': 2}]
54
+ 2023-04-02 19:19:00,219 INFO
55
+ cfg.DATA_CONFIG.DATA_AUGMENTOR = edict()
56
+ 2023-04-02 19:19:00,219 INFO cfg.DATA_CONFIG.DATA_AUGMENTOR.DISABLE_AUG_LIST: ['placeholder']
57
+ 2023-04-02 19:19:00,220 INFO cfg.DATA_CONFIG.DATA_AUGMENTOR.AUG_CONFIG_LIST: [{'NAME': 'global_alignment', 'rotation_axis': 2}]
58
+ 2023-04-02 19:19:00,220 INFO
59
+ cfg.DATA_CONFIG.POINT_FEATURE_ENCODING = edict()
60
+ 2023-04-02 19:19:00,221 INFO cfg.DATA_CONFIG.POINT_FEATURE_ENCODING.encoding_type: absolute_coordinates_encoding
61
+ 2023-04-02 19:19:00,222 INFO cfg.DATA_CONFIG.POINT_FEATURE_ENCODING.used_feature_list: ['x', 'y', 'z', 'r', 'g', 'b']
62
+ 2023-04-02 19:19:00,222 INFO cfg.DATA_CONFIG.POINT_FEATURE_ENCODING.src_feature_list: ['x', 'y', 'z', 'r', 'g', 'b']
63
+ 2023-04-02 19:19:00,223 INFO cfg.DATA_CONFIG.DATA_PROCESSOR: [{'NAME': 'mask_points_and_boxes_outside_range', 'REMOVE_OUTSIDE_BOXES': True}]
64
+ 2023-04-02 19:19:00,223 INFO cfg.DATA_CONFIG._BASE_CONFIG_: cfgs/dataset_configs/scannet_dataset.yaml
65
+ 2023-04-02 19:19:00,224 INFO cfg.VOXEL_SIZE: 0.02
66
+ 2023-04-02 19:19:00,224 INFO cfg.N_CLASSES: 18
67
+ 2023-04-02 19:19:00,224 INFO cfg.SEMANTIC_THR: 0.15
68
+ 2023-04-02 19:19:00,225 INFO
69
+ cfg.MODEL = edict()
70
+ 2023-04-02 19:19:00,225 INFO cfg.MODEL.NAME: CAGroup3D
71
+ 2023-04-02 19:19:00,226 INFO cfg.MODEL.VOXEL_SIZE: 0.02
72
+ 2023-04-02 19:19:00,226 INFO cfg.MODEL.SEMANTIC_MIN_THR: 0.05
73
+ 2023-04-02 19:19:00,227 INFO cfg.MODEL.SEMANTIC_ITER_VALUE: 0.02
74
+ 2023-04-02 19:19:00,227 INFO cfg.MODEL.SEMANTIC_THR: 0.15
75
+ 2023-04-02 19:19:00,227 INFO
76
+ cfg.MODEL.BACKBONE_3D = edict()
77
+ 2023-04-02 19:19:00,228 INFO cfg.MODEL.BACKBONE_3D.NAME: BiResNet
78
+ 2023-04-02 19:19:00,228 INFO cfg.MODEL.BACKBONE_3D.IN_CHANNELS: 3
79
+ 2023-04-02 19:19:00,229 INFO cfg.MODEL.BACKBONE_3D.OUT_CHANNELS: 64
80
+ 2023-04-02 19:19:00,229 INFO
81
+ cfg.MODEL.DENSE_HEAD = edict()
82
+ 2023-04-02 19:19:00,230 INFO cfg.MODEL.DENSE_HEAD.NAME: CAGroup3DHead
83
+ 2023-04-02 19:19:00,230 INFO cfg.MODEL.DENSE_HEAD.IN_CHANNELS: [64, 128, 256, 512]
84
+ 2023-04-02 19:19:00,231 INFO cfg.MODEL.DENSE_HEAD.OUT_CHANNELS: 64
85
+ 2023-04-02 19:19:00,231 INFO cfg.MODEL.DENSE_HEAD.SEMANTIC_THR: 0.15
86
+ 2023-04-02 19:19:00,232 INFO cfg.MODEL.DENSE_HEAD.VOXEL_SIZE: 0.02
87
+ 2023-04-02 19:19:00,233 INFO cfg.MODEL.DENSE_HEAD.N_CLASSES: 18
88
+ 2023-04-02 19:19:00,233 INFO cfg.MODEL.DENSE_HEAD.N_REG_OUTS: 6
89
+ 2023-04-02 19:19:00,233 INFO cfg.MODEL.DENSE_HEAD.CLS_KERNEL: 9
90
+ 2023-04-02 19:19:00,234 INFO cfg.MODEL.DENSE_HEAD.WITH_YAW: False
91
+ 2023-04-02 19:19:00,234 INFO cfg.MODEL.DENSE_HEAD.USE_SEM_SCORE: False
92
+ 2023-04-02 19:19:00,235 INFO cfg.MODEL.DENSE_HEAD.EXPAND_RATIO: 3
93
+ 2023-04-02 19:19:00,235 INFO
94
+ cfg.MODEL.DENSE_HEAD.ASSIGNER = edict()
95
+ 2023-04-02 19:19:00,236 INFO cfg.MODEL.DENSE_HEAD.ASSIGNER.NAME: CAGroup3DAssigner
96
+ 2023-04-02 19:19:00,237 INFO cfg.MODEL.DENSE_HEAD.ASSIGNER.LIMIT: 27
97
+ 2023-04-02 19:19:00,237 INFO cfg.MODEL.DENSE_HEAD.ASSIGNER.TOPK: 18
98
+ 2023-04-02 19:19:00,237 INFO cfg.MODEL.DENSE_HEAD.ASSIGNER.N_SCALES: 4
99
+ 2023-04-02 19:19:00,238 INFO
100
+ cfg.MODEL.DENSE_HEAD.LOSS_OFFSET = edict()
101
+ 2023-04-02 19:19:00,239 INFO cfg.MODEL.DENSE_HEAD.LOSS_OFFSET.NAME: SmoothL1Loss
102
+ 2023-04-02 19:19:00,239 INFO cfg.MODEL.DENSE_HEAD.LOSS_OFFSET.BETA: 0.04
103
+ 2023-04-02 19:19:00,240 INFO cfg.MODEL.DENSE_HEAD.LOSS_OFFSET.REDUCTION: sum
104
+ 2023-04-02 19:19:00,240 INFO cfg.MODEL.DENSE_HEAD.LOSS_OFFSET.LOSS_WEIGHT: 1.0
105
+ 2023-04-02 19:19:00,240 INFO
106
+ cfg.MODEL.DENSE_HEAD.LOSS_BBOX = edict()
107
+ 2023-04-02 19:19:00,241 INFO cfg.MODEL.DENSE_HEAD.LOSS_BBOX.NAME: IoU3DLoss
108
+ 2023-04-02 19:19:00,242 INFO cfg.MODEL.DENSE_HEAD.LOSS_BBOX.WITH_YAW: False
109
+ 2023-04-02 19:19:00,242 INFO cfg.MODEL.DENSE_HEAD.LOSS_BBOX.LOSS_WEIGHT: 1.0
110
+ 2023-04-02 19:19:00,243 INFO
111
+ cfg.MODEL.DENSE_HEAD.NMS_CONFIG = edict()
112
+ 2023-04-02 19:19:00,243 INFO cfg.MODEL.DENSE_HEAD.NMS_CONFIG.SCORE_THR: 0.01
113
+ 2023-04-02 19:19:00,244 INFO cfg.MODEL.DENSE_HEAD.NMS_CONFIG.NMS_PRE: 1000
114
+ 2023-04-02 19:19:00,244 INFO cfg.MODEL.DENSE_HEAD.NMS_CONFIG.IOU_THR: 0.5
115
+ 2023-04-02 19:19:00,245 INFO
116
+ cfg.MODEL.ROI_HEAD = edict()
117
+ 2023-04-02 19:19:00,246 INFO cfg.MODEL.ROI_HEAD.NAME: CAGroup3DRoIHead
118
+ 2023-04-02 19:19:00,246 INFO cfg.MODEL.ROI_HEAD.NUM_CLASSES: 18
119
+ 2023-04-02 19:19:00,247 INFO cfg.MODEL.ROI_HEAD.MIDDLE_FEATURE_SOURCE: [3]
120
+ 2023-04-02 19:19:00,247 INFO cfg.MODEL.ROI_HEAD.GRID_SIZE: 7
121
+ 2023-04-02 19:19:00,247 INFO cfg.MODEL.ROI_HEAD.VOXEL_SIZE: 0.02
122
+ 2023-04-02 19:19:00,248 INFO cfg.MODEL.ROI_HEAD.COORD_KEY: 2
123
+ 2023-04-02 19:19:00,248 INFO cfg.MODEL.ROI_HEAD.MLPS: [[64, 128, 128]]
124
+ 2023-04-02 19:19:00,249 INFO cfg.MODEL.ROI_HEAD.CODE_SIZE: 6
125
+ 2023-04-02 19:19:00,249 INFO cfg.MODEL.ROI_HEAD.ENCODE_SINCOS: False
126
+ 2023-04-02 19:19:00,250 INFO cfg.MODEL.ROI_HEAD.ROI_PER_IMAGE: 128
127
+ 2023-04-02 19:19:00,250 INFO cfg.MODEL.ROI_HEAD.ROI_FG_RATIO: 0.9
128
+ 2023-04-02 19:19:00,251 INFO cfg.MODEL.ROI_HEAD.REG_FG_THRESH: 0.3
129
+ 2023-04-02 19:19:00,251 INFO cfg.MODEL.ROI_HEAD.ROI_CONV_KERNEL: 5
130
+ 2023-04-02 19:19:00,251 INFO cfg.MODEL.ROI_HEAD.ENLARGE_RATIO: False
131
+ 2023-04-02 19:19:00,252 INFO cfg.MODEL.ROI_HEAD.USE_IOU_LOSS: False
132
+ 2023-04-02 19:19:00,252 INFO cfg.MODEL.ROI_HEAD.USE_GRID_OFFSET: False
133
+ 2023-04-02 19:19:00,253 INFO cfg.MODEL.ROI_HEAD.USE_SIMPLE_POOLING: True
134
+ 2023-04-02 19:19:00,253 INFO cfg.MODEL.ROI_HEAD.USE_CENTER_POOLING: True
135
+ 2023-04-02 19:19:00,254 INFO
136
+ cfg.MODEL.ROI_HEAD.LOSS_WEIGHTS = edict()
137
+ 2023-04-02 19:19:00,254 INFO cfg.MODEL.ROI_HEAD.LOSS_WEIGHTS.RCNN_CLS_WEIGHT: 1.0
138
+ 2023-04-02 19:19:00,254 INFO cfg.MODEL.ROI_HEAD.LOSS_WEIGHTS.RCNN_REG_WEIGHT: 1.0
139
+ 2023-04-02 19:19:00,255 INFO cfg.MODEL.ROI_HEAD.LOSS_WEIGHTS.RCNN_IOU_WEIGHT: 1.0
140
+ 2023-04-02 19:19:00,255 INFO cfg.MODEL.ROI_HEAD.LOSS_WEIGHTS.CODE_WEIGHT: [1.0, 1.0, 1.0, 1.0, 1.0, 1.0]
141
+ 2023-04-02 19:19:00,256 INFO
142
+ cfg.MODEL.POST_PROCESSING = edict()
143
+ 2023-04-02 19:19:00,256 INFO cfg.MODEL.POST_PROCESSING.RECALL_THRESH_LIST: [0.25, 0.5]
144
+ 2023-04-02 19:19:00,257 INFO cfg.MODEL.POST_PROCESSING.EVAL_METRIC: scannet
145
+ 2023-04-02 19:19:00,257 INFO
146
+ cfg.OPTIMIZATION = edict()
147
+ 2023-04-02 19:19:00,258 INFO cfg.OPTIMIZATION.BATCH_SIZE_PER_GPU: 16
148
+ 2023-04-02 19:19:00,258 INFO cfg.OPTIMIZATION.NUM_EPOCHS: 1
149
+ 2023-04-02 19:19:00,259 INFO cfg.OPTIMIZATION.OPTIMIZER: adamW
150
+ 2023-04-02 19:19:00,259 INFO cfg.OPTIMIZATION.LR: 0.001
151
+ 2023-04-02 19:19:00,259 INFO cfg.OPTIMIZATION.WEIGHT_DECAY: 0.0001
152
+ 2023-04-02 19:19:00,260 INFO cfg.OPTIMIZATION.DECAY_STEP_LIST: [7, 9]
153
+ 2023-04-02 19:19:00,261 INFO cfg.OPTIMIZATION.LR_DECAY: 0.1
154
+ 2023-04-02 19:19:00,261 INFO cfg.OPTIMIZATION.GRAD_NORM_CLIP: 10
155
+ 2023-04-02 19:19:00,261 INFO cfg.OPTIMIZATION.PCT_START: 0.4
156
+ 2023-04-02 19:19:00,262 INFO cfg.OPTIMIZATION.DIV_FACTOR: 10
157
+ 2023-04-02 19:19:00,262 INFO cfg.OPTIMIZATION.LR_CLIP: 1e-07
158
+ 2023-04-02 19:19:00,263 INFO cfg.OPTIMIZATION.LR_WARMUP: False
159
+ 2023-04-02 19:19:00,263 INFO cfg.OPTIMIZATION.WARMUP_EPOCH: 1
160
+ 2023-04-02 19:19:00,264 INFO cfg.TAG: CAGroup3D
161
+ 2023-04-02 19:19:00,264 INFO cfg.EXP_GROUP_PATH: scannet_models
162
+ 2023-04-02 19:19:00,295 INFO Loading SCANNET dataset
163
+ 2023-04-02 19:19:00,413 INFO Total samples for SCANNET dataset: 1201
164
+ 2023-04-02 19:19:03,525 INFO ==> Loading parameters from checkpoint ../output/scannet_models/CAGroup3D/cagroup3d-win10-scannet-train-good/ckpt/checkpoint_epoch_8.pth to CPU
165
+ 2023-04-02 19:19:04,589 INFO ==> Checkpoint trained from version: pcdet+0.5.2+0000000
166
+ 2023-04-02 19:19:04,732 INFO ==> Done (loaded 838/838)
167
+ 2023-04-02 19:19:04,914 INFO ==> Loading parameters from checkpoint ../output/scannet_models/CAGroup3D/cagroup3d-win10-scannet-train-good/ckpt/checkpoint_epoch_8.pth to CPU
168
+ 2023-04-02 19:19:06,073 INFO ==> Loading optimizer parameters from checkpoint ../output/scannet_models/CAGroup3D/cagroup3d-win10-scannet-train-good/ckpt/checkpoint_epoch_8.pth to CPU
169
+ 2023-04-02 19:19:06,413 INFO ==> Done
170
+ 2023-04-02 19:19:06,797 INFO DistributedDataParallel(
171
+ (module): CAGroup3D(
172
+ (vfe): None
173
+ (backbone_3d): BiResNet(
174
+ (conv1): Sequential(
175
+ (0): MinkowskiConvolution(in=3, out=64, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1])
176
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
177
+ (2): MinkowskiReLU()
178
+ (3): MinkowskiConvolution(in=64, out=64, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1])
179
+ (4): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
180
+ (5): MinkowskiReLU()
181
+ )
182
+ (relu): MinkowskiReLU()
183
+ (layer1): Sequential(
184
+ (0): BasicBlock(
185
+ (conv1): MinkowskiConvolution(in=64, out=64, kernel_size=[3, 3, 3], stride=[2, 2, 2], dilation=[1, 1, 1])
186
+ (norm1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
187
+ (conv2): MinkowskiConvolution(in=64, out=64, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1])
188
+ (norm2): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
189
+ (relu): MinkowskiReLU()
190
+ (downsample): Sequential(
191
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[1, 1, 1], stride=[2, 2, 2], dilation=[1, 1, 1])
192
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
193
+ )
194
+ )
195
+ (1): BasicBlock(
196
+ (conv1): MinkowskiConvolution(in=64, out=64, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1])
197
+ (norm1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
198
+ (conv2): MinkowskiConvolution(in=64, out=64, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1])
199
+ (norm2): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
200
+ (relu): MinkowskiReLU()
201
+ )
202
+ )
203
+ (layer2): Sequential(
204
+ (0): BasicBlock(
205
+ (conv1): MinkowskiConvolution(in=64, out=128, kernel_size=[3, 3, 3], stride=[2, 2, 2], dilation=[1, 1, 1])
206
+ (norm1): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
207
+ (conv2): MinkowskiConvolution(in=128, out=128, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1])
208
+ (norm2): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
209
+ (relu): MinkowskiReLU()
210
+ (downsample): Sequential(
211
+ (0): MinkowskiConvolution(in=64, out=128, kernel_size=[1, 1, 1], stride=[2, 2, 2], dilation=[1, 1, 1])
212
+ (1): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
213
+ )
214
+ )
215
+ (1): BasicBlock(
216
+ (conv1): MinkowskiConvolution(in=128, out=128, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1])
217
+ (norm1): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
218
+ (conv2): MinkowskiConvolution(in=128, out=128, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1])
219
+ (norm2): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
220
+ (relu): MinkowskiReLU()
221
+ )
222
+ )
223
+ (layer3): Sequential(
224
+ (0): BasicBlock(
225
+ (conv1): MinkowskiConvolution(in=128, out=256, kernel_size=[3, 3, 3], stride=[2, 2, 2], dilation=[1, 1, 1])
226
+ (norm1): MinkowskiBatchNorm(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
227
+ (conv2): MinkowskiConvolution(in=256, out=256, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1])
228
+ (norm2): MinkowskiBatchNorm(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
229
+ (relu): MinkowskiReLU()
230
+ (downsample): Sequential(
231
+ (0): MinkowskiConvolution(in=128, out=256, kernel_size=[1, 1, 1], stride=[2, 2, 2], dilation=[1, 1, 1])
232
+ (1): MinkowskiBatchNorm(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
233
+ )
234
+ )
235
+ (1): BasicBlock(
236
+ (conv1): MinkowskiConvolution(in=256, out=256, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1])
237
+ (norm1): MinkowskiBatchNorm(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
238
+ (conv2): MinkowskiConvolution(in=256, out=256, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1])
239
+ (norm2): MinkowskiBatchNorm(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
240
+ (relu): MinkowskiReLU()
241
+ )
242
+ )
243
+ (layer4): Sequential(
244
+ (0): BasicBlock(
245
+ (conv1): MinkowskiConvolution(in=256, out=512, kernel_size=[3, 3, 3], stride=[2, 2, 2], dilation=[1, 1, 1])
246
+ (norm1): MinkowskiBatchNorm(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
247
+ (conv2): MinkowskiConvolution(in=512, out=512, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1])
248
+ (norm2): MinkowskiBatchNorm(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
249
+ (relu): MinkowskiReLU()
250
+ (downsample): Sequential(
251
+ (0): MinkowskiConvolution(in=256, out=512, kernel_size=[1, 1, 1], stride=[2, 2, 2], dilation=[1, 1, 1])
252
+ (1): MinkowskiBatchNorm(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
253
+ )
254
+ )
255
+ (1): BasicBlock(
256
+ (conv1): MinkowskiConvolution(in=512, out=512, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1])
257
+ (norm1): MinkowskiBatchNorm(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
258
+ (conv2): MinkowskiConvolution(in=512, out=512, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1])
259
+ (norm2): MinkowskiBatchNorm(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
260
+ (relu): MinkowskiReLU()
261
+ )
262
+ )
263
+ (compression3): Sequential(
264
+ (0): MinkowskiConvolution(in=256, out=128, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
265
+ (1): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
266
+ )
267
+ (compression4): Sequential(
268
+ (0): MinkowskiConvolution(in=512, out=128, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
269
+ (1): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
270
+ )
271
+ (down3): Sequential(
272
+ (0): MinkowskiConvolution(in=128, out=256, kernel_size=[3, 3, 3], stride=[2, 2, 2], dilation=[1, 1, 1])
273
+ (1): MinkowskiBatchNorm(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
274
+ )
275
+ (down4): Sequential(
276
+ (0): MinkowskiConvolution(in=128, out=256, kernel_size=[3, 3, 3], stride=[2, 2, 2], dilation=[1, 1, 1])
277
+ (1): MinkowskiBatchNorm(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
278
+ (2): MinkowskiReLU()
279
+ (3): MinkowskiConvolution(in=256, out=512, kernel_size=[3, 3, 3], stride=[2, 2, 2], dilation=[1, 1, 1])
280
+ (4): MinkowskiBatchNorm(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
281
+ )
282
+ (layer3_): Sequential(
283
+ (0): BasicBlock(
284
+ (conv1): MinkowskiConvolution(in=128, out=128, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1])
285
+ (norm1): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
286
+ (conv2): MinkowskiConvolution(in=128, out=128, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1])
287
+ (norm2): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
288
+ (relu): MinkowskiReLU()
289
+ )
290
+ (1): BasicBlock(
291
+ (conv1): MinkowskiConvolution(in=128, out=128, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1])
292
+ (norm1): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
293
+ (conv2): MinkowskiConvolution(in=128, out=128, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1])
294
+ (norm2): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
295
+ (relu): MinkowskiReLU()
296
+ )
297
+ )
298
+ (layer4_): Sequential(
299
+ (0): BasicBlock(
300
+ (conv1): MinkowskiConvolution(in=128, out=128, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1])
301
+ (norm1): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
302
+ (conv2): MinkowskiConvolution(in=128, out=128, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1])
303
+ (norm2): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
304
+ (relu): MinkowskiReLU()
305
+ )
306
+ (1): BasicBlock(
307
+ (conv1): MinkowskiConvolution(in=128, out=128, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1])
308
+ (norm1): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
309
+ (conv2): MinkowskiConvolution(in=128, out=128, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1])
310
+ (norm2): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
311
+ (relu): MinkowskiReLU()
312
+ )
313
+ )
314
+ (layer5_): Sequential(
315
+ (0): Bottleneck(
316
+ (conv1): MinkowskiConvolution(in=128, out=128, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
317
+ (norm1): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
318
+ (conv2): MinkowskiConvolution(in=128, out=128, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1])
319
+ (norm2): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
320
+ (conv3): MinkowskiConvolution(in=128, out=256, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
321
+ (norm3): MinkowskiBatchNorm(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
322
+ (relu): MinkowskiReLU()
323
+ (downsample): Sequential(
324
+ (0): MinkowskiConvolution(in=128, out=256, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
325
+ (1): MinkowskiBatchNorm(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
326
+ )
327
+ )
328
+ )
329
+ (layer5): Sequential(
330
+ (0): Bottleneck(
331
+ (conv1): MinkowskiConvolution(in=512, out=512, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
332
+ (norm1): MinkowskiBatchNorm(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
333
+ (conv2): MinkowskiConvolution(in=512, out=512, kernel_size=[3, 3, 3], stride=[2, 2, 2], dilation=[1, 1, 1])
334
+ (norm2): MinkowskiBatchNorm(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
335
+ (conv3): MinkowskiConvolution(in=512, out=1024, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
336
+ (norm3): MinkowskiBatchNorm(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
337
+ (relu): MinkowskiReLU()
338
+ (downsample): Sequential(
339
+ (0): MinkowskiConvolution(in=512, out=1024, kernel_size=[1, 1, 1], stride=[2, 2, 2], dilation=[1, 1, 1])
340
+ (1): MinkowskiBatchNorm(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
341
+ )
342
+ )
343
+ )
344
+ (spp): DAPPM(
345
+ (scale1): Sequential(
346
+ (0): MinkowskiAvgPooling(kernel_size=[5, 5, 5], stride=[2, 2, 2], dilation=[1, 1, 1])
347
+ (1): MinkowskiBatchNorm(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
348
+ (2): MinkowskiReLU()
349
+ (3): MinkowskiConvolution(in=1024, out=128, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
350
+ )
351
+ (scale2): Sequential(
352
+ (0): MinkowskiAvgPooling(kernel_size=[9, 9, 9], stride=[4, 4, 4], dilation=[1, 1, 1])
353
+ (1): MinkowskiBatchNorm(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
354
+ (2): MinkowskiReLU()
355
+ (3): MinkowskiConvolution(in=1024, out=128, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
356
+ )
357
+ (scale3): Sequential(
358
+ (0): MinkowskiAvgPooling(kernel_size=[17, 17, 17], stride=[8, 8, 8], dilation=[1, 1, 1])
359
+ (1): MinkowskiBatchNorm(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
360
+ (2): MinkowskiReLU()
361
+ (3): MinkowskiConvolution(in=1024, out=128, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
362
+ )
363
+ (scale4): Sequential(
364
+ (0): MinkowskiAvgPooling(kernel_size=[33, 33, 33], stride=[16, 16, 16], dilation=[1, 1, 1])
365
+ (1): MinkowskiBatchNorm(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
366
+ (2): MinkowskiReLU()
367
+ (3): MinkowskiConvolution(in=1024, out=128, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
368
+ )
369
+ (scale0): Sequential(
370
+ (0): MinkowskiBatchNorm(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
371
+ (1): MinkowskiReLU()
372
+ (2): MinkowskiConvolution(in=1024, out=128, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
373
+ )
374
+ (process1): Sequential(
375
+ (0): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
376
+ (1): MinkowskiReLU()
377
+ (2): MinkowskiConvolution(in=128, out=128, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1])
378
+ )
379
+ (process2): Sequential(
380
+ (0): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
381
+ (1): MinkowskiReLU()
382
+ (2): MinkowskiConvolution(in=128, out=128, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1])
383
+ )
384
+ (process3): Sequential(
385
+ (0): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
386
+ (1): MinkowskiReLU()
387
+ (2): MinkowskiConvolution(in=128, out=128, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1])
388
+ )
389
+ (process4): Sequential(
390
+ (0): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
391
+ (1): MinkowskiReLU()
392
+ (2): MinkowskiConvolution(in=128, out=128, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1])
393
+ )
394
+ (compression): Sequential(
395
+ (0): MinkowskiBatchNorm(640, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
396
+ (1): MinkowskiReLU()
397
+ (2): MinkowskiConvolution(in=640, out=256, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
398
+ )
399
+ (shortcut): Sequential(
400
+ (0): MinkowskiBatchNorm(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
401
+ (1): MinkowskiReLU()
402
+ (2): MinkowskiConvolution(in=1024, out=256, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
403
+ )
404
+ )
405
+ (out): Sequential(
406
+ (0): MinkowskiConvolutionTranspose(in=256, out=256, kernel_size=[2, 2, 2], stride=[2, 2, 2], dilation=[1, 1, 1])
407
+ (1): MinkowskiBatchNorm(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
408
+ (2): MinkowskiReLU()
409
+ (3): MinkowskiConvolution(in=256, out=64, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
410
+ (4): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
411
+ (5): MinkowskiReLU()
412
+ )
413
+ )
414
+ (map_to_bev_module): None
415
+ (pfe): None
416
+ (backbone_2d): None
417
+ (dense_head): CAGroup3DHead(
418
+ (loss_centerness): CrossEntropy()
419
+ (loss_bbox): IoU3DLoss()
420
+ (loss_cls): FocalLoss()
421
+ (loss_sem): FocalLoss()
422
+ (loss_offset): SmoothL1Loss()
423
+ (offset_block): Sequential(
424
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
425
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
426
+ (2): MinkowskiELU()
427
+ (3): MinkowskiConvolution(in=64, out=64, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
428
+ (4): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
429
+ (5): MinkowskiELU()
430
+ (6): MinkowskiConvolution(in=64, out=3, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
431
+ )
432
+ (feature_offset): Sequential(
433
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[3, 3, 3], stride=[1, 1, 1], dilation=[1, 1, 1])
434
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
435
+ (2): MinkowskiELU()
436
+ )
437
+ (semantic_conv): MinkowskiConvolution(in=64, out=18, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
438
+ (centerness_conv): MinkowskiConvolution(in=64, out=1, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
439
+ (reg_conv): MinkowskiConvolution(in=64, out=6, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
440
+ (cls_conv): MinkowskiConvolution(in=64, out=18, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
441
+ (scales): ModuleList(
442
+ (0): Scale()
443
+ (1): Scale()
444
+ (2): Scale()
445
+ (3): Scale()
446
+ (4): Scale()
447
+ (5): Scale()
448
+ (6): Scale()
449
+ (7): Scale()
450
+ (8): Scale()
451
+ (9): Scale()
452
+ (10): Scale()
453
+ (11): Scale()
454
+ (12): Scale()
455
+ (13): Scale()
456
+ (14): Scale()
457
+ (15): Scale()
458
+ (16): Scale()
459
+ (17): Scale()
460
+ )
461
+ (cls_individual_out): ModuleList(
462
+ (0): Sequential(
463
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[9, 9, 9], stride=[1, 1, 1], dilation=[1, 1, 1])
464
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
465
+ (2): MinkowskiELU()
466
+ )
467
+ (1): Sequential(
468
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[9, 9, 9], stride=[1, 1, 1], dilation=[1, 1, 1])
469
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
470
+ (2): MinkowskiELU()
471
+ )
472
+ (2): Sequential(
473
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[9, 9, 9], stride=[1, 1, 1], dilation=[1, 1, 1])
474
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
475
+ (2): MinkowskiELU()
476
+ )
477
+ (3): Sequential(
478
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[9, 9, 9], stride=[1, 1, 1], dilation=[1, 1, 1])
479
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
480
+ (2): MinkowskiELU()
481
+ )
482
+ (4): Sequential(
483
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[9, 9, 9], stride=[1, 1, 1], dilation=[1, 1, 1])
484
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
485
+ (2): MinkowskiELU()
486
+ )
487
+ (5): Sequential(
488
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[9, 9, 9], stride=[1, 1, 1], dilation=[1, 1, 1])
489
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
490
+ (2): MinkowskiELU()
491
+ )
492
+ (6): Sequential(
493
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[9, 9, 9], stride=[1, 1, 1], dilation=[1, 1, 1])
494
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
495
+ (2): MinkowskiELU()
496
+ )
497
+ (7): Sequential(
498
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[9, 9, 9], stride=[1, 1, 1], dilation=[1, 1, 1])
499
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
500
+ (2): MinkowskiELU()
501
+ )
502
+ (8): Sequential(
503
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[9, 9, 9], stride=[1, 1, 1], dilation=[1, 1, 1])
504
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
505
+ (2): MinkowskiELU()
506
+ )
507
+ (9): Sequential(
508
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[9, 9, 9], stride=[1, 1, 1], dilation=[1, 1, 1])
509
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
510
+ (2): MinkowskiELU()
511
+ )
512
+ (10): Sequential(
513
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[9, 9, 9], stride=[1, 1, 1], dilation=[1, 1, 1])
514
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
515
+ (2): MinkowskiELU()
516
+ )
517
+ (11): Sequential(
518
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[9, 9, 9], stride=[1, 1, 1], dilation=[1, 1, 1])
519
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
520
+ (2): MinkowskiELU()
521
+ )
522
+ (12): Sequential(
523
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[9, 9, 9], stride=[1, 1, 1], dilation=[1, 1, 1])
524
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
525
+ (2): MinkowskiELU()
526
+ )
527
+ (13): Sequential(
528
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[9, 9, 9], stride=[1, 1, 1], dilation=[1, 1, 1])
529
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
530
+ (2): MinkowskiELU()
531
+ )
532
+ (14): Sequential(
533
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[9, 9, 9], stride=[1, 1, 1], dilation=[1, 1, 1])
534
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
535
+ (2): MinkowskiELU()
536
+ )
537
+ (15): Sequential(
538
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[9, 9, 9], stride=[1, 1, 1], dilation=[1, 1, 1])
539
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
540
+ (2): MinkowskiELU()
541
+ )
542
+ (16): Sequential(
543
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[9, 9, 9], stride=[1, 1, 1], dilation=[1, 1, 1])
544
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
545
+ (2): MinkowskiELU()
546
+ )
547
+ (17): Sequential(
548
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[9, 9, 9], stride=[1, 1, 1], dilation=[1, 1, 1])
549
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
550
+ (2): MinkowskiELU()
551
+ )
552
+ )
553
+ (cls_individual_up): ModuleList(
554
+ (0): ModuleList(
555
+ (0): MinkowskiGenerativeConvolutionTranspose(in=64, out=64, kernel_size=[3, 3, 3], stride=[3, 3, 3], dilation=[1, 1, 1])
556
+ (1): Sequential(
557
+ (0): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
558
+ (1): MinkowskiELU()
559
+ )
560
+ )
561
+ (1): ModuleList(
562
+ (0): MinkowskiGenerativeConvolutionTranspose(in=64, out=64, kernel_size=[3, 3, 3], stride=[3, 3, 3], dilation=[1, 1, 1])
563
+ (1): Sequential(
564
+ (0): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
565
+ (1): MinkowskiELU()
566
+ )
567
+ )
568
+ (2): ModuleList(
569
+ (0): MinkowskiGenerativeConvolutionTranspose(in=64, out=64, kernel_size=[3, 3, 3], stride=[3, 3, 3], dilation=[1, 1, 1])
570
+ (1): Sequential(
571
+ (0): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
572
+ (1): MinkowskiELU()
573
+ )
574
+ )
575
+ (3): ModuleList(
576
+ (0): MinkowskiGenerativeConvolutionTranspose(in=64, out=64, kernel_size=[3, 3, 3], stride=[3, 3, 3], dilation=[1, 1, 1])
577
+ (1): Sequential(
578
+ (0): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
579
+ (1): MinkowskiELU()
580
+ )
581
+ )
582
+ (4): ModuleList(
583
+ (0): MinkowskiGenerativeConvolutionTranspose(in=64, out=64, kernel_size=[3, 3, 3], stride=[3, 3, 3], dilation=[1, 1, 1])
584
+ (1): Sequential(
585
+ (0): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
586
+ (1): MinkowskiELU()
587
+ )
588
+ )
589
+ (5): ModuleList(
590
+ (0): MinkowskiGenerativeConvolutionTranspose(in=64, out=64, kernel_size=[3, 3, 3], stride=[3, 3, 3], dilation=[1, 1, 1])
591
+ (1): Sequential(
592
+ (0): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
593
+ (1): MinkowskiELU()
594
+ )
595
+ )
596
+ (6): ModuleList(
597
+ (0): MinkowskiGenerativeConvolutionTranspose(in=64, out=64, kernel_size=[3, 3, 3], stride=[3, 3, 3], dilation=[1, 1, 1])
598
+ (1): Sequential(
599
+ (0): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
600
+ (1): MinkowskiELU()
601
+ )
602
+ )
603
+ (7): ModuleList(
604
+ (0): MinkowskiGenerativeConvolutionTranspose(in=64, out=64, kernel_size=[3, 3, 3], stride=[3, 3, 3], dilation=[1, 1, 1])
605
+ (1): Sequential(
606
+ (0): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
607
+ (1): MinkowskiELU()
608
+ )
609
+ )
610
+ (8): ModuleList(
611
+ (0): MinkowskiGenerativeConvolutionTranspose(in=64, out=64, kernel_size=[3, 3, 3], stride=[3, 3, 3], dilation=[1, 1, 1])
612
+ (1): Sequential(
613
+ (0): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
614
+ (1): MinkowskiELU()
615
+ )
616
+ )
617
+ (9): ModuleList(
618
+ (0): MinkowskiGenerativeConvolutionTranspose(in=64, out=64, kernel_size=[3, 3, 3], stride=[3, 3, 3], dilation=[1, 1, 1])
619
+ (1): Sequential(
620
+ (0): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
621
+ (1): MinkowskiELU()
622
+ )
623
+ )
624
+ (10): ModuleList(
625
+ (0): MinkowskiGenerativeConvolutionTranspose(in=64, out=64, kernel_size=[3, 3, 3], stride=[3, 3, 3], dilation=[1, 1, 1])
626
+ (1): Sequential(
627
+ (0): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
628
+ (1): MinkowskiELU()
629
+ )
630
+ )
631
+ (11): ModuleList(
632
+ (0): MinkowskiGenerativeConvolutionTranspose(in=64, out=64, kernel_size=[3, 3, 3], stride=[3, 3, 3], dilation=[1, 1, 1])
633
+ (1): Sequential(
634
+ (0): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
635
+ (1): MinkowskiELU()
636
+ )
637
+ )
638
+ (12): ModuleList(
639
+ (0): MinkowskiGenerativeConvolutionTranspose(in=64, out=64, kernel_size=[3, 3, 3], stride=[3, 3, 3], dilation=[1, 1, 1])
640
+ (1): Sequential(
641
+ (0): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
642
+ (1): MinkowskiELU()
643
+ )
644
+ )
645
+ (13): ModuleList(
646
+ (0): MinkowskiGenerativeConvolutionTranspose(in=64, out=64, kernel_size=[3, 3, 3], stride=[3, 3, 3], dilation=[1, 1, 1])
647
+ (1): Sequential(
648
+ (0): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
649
+ (1): MinkowskiELU()
650
+ )
651
+ )
652
+ (14): ModuleList(
653
+ (0): MinkowskiGenerativeConvolutionTranspose(in=64, out=64, kernel_size=[3, 3, 3], stride=[3, 3, 3], dilation=[1, 1, 1])
654
+ (1): Sequential(
655
+ (0): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
656
+ (1): MinkowskiELU()
657
+ )
658
+ )
659
+ (15): ModuleList(
660
+ (0): MinkowskiGenerativeConvolutionTranspose(in=64, out=64, kernel_size=[3, 3, 3], stride=[3, 3, 3], dilation=[1, 1, 1])
661
+ (1): Sequential(
662
+ (0): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
663
+ (1): MinkowskiELU()
664
+ )
665
+ )
666
+ (16): ModuleList(
667
+ (0): MinkowskiGenerativeConvolutionTranspose(in=64, out=64, kernel_size=[3, 3, 3], stride=[3, 3, 3], dilation=[1, 1, 1])
668
+ (1): Sequential(
669
+ (0): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
670
+ (1): MinkowskiELU()
671
+ )
672
+ )
673
+ (17): ModuleList(
674
+ (0): MinkowskiGenerativeConvolutionTranspose(in=64, out=64, kernel_size=[3, 3, 3], stride=[3, 3, 3], dilation=[1, 1, 1])
675
+ (1): Sequential(
676
+ (0): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
677
+ (1): MinkowskiELU()
678
+ )
679
+ )
680
+ )
681
+ (cls_individual_fuse): ModuleList(
682
+ (0): Sequential(
683
+ (0): MinkowskiConvolution(in=128, out=64, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
684
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
685
+ (2): MinkowskiELU()
686
+ )
687
+ (1): Sequential(
688
+ (0): MinkowskiConvolution(in=128, out=64, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
689
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
690
+ (2): MinkowskiELU()
691
+ )
692
+ (2): Sequential(
693
+ (0): MinkowskiConvolution(in=128, out=64, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
694
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
695
+ (2): MinkowskiELU()
696
+ )
697
+ (3): Sequential(
698
+ (0): MinkowskiConvolution(in=128, out=64, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
699
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
700
+ (2): MinkowskiELU()
701
+ )
702
+ (4): Sequential(
703
+ (0): MinkowskiConvolution(in=128, out=64, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
704
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
705
+ (2): MinkowskiELU()
706
+ )
707
+ (5): Sequential(
708
+ (0): MinkowskiConvolution(in=128, out=64, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
709
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
710
+ (2): MinkowskiELU()
711
+ )
712
+ (6): Sequential(
713
+ (0): MinkowskiConvolution(in=128, out=64, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
714
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
715
+ (2): MinkowskiELU()
716
+ )
717
+ (7): Sequential(
718
+ (0): MinkowskiConvolution(in=128, out=64, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
719
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
720
+ (2): MinkowskiELU()
721
+ )
722
+ (8): Sequential(
723
+ (0): MinkowskiConvolution(in=128, out=64, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
724
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
725
+ (2): MinkowskiELU()
726
+ )
727
+ (9): Sequential(
728
+ (0): MinkowskiConvolution(in=128, out=64, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
729
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
730
+ (2): MinkowskiELU()
731
+ )
732
+ (10): Sequential(
733
+ (0): MinkowskiConvolution(in=128, out=64, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
734
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
735
+ (2): MinkowskiELU()
736
+ )
737
+ (11): Sequential(
738
+ (0): MinkowskiConvolution(in=128, out=64, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
739
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
740
+ (2): MinkowskiELU()
741
+ )
742
+ (12): Sequential(
743
+ (0): MinkowskiConvolution(in=128, out=64, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
744
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
745
+ (2): MinkowskiELU()
746
+ )
747
+ (13): Sequential(
748
+ (0): MinkowskiConvolution(in=128, out=64, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
749
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
750
+ (2): MinkowskiELU()
751
+ )
752
+ (14): Sequential(
753
+ (0): MinkowskiConvolution(in=128, out=64, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
754
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
755
+ (2): MinkowskiELU()
756
+ )
757
+ (15): Sequential(
758
+ (0): MinkowskiConvolution(in=128, out=64, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
759
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
760
+ (2): MinkowskiELU()
761
+ )
762
+ (16): Sequential(
763
+ (0): MinkowskiConvolution(in=128, out=64, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
764
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
765
+ (2): MinkowskiELU()
766
+ )
767
+ (17): Sequential(
768
+ (0): MinkowskiConvolution(in=128, out=64, kernel_size=[1, 1, 1], stride=[1, 1, 1], dilation=[1, 1, 1])
769
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
770
+ (2): MinkowskiELU()
771
+ )
772
+ )
773
+ (cls_individual_expand_out): ModuleList(
774
+ (0): Sequential(
775
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[5, 5, 5], stride=[1, 1, 1], dilation=[1, 1, 1])
776
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
777
+ (2): MinkowskiELU()
778
+ )
779
+ (1): Sequential(
780
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[5, 5, 5], stride=[1, 1, 1], dilation=[1, 1, 1])
781
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
782
+ (2): MinkowskiELU()
783
+ )
784
+ (2): Sequential(
785
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[5, 5, 5], stride=[1, 1, 1], dilation=[1, 1, 1])
786
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
787
+ (2): MinkowskiELU()
788
+ )
789
+ (3): Sequential(
790
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[5, 5, 5], stride=[1, 1, 1], dilation=[1, 1, 1])
791
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
792
+ (2): MinkowskiELU()
793
+ )
794
+ (4): Sequential(
795
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[5, 5, 5], stride=[1, 1, 1], dilation=[1, 1, 1])
796
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
797
+ (2): MinkowskiELU()
798
+ )
799
+ (5): Sequential(
800
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[5, 5, 5], stride=[1, 1, 1], dilation=[1, 1, 1])
801
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
802
+ (2): MinkowskiELU()
803
+ )
804
+ (6): Sequential(
805
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[5, 5, 5], stride=[1, 1, 1], dilation=[1, 1, 1])
806
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
807
+ (2): MinkowskiELU()
808
+ )
809
+ (7): Sequential(
810
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[5, 5, 5], stride=[1, 1, 1], dilation=[1, 1, 1])
811
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
812
+ (2): MinkowskiELU()
813
+ )
814
+ (8): Sequential(
815
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[5, 5, 5], stride=[1, 1, 1], dilation=[1, 1, 1])
816
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
817
+ (2): MinkowskiELU()
818
+ )
819
+ (9): Sequential(
820
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[5, 5, 5], stride=[1, 1, 1], dilation=[1, 1, 1])
821
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
822
+ (2): MinkowskiELU()
823
+ )
824
+ (10): Sequential(
825
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[5, 5, 5], stride=[1, 1, 1], dilation=[1, 1, 1])
826
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
827
+ (2): MinkowskiELU()
828
+ )
829
+ (11): Sequential(
830
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[5, 5, 5], stride=[1, 1, 1], dilation=[1, 1, 1])
831
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
832
+ (2): MinkowskiELU()
833
+ )
834
+ (12): Sequential(
835
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[5, 5, 5], stride=[1, 1, 1], dilation=[1, 1, 1])
836
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
837
+ (2): MinkowskiELU()
838
+ )
839
+ (13): Sequential(
840
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[5, 5, 5], stride=[1, 1, 1], dilation=[1, 1, 1])
841
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
842
+ (2): MinkowskiELU()
843
+ )
844
+ (14): Sequential(
845
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[5, 5, 5], stride=[1, 1, 1], dilation=[1, 1, 1])
846
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
847
+ (2): MinkowskiELU()
848
+ )
849
+ (15): Sequential(
850
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[5, 5, 5], stride=[1, 1, 1], dilation=[1, 1, 1])
851
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
852
+ (2): MinkowskiELU()
853
+ )
854
+ (16): Sequential(
855
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[5, 5, 5], stride=[1, 1, 1], dilation=[1, 1, 1])
856
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
857
+ (2): MinkowskiELU()
858
+ )
859
+ (17): Sequential(
860
+ (0): MinkowskiConvolution(in=64, out=64, kernel_size=[5, 5, 5], stride=[1, 1, 1], dilation=[1, 1, 1])
861
+ (1): MinkowskiBatchNorm(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
862
+ (2): MinkowskiELU()
863
+ )
864
+ )
865
+ )
866
+ (point_head): None
867
+ (roi_head): CAGroup3DRoIHead(
868
+ (proposal_target_layer): ProposalTargetLayer()
869
+ (reg_loss_func): WeightedSmoothL1Loss()
870
+ (roi_grid_pool_layers): ModuleList(
871
+ (0): SimplePoolingLayer(
872
+ (grid_conv): MinkowskiConvolution(in=64, out=128, kernel_size=[5, 5, 5], stride=[1, 1, 1], dilation=[1, 1, 1])
873
+ (grid_bn): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
874
+ (grid_relu): MinkowskiELU()
875
+ (pooling_conv): MinkowskiConvolution(in=128, out=128, kernel_size=[7, 7, 7], stride=[1, 1, 1], dilation=[1, 1, 1])
876
+ (pooling_bn): MinkowskiBatchNorm(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
877
+ )
878
+ )
879
+ (reg_fc_layers): Sequential(
880
+ (0): Linear(in_features=128, out_features=256, bias=False)
881
+ (1): BatchNorm1d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
882
+ (2): ReLU()
883
+ (3): Dropout(p=0.3, inplace=False)
884
+ (4): Linear(in_features=256, out_features=256, bias=False)
885
+ (5): BatchNorm1d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
886
+ (6): ReLU()
887
+ )
888
+ (reg_pred_layer): Linear(in_features=256, out_features=6, bias=True)
889
+ )
890
+ )
891
+ )
892
+ 2023-04-02 19:19:06,861 INFO **********************Start training scannet_models/CAGroup3D(cagroup3d-win10-scannet-train)**********************
893
+ 2023-04-03 02:48:23,235 INFO Epoch [ 9][ 50]/[ 751] : lr: 1.000e-04, sem_thr: 0.05, loss_centerness: 0.6437654900550842, loss_bbox: 0.24313656240701675, loss_cls: 0.12552095234394073, loss_sem: 0.0908023527264595, loss_vote: 0.4054640585184097, one_stage_loss: 1.5086894202232362, rcnn_loss_reg: 0.18907397538423537, loss_two_stage: 0.18907397538423537,
894
+ 2023-04-03 10:39:11,381 INFO Epoch [ 9][ 100]/[ 751] : lr: 1.000e-04, sem_thr: 0.05, loss_centerness: 0.6452190101146698, loss_bbox: 0.2506902211904526, loss_cls: 0.13058093473315238, loss_sem: 0.09085427075624466, loss_vote: 0.3957407087087631, one_stage_loss: 1.5130851411819457, rcnn_loss_reg: 0.18459385246038437, loss_two_stage: 0.18459385246038437,
895
+ 2023-04-03 18:37:23,817 INFO Epoch [ 9][ 150]/[ 751] : lr: 1.000e-04, sem_thr: 0.05, loss_centerness: 0.6455395030975342, loss_bbox: 0.26197255998849867, loss_cls: 0.13516157254576683, loss_sem: 0.09203423753380775, loss_vote: 0.3918899363279343, one_stage_loss: 1.5265977954864502, rcnn_loss_reg: 0.19318852871656417, loss_two_stage: 0.19318852871656417,
896
+ 2023-04-04 02:50:19,460 INFO Epoch [ 9][ 200]/[ 751] : lr: 1.000e-04, sem_thr: 0.05, loss_centerness: 0.6455828678607941, loss_bbox: 0.26617022305727006, loss_cls: 0.13564770326018333, loss_sem: 0.09159276977181435, loss_vote: 0.3895376515388489, one_stage_loss: 1.5285312223434449, rcnn_loss_reg: 0.19834407955408095, loss_two_stage: 0.19834407955408095,
897
+ 2023-04-04 10:54:09,210 INFO Epoch [ 9][ 250]/[ 751] : lr: 1.000e-04, sem_thr: 0.05, loss_centerness: 0.6433031773567199, loss_bbox: 0.2684298923611641, loss_cls: 0.1351591469347477, loss_sem: 0.0898670071363449, loss_vote: 0.38588442504405973, one_stage_loss: 1.5226436424255372, rcnn_loss_reg: 0.19586090356111527, loss_two_stage: 0.19586090356111527,
898
+ 2023-04-04 18:54:23,838 INFO Epoch [ 9][ 300]/[ 751] : lr: 1.000e-04, sem_thr: 0.05, loss_centerness: 0.6456489896774292, loss_bbox: 0.26939463675022124, loss_cls: 0.13399980276823042, loss_sem: 0.08941386982798577, loss_vote: 0.38932142794132235, one_stage_loss: 1.5277787280082702, rcnn_loss_reg: 0.19743700653314591, loss_two_stage: 0.19743700653314591,
899
+ 2023-04-05 03:03:32,943 INFO Epoch [ 9][ 350]/[ 751] : lr: 1.000e-04, sem_thr: 0.05, loss_centerness: 0.6423018944263458, loss_bbox: 0.2762458199262619, loss_cls: 0.13613657861948014, loss_sem: 0.08952155798673629, loss_vote: 0.3827694195508957, one_stage_loss: 1.526975281238556, rcnn_loss_reg: 0.2028840947151184, loss_two_stage: 0.2028840947151184,
900
+ 2023-04-05 11:14:13,601 INFO Epoch [ 9][ 400]/[ 751] : lr: 1.000e-04, sem_thr: 0.05, loss_centerness: 0.6456181585788727, loss_bbox: 0.27737378895282744, loss_cls: 0.1374327675998211, loss_sem: 0.08987010061740876, loss_vote: 0.3778589928150177, one_stage_loss: 1.5281538224220277, rcnn_loss_reg: 0.19922083646059036, loss_two_stage: 0.19922083646059036,
901
+ 2023-04-05 19:27:24,824 INFO Epoch [ 9][ 450]/[ 751] : lr: 1.000e-04, sem_thr: 0.05, loss_centerness: 0.6425622916221618, loss_bbox: 0.27846161276102066, loss_cls: 0.1391485698521137, loss_sem: 0.09056971445679665, loss_vote: 0.3740858173370361, one_stage_loss: 1.524828016757965, rcnn_loss_reg: 0.2066230583190918, loss_two_stage: 0.2066230583190918,
902
+ 2023-04-06 03:42:50,486 INFO Epoch [ 9][ 500]/[ 751] : lr: 1.000e-04, sem_thr: 0.05, loss_centerness: 0.6431539249420166, loss_bbox: 0.29076395750045775, loss_cls: 0.14162053808569908, loss_sem: 0.09268154501914978, loss_vote: 0.37324252247810363, one_stage_loss: 1.5414625024795532, rcnn_loss_reg: 0.21089414656162261, loss_two_stage: 0.21089414656162261,
903
+ 2023-04-06 12:02:31,235 INFO Epoch [ 9][ 550]/[ 751] : lr: 1.000e-04, sem_thr: 0.05, loss_centerness: 0.6435322630405426, loss_bbox: 0.2865727955102921, loss_cls: 0.14131027206778526, loss_sem: 0.08974318355321884, loss_vote: 0.36962947726249695, one_stage_loss: 1.5307879900932313, rcnn_loss_reg: 0.21048259049654006, loss_two_stage: 0.21048259049654006,
904
+ 2023-04-06 20:02:56,351 INFO Epoch [ 9][ 600]/[ 751] : lr: 1.000e-04, sem_thr: 0.05, loss_centerness: 0.6414503383636475, loss_bbox: 0.29402961790561677, loss_cls: 0.14405513793230057, loss_sem: 0.09276293635368348, loss_vote: 0.34642710983753205, one_stage_loss: 1.5187251472473144, rcnn_loss_reg: 0.21440828204154969, loss_two_stage: 0.21440828204154969,
905
+ 2023-04-07 04:17:17,115 INFO Epoch [ 9][ 650]/[ 751] : lr: 1.000e-04, sem_thr: 0.05, loss_centerness: 0.6445023834705352, loss_bbox: 0.29507197201251983, loss_cls: 0.1420440413057804, loss_sem: 0.08918493255972862, loss_vote: 0.36039295852184294, one_stage_loss: 1.5311962819099427, rcnn_loss_reg: 0.21248085170984268, loss_two_stage: 0.21248085170984268,
906
+ 2023-04-07 12:27:17,390 INFO Epoch [ 9][ 700]/[ 751] : lr: 1.000e-04, sem_thr: 0.05, loss_centerness: 0.6422169864177704, loss_bbox: 0.29913002490997315, loss_cls: 0.1449529528617859, loss_sem: 0.09117808878421783, loss_vote: 0.35617750346660615, one_stage_loss: 1.5336555647850036, rcnn_loss_reg: 0.21863267749547957, loss_two_stage: 0.21863267749547957,
907
+ 2023-04-07 20:27:53,668 INFO Epoch [ 9][ 750]/[ 751] : lr: 1.000e-04, sem_thr: 0.05, loss_centerness: 0.6417906320095063, loss_bbox: 0.30550686359405516, loss_cls: 0.14678879588842392, loss_sem: 0.09153819352388382, loss_vote: 0.33501013278961184, one_stage_loss: 1.5206346249580383, rcnn_loss_reg: 0.21843280464410783, loss_two_stage: 0.21843280464410783,
908
+ 2023-04-07 20:40:13,681 INFO **********************End training scannet_models/CAGroup3D(cagroup3d-win10-scannet-train)**********************
909
+
910
+
911
+
912
+ 2023-04-07 20:40:13,682 INFO **********************Start evaluation scannet_models/CAGroup3D(cagroup3d-win10-scannet-train)**********************
913
+ 2023-04-07 20:40:13,683 INFO Loading SCANNET dataset
914
+ 2023-04-07 20:40:13,727 INFO Total samples for SCANNET dataset: 312
915
+ 2023-04-07 20:40:13,734 INFO ==> Loading parameters from checkpoint C:\CITYU\CS5182\proj\CAGroup3D\output\scannet_models\CAGroup3D\cagroup3d-win10-scannet-train\ckpt\checkpoint_epoch_9.pth to CPU
916
+ 2023-04-07 20:40:15,066 INFO ==> Checkpoint trained from version: pcdet+0.5.2+4ae8a35+pyde9d900
917
+ 2023-04-07 20:40:15,285 INFO ==> Done (loaded 838/838)
918
+ 2023-04-07 20:40:15,536 INFO *************** EPOCH 9 EVALUATION *****************
919
+ 2023-04-07 23:42:31,883 INFO *************** Performance of EPOCH 9 *****************
920
+ 2023-04-07 23:42:31,884 INFO Generate label finished(sec_per_example: 35.0515 second).
921
+ 2023-04-07 23:42:31,884 INFO recall_roi_0.25: 0.000000
922
+ 2023-04-07 23:42:31,885 INFO recall_rcnn_0.25: 0.000000
923
+ 2023-04-07 23:42:31,885 INFO recall_roi_0.5: 0.000000
924
+ 2023-04-07 23:42:31,886 INFO recall_rcnn_0.5: 0.000000
925
+ 2023-04-07 23:42:31,886 INFO Average predicted number of objects(312 samples): 615.288
926
+ 2023-04-07 23:42:50,866 INFO {'cabinet_AP_0.25': 0.48540377616882324, 'bed_AP_0.25': 0.8844786882400513, 'chair_AP_0.25': 0.9513316750526428, 'sofa_AP_0.25': 0.897523820400238, 'table_AP_0.25': 0.6726281046867371, 'door_AP_0.25': 0.6615963578224182, 'window_AP_0.25': 0.6129509806632996, 'bookshelf_AP_0.25': 0.6147690415382385, 'picture_AP_0.25': 0.3527411222457886, 'counter_AP_0.25': 0.6677877902984619, 'desk_AP_0.25': 0.8030293583869934, 'curtain_AP_0.25': 0.6958670020103455, 'refrigerator_AP_0.25': 0.5268049240112305, 'showercurtrain_AP_0.25': 0.7306337356567383, 'toilet_AP_0.25': 0.9988548159599304, 'sink_AP_0.25': 0.7495102286338806, 'bathtub_AP_0.25': 0.8837810754776001, 'garbagebin_AP_0.25': 0.6311063170433044, 'mAP_0.25': 0.7122666239738464, 'cabinet_rec_0.25': 0.9005376344086021, 'bed_rec_0.25': 0.9135802469135802, 'chair_rec_0.25': 0.9692982456140351, 'sofa_rec_0.25': 0.979381443298969, 'table_rec_0.25': 0.8514285714285714, 'door_rec_0.25': 0.9079229122055674, 'window_rec_0.25': 0.900709219858156, 'bookshelf_rec_0.25': 0.8701298701298701, 'picture_rec_0.25': 0.6576576576576577, 'counter_rec_0.25': 0.9423076923076923, 'desk_rec_0.25': 0.9606299212598425, 'curtain_rec_0.25': 0.8656716417910447, 'refrigerator_rec_0.25': 0.8947368421052632, 'showercurtrain_rec_0.25': 0.9642857142857143, 'toilet_rec_0.25': 1.0, 'sink_rec_0.25': 0.8367346938775511, 'bathtub_rec_0.25': 0.9032258064516129, 'garbagebin_rec_0.25': 0.8622641509433963, 'mAR_0.25': 0.8989167924742847, 'cabinet_AP_0.50': 0.3409128785133362, 'bed_AP_0.50': 0.8349310159683228, 'chair_AP_0.50': 0.8949430584907532, 'sofa_AP_0.50': 0.8081077337265015, 'table_AP_0.50': 0.6181142330169678, 'door_AP_0.50': 0.5156012773513794, 'window_AP_0.50': 0.31114932894706726, 'bookshelf_AP_0.50': 0.5378049612045288, 'picture_AP_0.50': 0.22183111310005188, 'counter_AP_0.50': 0.3777454197406769, 'desk_AP_0.50': 0.6232985854148865, 'curtain_AP_0.50': 0.40846431255340576, 'refrigerator_AP_0.50': 0.44400057196617126, 'showercurtrain_AP_0.50': 0.43062731623649597, 'toilet_AP_0.50': 0.9467570781707764, 'sink_AP_0.50': 0.5182515382766724, 'bathtub_AP_0.50': 0.8415195941925049, 'garbagebin_AP_0.50': 0.5481723546981812, 'mAP_0.50': 0.5679017901420593, 'cabinet_rec_0.50': 0.6935483870967742, 'bed_rec_0.50': 0.8641975308641975, 'chair_rec_0.50': 0.9232456140350878, 'sofa_rec_0.50': 0.9175257731958762, 'table_rec_0.50': 0.7571428571428571, 'door_rec_0.50': 0.734475374732334, 'window_rec_0.50': 0.5921985815602837, 'bookshelf_rec_0.50': 0.7662337662337663, 'picture_rec_0.50': 0.42342342342342343, 'counter_rec_0.50': 0.6153846153846154, 'desk_rec_0.50': 0.8346456692913385, 'curtain_rec_0.50': 0.6119402985074627, 'refrigerator_rec_0.50': 0.7719298245614035, 'showercurtrain_rec_0.50': 0.6071428571428571, 'toilet_rec_0.50': 0.9482758620689655, 'sink_rec_0.50': 0.6428571428571429, 'bathtub_rec_0.50': 0.8709677419354839, 'garbagebin_rec_0.50': 0.7169811320754716, 'mAR_0.50': 0.7384509140060744}
927
+ 2023-04-07 23:42:50,872 INFO Result is save to C:\CITYU\CS5182\proj\CAGroup3D\output\scannet_models\CAGroup3D\cagroup3d-win10-scannet-train\eval\eval_with_train\epoch_9\val
928
+ 2023-04-07 23:42:50,873 INFO ****************Evaluation done.*****************
929
+ 2023-04-07 23:42:51,201 INFO Epoch 9 has been evaluated
930
+ 2023-04-07 23:43:21,209 INFO **********************End evaluation scannet_models/CAGroup3D(cagroup3d-win10-scannet-train)**********************
tensorboard/events.out.tfevents.1680434340.DESKTOP-3FL13RB ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ba189ca5210fae4f752fbce1ed9366602d11ab5f358baa8c7824126e48a2ec6e
3
+ size 507716