Comments (11)
I had the same problem and I figured out that this error comes from pcdet/model/rpn/pillar_scatter.py.
The error occurs in line 32 and 33:
[32] indices = this_coords[:, 1] * nz + this_coords[:, 2] * nx + this_coords[:, 3]
[33] indices = indices.type(torch.long)
This casting produces negative indices which leads to an error. My fix looks like this:
indices = this_coords[:, 1].type(torch.long) * nz + this_coords[:, 2].type(torch.long) * nx + this_coords[:, 3].type(torch.long)
However, I don't understand why this_coords.type(torch.long)
before the mentioned lines doesn't fix the error.
from openpcdet.
ADN,when i set batch_size = 1,there has another problem as:
2020-03-22 19:05:12,088 INFO Start logging
2020-03-22 19:05:12,089 INFO CUDA_VISIBLE_DEVICES=ALL
2020-03-22 19:05:12,090 INFO cfg_file cfgs/pointpillar.yaml
2020-03-22 19:05:12,091 INFO data_dir None
2020-03-22 19:05:12,092 INFO batch_size 1
2020-03-22 19:05:12,092 INFO epochs 80
2020-03-22 19:05:12,093 INFO workers 4
2020-03-22 19:05:12,094 INFO extra_tag default
2020-03-22 19:05:12,094 INFO ckpt pointpillar.pth
2020-03-22 19:05:12,095 INFO mgpus False
2020-03-22 19:05:12,095 INFO launcher none
2020-03-22 19:05:12,096 INFO tcp_port 18888
2020-03-22 19:05:12,096 INFO local_rank 0
2020-03-22 19:05:12,097 INFO set_cfgs None
2020-03-22 19:05:12,098 INFO max_waiting_mins 30
2020-03-22 19:05:12,098 INFO start_epoch 0
2020-03-22 19:05:12,099 INFO eval_tag default
2020-03-22 19:05:12,100 INFO eval_all False
2020-03-22 19:05:12,100 INFO ckpt_dir None
2020-03-22 19:05:12,101 INFO save_to_file False
2020-03-22 19:05:12,101 INFO cfg.ROOT_DIR: /media/buaa/My Passport/PCDet
2020-03-22 19:05:12,102 INFO cfg.LOCAL_RANK: 0
2020-03-22 19:05:12,103 INFO cfg.CLASS_NAMES: ['Car', 'Pedestrian', 'Cyclist']
2020-03-22 19:05:12,104 INFO
cfg.DATA_CONFIG = edict()
2020-03-22 19:05:12,104 INFO cfg.DATA_CONFIG.DATASET: KittiDataset
2020-03-22 19:05:12,105 INFO cfg.DATA_CONFIG.DATA_DIR: data/kitti
2020-03-22 19:05:12,106 INFO cfg.DATA_CONFIG.FOV_POINTS_ONLY: True
2020-03-22 19:05:12,106 INFO
cfg.DATA_CONFIG.NUM_POINT_FEATURES = edict()
2020-03-22 19:05:12,107 INFO cfg.DATA_CONFIG.NUM_POINT_FEATURES.total: 4
2020-03-22 19:05:12,107 INFO cfg.DATA_CONFIG.NUM_POINT_FEATURES.use: 4
2020-03-22 19:05:12,108 INFO cfg.DATA_CONFIG.POINT_CLOUD_RANGE: [0, -39.68, -3, 69.12, 39.68, 1]
2020-03-22 19:05:12,109 INFO cfg.DATA_CONFIG.MASK_POINTS_BY_RANGE: True
2020-03-22 19:05:12,109 INFO
cfg.DATA_CONFIG.TRAIN = edict()
2020-03-22 19:05:12,110 INFO cfg.DATA_CONFIG.TRAIN.INFO_PATH: ['data/kitti/kitti_infos_train.pkl']
2020-03-22 19:05:12,111 INFO cfg.DATA_CONFIG.TRAIN.SHUFFLE_POINTS: True
2020-03-22 19:05:12,111 INFO cfg.DATA_CONFIG.TRAIN.MAX_NUMBER_OF_VOXELS: 16000
2020-03-22 19:05:12,112 INFO
cfg.DATA_CONFIG.TEST = edict()
2020-03-22 19:05:12,113 INFO cfg.DATA_CONFIG.TEST.INFO_PATH: ['data/kitti/kitti_infos_val.pkl']
2020-03-22 19:05:12,113 INFO cfg.DATA_CONFIG.TEST.SHUFFLE_POINTS: False
2020-03-22 19:05:12,114 INFO cfg.DATA_CONFIG.TEST.MAX_NUMBER_OF_VOXELS: 40000
2020-03-22 19:05:12,115 INFO
cfg.DATA_CONFIG.AUGMENTATION = edict()
2020-03-22 19:05:12,115 INFO
cfg.DATA_CONFIG.AUGMENTATION.NOISE_PER_OBJECT = edict()
2020-03-22 19:05:12,116 INFO cfg.DATA_CONFIG.AUGMENTATION.NOISE_PER_OBJECT.ENABLED: True
2020-03-22 19:05:12,116 INFO cfg.DATA_CONFIG.AUGMENTATION.NOISE_PER_OBJECT.GT_LOC_NOISE_STD: [1.0, 1.0, 0.1]
2020-03-22 19:05:12,117 INFO cfg.DATA_CONFIG.AUGMENTATION.NOISE_PER_OBJECT.GT_ROT_UNIFORM_NOISE: [-0.78539816, 0.78539816]
2020-03-22 19:05:12,118 INFO
cfg.DATA_CONFIG.AUGMENTATION.NOISE_GLOBAL_SCENE = edict()
2020-03-22 19:05:12,118 INFO cfg.DATA_CONFIG.AUGMENTATION.NOISE_GLOBAL_SCENE.ENABLED: True
2020-03-22 19:05:12,119 INFO cfg.DATA_CONFIG.AUGMENTATION.NOISE_GLOBAL_SCENE.GLOBAL_ROT_UNIFORM_NOISE: [-0.78539816, 0.78539816]
2020-03-22 19:05:12,120 INFO cfg.DATA_CONFIG.AUGMENTATION.NOISE_GLOBAL_SCENE.GLOBAL_SCALING_UNIFORM_NOISE: [0.95, 1.05]
2020-03-22 19:05:12,120 INFO
cfg.DATA_CONFIG.AUGMENTATION.DB_SAMPLER = edict()
2020-03-22 19:05:12,121 INFO cfg.DATA_CONFIG.AUGMENTATION.DB_SAMPLER.ENABLED: True
2020-03-22 19:05:12,122 INFO cfg.DATA_CONFIG.AUGMENTATION.DB_SAMPLER.DB_INFO_PATH: ['data/kitti/kitti_dbinfos_train.pkl']
2020-03-22 19:05:12,122 INFO
cfg.DATA_CONFIG.AUGMENTATION.DB_SAMPLER.PREPARE = edict()
2020-03-22 19:05:12,123 INFO cfg.DATA_CONFIG.AUGMENTATION.DB_SAMPLER.PREPARE.filter_by_difficulty: [-1]
2020-03-22 19:05:12,124 INFO cfg.DATA_CONFIG.AUGMENTATION.DB_SAMPLER.PREPARE.filter_by_min_points: ['Car:5', 'Pedestrian:5', 'Cyclist:5']
2020-03-22 19:05:12,125 INFO cfg.DATA_CONFIG.AUGMENTATION.DB_SAMPLER.RATE: 1.0
2020-03-22 19:05:12,126 INFO cfg.DATA_CONFIG.AUGMENTATION.DB_SAMPLER.SAMPLE_GROUPS: ['Car:15', 'Pedestrian:10', 'Cyclist:10']
2020-03-22 19:05:12,127 INFO cfg.DATA_CONFIG.AUGMENTATION.DB_SAMPLER.USE_ROAD_PLANE: True
2020-03-22 19:05:12,128 INFO
cfg.DATA_CONFIG.VOXEL_GENERATOR = edict()
2020-03-22 19:05:12,129 INFO cfg.DATA_CONFIG.VOXEL_GENERATOR.MAX_POINTS_PER_VOXEL: 32
2020-03-22 19:05:12,130 INFO cfg.DATA_CONFIG.VOXEL_GENERATOR.VOXEL_SIZE: [0.16, 0.16, 4]
2020-03-22 19:05:12,130 INFO
cfg.MODEL = edict()
2020-03-22 19:05:12,131 INFO cfg.MODEL.NAME: PointPillar
2020-03-22 19:05:12,132 INFO
cfg.MODEL.VFE = edict()
2020-03-22 19:05:12,133 INFO cfg.MODEL.VFE.NAME: PillarFeatureNetOld2
2020-03-22 19:05:12,133 INFO
cfg.MODEL.VFE.ARGS = edict()
2020-03-22 19:05:12,134 INFO cfg.MODEL.VFE.ARGS.use_norm: True
2020-03-22 19:05:12,135 INFO cfg.MODEL.VFE.ARGS.num_filters: [64]
2020-03-22 19:05:12,135 INFO cfg.MODEL.VFE.ARGS.with_distance: False
2020-03-22 19:05:12,136 INFO
cfg.MODEL.RPN = edict()
2020-03-22 19:05:12,137 INFO cfg.MODEL.RPN.PARAMS_FIXED: False
2020-03-22 19:05:12,137 INFO
cfg.MODEL.RPN.BACKBONE = edict()
2020-03-22 19:05:12,138 INFO cfg.MODEL.RPN.BACKBONE.NAME: PointPillarsScatter
2020-03-22 19:05:12,138 INFO
cfg.MODEL.RPN.BACKBONE.ARGS = edict()
2020-03-22 19:05:12,139 INFO
cfg.MODEL.RPN.RPN_HEAD = edict()
2020-03-22 19:05:12,140 INFO cfg.MODEL.RPN.RPN_HEAD.NAME: RPNV2
2020-03-22 19:05:12,140 INFO cfg.MODEL.RPN.RPN_HEAD.DOWNSAMPLE_FACTOR: 8
2020-03-22 19:05:12,141 INFO
cfg.MODEL.RPN.RPN_HEAD.ARGS = edict()
2020-03-22 19:05:12,142 INFO cfg.MODEL.RPN.RPN_HEAD.ARGS.use_norm: True
2020-03-22 19:05:12,142 INFO cfg.MODEL.RPN.RPN_HEAD.ARGS.concat_input: False
2020-03-22 19:05:12,143 INFO cfg.MODEL.RPN.RPN_HEAD.ARGS.num_input_features: 64
2020-03-22 19:05:12,144 INFO cfg.MODEL.RPN.RPN_HEAD.ARGS.layer_nums: [3, 5, 5]
2020-03-22 19:05:12,145 INFO cfg.MODEL.RPN.RPN_HEAD.ARGS.layer_strides: [2, 2, 2]
2020-03-22 19:05:12,145 INFO cfg.MODEL.RPN.RPN_HEAD.ARGS.num_filters: [64, 128, 256]
2020-03-22 19:05:12,146 INFO cfg.MODEL.RPN.RPN_HEAD.ARGS.upsample_strides: [1, 2, 4]
2020-03-22 19:05:12,146 INFO cfg.MODEL.RPN.RPN_HEAD.ARGS.num_upsample_filters: [128, 128, 128]
2020-03-22 19:05:12,149 INFO cfg.MODEL.RPN.RPN_HEAD.ARGS.encode_background_as_zeros: True
2020-03-22 19:05:12,149 INFO cfg.MODEL.RPN.RPN_HEAD.ARGS.use_direction_classifier: True
2020-03-22 19:05:12,150 INFO cfg.MODEL.RPN.RPN_HEAD.ARGS.num_direction_bins: 2
2020-03-22 19:05:12,151 INFO cfg.MODEL.RPN.RPN_HEAD.ARGS.dir_offset: 0.78539
2020-03-22 19:05:12,152 INFO cfg.MODEL.RPN.RPN_HEAD.ARGS.dir_limit_offset: 0.0
2020-03-22 19:05:12,152 INFO cfg.MODEL.RPN.RPN_HEAD.ARGS.use_binary_dir_classifier: False
2020-03-22 19:05:12,153 INFO
cfg.MODEL.RPN.RPN_HEAD.TARGET_CONFIG = edict()
2020-03-22 19:05:12,153 INFO cfg.MODEL.RPN.RPN_HEAD.TARGET_CONFIG.DOWNSAMPLED_FACTOR: 2
2020-03-22 19:05:12,154 INFO cfg.MODEL.RPN.RPN_HEAD.TARGET_CONFIG.BOX_CODER: ResidualCoder
2020-03-22 19:05:12,155 INFO cfg.MODEL.RPN.RPN_HEAD.TARGET_CONFIG.REGION_SIMILARITY_FN: nearest_iou_similarity
2020-03-22 19:05:12,155 INFO cfg.MODEL.RPN.RPN_HEAD.TARGET_CONFIG.SAMPLE_POS_FRACTION: -1.0
2020-03-22 19:05:12,156 INFO cfg.MODEL.RPN.RPN_HEAD.TARGET_CONFIG.SAMPLE_SIZE: 512
2020-03-22 19:05:12,157 INFO cfg.MODEL.RPN.RPN_HEAD.TARGET_CONFIG.ANCHOR_GENERATOR: [{'anchor_range': [0, -40.0, -1.78, 70.4, 40.0, -1.78], 'sizes': [[1.6, 3.9, 1.56]], 'rotations': [0, 1.57], 'matched_threshold': 0.6, 'unmatched_threshold': 0.45, 'class_name': 'Car'}, {'anchor_range': [0, -40, -0.6, 70.4, 40, -0.6], 'sizes': [[0.6, 0.8, 1.73]], 'rotations': [0, 1.57], 'matched_threshold': 0.5, 'unmatched_threshold': 0.35, 'class_name': 'Pedestrian'}, {'anchor_range': [0, -40, -0.6, 70.4, 40, -0.6], 'sizes': [[0.6, 1.76, 1.73]], 'rotations': [0, 1.57], 'matched_threshold': 0.5, 'unmatched_threshold': 0.35, 'class_name': 'Cyclist'}]
2020-03-22 19:05:12,159 INFO
cfg.MODEL.RCNN = edict()
2020-03-22 19:05:12,160 INFO cfg.MODEL.RCNN.ENABLED: False
2020-03-22 19:05:12,161 INFO
cfg.MODEL.LOSSES = edict()
2020-03-22 19:05:12,161 INFO cfg.MODEL.LOSSES.RPN_REG_LOSS: smooth-l1
2020-03-22 19:05:12,162 INFO
cfg.MODEL.LOSSES.LOSS_WEIGHTS = edict()
2020-03-22 19:05:12,162 INFO cfg.MODEL.LOSSES.LOSS_WEIGHTS.rpn_cls_weight: 1.0
2020-03-22 19:05:12,163 INFO cfg.MODEL.LOSSES.LOSS_WEIGHTS.rpn_loc_weight: 2.0
2020-03-22 19:05:12,164 INFO cfg.MODEL.LOSSES.LOSS_WEIGHTS.rpn_dir_weight: 0.2
2020-03-22 19:05:12,164 INFO cfg.MODEL.LOSSES.LOSS_WEIGHTS.code_weights: [1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0]
2020-03-22 19:05:12,165 INFO
cfg.MODEL.TRAIN = edict()
2020-03-22 19:05:12,166 INFO cfg.MODEL.TRAIN.SPLIT: train
2020-03-22 19:05:12,166 INFO
cfg.MODEL.TRAIN.OPTIMIZATION = edict()
2020-03-22 19:05:12,167 INFO cfg.MODEL.TRAIN.OPTIMIZATION.OPTIMIZER: adam_onecycle
2020-03-22 19:05:12,168 INFO cfg.MODEL.TRAIN.OPTIMIZATION.LR: 0.003
2020-03-22 19:05:12,168 INFO cfg.MODEL.TRAIN.OPTIMIZATION.WEIGHT_DECAY: 0.01
2020-03-22 19:05:12,169 INFO cfg.MODEL.TRAIN.OPTIMIZATION.MOMENTUM: 0.9
2020-03-22 19:05:12,169 INFO cfg.MODEL.TRAIN.OPTIMIZATION.MOMS: [0.95, 0.85]
2020-03-22 19:05:12,170 INFO cfg.MODEL.TRAIN.OPTIMIZATION.PCT_START: 0.4
2020-03-22 19:05:12,171 INFO cfg.MODEL.TRAIN.OPTIMIZATION.DIV_FACTOR: 10
2020-03-22 19:05:12,171 INFO cfg.MODEL.TRAIN.OPTIMIZATION.DECAY_STEP_LIST: [35, 45]
2020-03-22 19:05:12,172 INFO cfg.MODEL.TRAIN.OPTIMIZATION.LR_DECAY: 0.1
2020-03-22 19:05:12,172 INFO cfg.MODEL.TRAIN.OPTIMIZATION.LR_CLIP: 1e-07
2020-03-22 19:05:12,173 INFO cfg.MODEL.TRAIN.OPTIMIZATION.LR_WARMUP: False
2020-03-22 19:05:12,174 INFO cfg.MODEL.TRAIN.OPTIMIZATION.WARMUP_EPOCH: 1
2020-03-22 19:05:12,174 INFO cfg.MODEL.TRAIN.OPTIMIZATION.GRAD_NORM_CLIP: 10
2020-03-22 19:05:12,175 INFO
cfg.MODEL.TEST = edict()
2020-03-22 19:05:12,176 INFO cfg.MODEL.TEST.SPLIT: val
2020-03-22 19:05:12,176 INFO cfg.MODEL.TEST.NMS_TYPE: nms_gpu
2020-03-22 19:05:12,177 INFO cfg.MODEL.TEST.MULTI_CLASSES_NMS: False
2020-03-22 19:05:12,177 INFO cfg.MODEL.TEST.NMS_THRESH: 0.01
2020-03-22 19:05:12,178 INFO cfg.MODEL.TEST.SCORE_THRESH: 0.1
2020-03-22 19:05:12,179 INFO cfg.MODEL.TEST.USE_RAW_SCORE: True
2020-03-22 19:05:12,179 INFO cfg.MODEL.TEST.NMS_PRE_MAXSIZE_LAST: 4096
2020-03-22 19:05:12,180 INFO cfg.MODEL.TEST.NMS_POST_MAXSIZE_LAST: 500
2020-03-22 19:05:12,180 INFO cfg.MODEL.TEST.RECALL_THRESH_LIST: [0.5, 0.7]
2020-03-22 19:05:12,181 INFO cfg.MODEL.TEST.EVAL_METRIC: kitti
2020-03-22 19:05:12,181 INFO
cfg.MODEL.TEST.BOX_FILTER = edict()
2020-03-22 19:05:12,182 INFO cfg.MODEL.TEST.BOX_FILTER.USE_IMAGE_AREA_FILTER: True
2020-03-22 19:05:12,183 INFO cfg.MODEL.TEST.BOX_FILTER.LIMIT_RANGE: [0, -40, -3.0, 70.4, 40, 3.0]
2020-03-22 19:05:12,183 INFO cfg.TAG: pointpillar
2020-03-22 19:05:12,194 INFO Loading KITTI dataset
2020-03-22 19:05:12,815 INFO Total samples for KITTI dataset: 3769
2020-03-22 19:05:18,177 INFO ==> Loading parameters from checkpoint pointpillar.pth to GPU
2020-03-22 19:05:18,559 INFO ==> Done (loaded 127/127)
2020-03-22 19:05:18,657 INFO *************** EPOCH no_number EVALUATION *****************
eval: 0%| | 4/3769 [00:05<1:48:04, 1.72s/it]Traceback (most recent call last):
File "test.py", line 181, in
main()
File "test.py", line 177, in main
eval_single_ckpt(model, test_loader, args, eval_output_dir, logger, epoch_id)
File "test.py", line 59, in eval_single_ckpt
model, test_loader, epoch_id, logger, result_dir=eval_output_dir, save_to_file=args.save_to_file
File "/media/buaa/My Passport/PCDet/tools/eval_utils/eval_utils.py", line 46, in eval_one_epoch
pred_dicts, ret_dict = model(input_dict)
File "/usr/local/lib/python3.6/dist-packages/torch/nn/modules/module.py", line 541, in call
result = self.forward(*input, **kwargs)
File "/media/buaa/My Passport/PCDet/pcdet/models/detectors/pointpillar.py", line 34, in forward
rpn_ret_dict = self.forward_rpn(**input_dict)
File "/media/buaa/My Passport/PCDet/pcdet/models/detectors/pointpillar.py", line 22, in forward_rpn
**{'gt_boxes': kwargs.get('gt_boxes', None)}
File "/usr/local/lib/python3.6/dist-packages/torch/nn/modules/module.py", line 541, in call
result = self.forward(*input, **kwargs)
File "/media/buaa/My Passport/PCDet/pcdet/models/bbox_heads/rpn_head.py", line 292, in forward
x = self.blocksi
File "/usr/local/lib/python3.6/dist-packages/torch/nn/modules/module.py", line 541, in call
result = self.forward(*input, **kwargs)
File "/media/buaa/My Passport/PCDet/pcdet/models/model_utils/pytorch_utils.py", line 88, in forward
input = module(input)
File "/usr/local/lib/python3.6/dist-packages/torch/nn/modules/module.py", line 541, in call
result = self.forward(*input, **kwargs)
File "/usr/local/lib/python3.6/dist-packages/torch/nn/modules/conv.py", line 345, in forward
return self.conv2d_forward(input, self.weight)
File "/usr/local/lib/python3.6/dist-packages/torch/nn/modules/conv.py", line 342, in conv2d_forward
self.padding, self.dilation, self.groups)
RuntimeError: cuDNN error: CUDNN_STATUS_MAPPING_ERROR
eval: 0%| | 4/3769 [00:05<1:28:06, 1.40s/it]
from openpcdet.
I had the same problem and I figured out that this error comes from pcdet/model/rpn/pillar_scatter.py.
The error occurs in line 32 and 33:
[32]indices = this_coords[:, 1] * nz + this_coords[:, 2] * nx + this_coords[:, 3]
[33]indices = indices.type(torch.long)
This casting produces negative indices which leads to an error. My fix looks like this:
indices = this_coords[:, 1].type(torch.long) * nz + this_coords[:, 2].type(torch.long) * nx + this_coords[:, 3].type(torch.long)
However, I don't understand why
this_coords.type(torch.long)
before the mentioned lines doesn't fix the error.
It worked for me ,I am very appreciate your advice, but I also wana to know why it worked,Can i explain it for a while?
from openpcdet.
Do you get the good result?
I fix like this:
#indices = this_coords[:, 1] * nz + this_coords[:, 2] * nx + this_coords[:, 3]
#indices = indices.type(torch.long)
indices = this_coords[:, 1].type(torch.long) * nz + this_coords[:, 2].type(torch.long) * nx + this_coords[:, 3].type(torch.long)
indices = indices.type(torch.long)
But i get the wrong result:
2020-05-23 01:57:55,972 INFO cfg.MODEL.TEST.BOX_FILTER.USE_IMAGE_AREA_FILTER: True
2020-05-23 01:57:55,972 INFO cfg.MODEL.TEST.BOX_FILTER.LIMIT_RANGE: [0, -40, -3.0, 70.4, 40, 3.0]
2020-05-23 01:57:55,972 INFO cfg.TAG: pointpillar
2020-05-23 01:57:56,044 INFO Loading KITTI dataset
2020-05-23 01:57:56,331 INFO Total samples for KITTI dataset: 3769
2020-05-23 01:58:02,530 INFO ==> Loading parameters from checkpoint ./weight/pointpillar.pth to GPU
2020-05-23 01:58:03,157 INFO ==> Done (loaded 127/127)
2020-05-23 01:58:03,182 INFO *************** EPOCH no_number EVALUATION *****************
2020-05-23 02:02:46,837 INFO *************** Performance of EPOCH no_number *****************
2020-05-23 02:02:46,842 INFO Generate label finished(sec_per_example: 0.0752 second).
2020-05-23 02:02:46,844 INFO Average predicted number of objects(3769 samples): 25.871
2020-05-23 02:03:11,634 INFO Car [email protected], 0.70, 0.70:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Car [email protected], 0.70, 0.70:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Car [email protected], 0.50, 0.50:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Car [email protected], 0.50, 0.50:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Pedestrian [email protected], 0.50, 0.50:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Pedestrian [email protected], 0.50, 0.50:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Pedestrian [email protected], 0.25, 0.25:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Pedestrian [email protected], 0.25, 0.25:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Cyclist [email protected], 0.50, 0.50:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Cyclist [email protected], 0.50, 0.50:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Cyclist [email protected], 0.25, 0.25:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Cyclist [email protected], 0.25, 0.25:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
@Shaoqing26
from openpcdet.
Do you get the good result?
I fix like this:
#indices = this_coords[:, 1] * nz + this_coords[:, 2] * nx + this_coords[:, 3]
#indices = indices.type(torch.long)
indices = this_coords[:, 1].type(torch.long) * nz + this_coords[:, 2].type(torch.long) * nx + this_coords[:, 3].type(torch.long)
indices = indices.type(torch.long)
But i get the wrong result:
2020-05-23 01:57:55,972 INFO cfg.MODEL.TEST.BOX_FILTER.USE_IMAGE_AREA_FILTER: True
2020-05-23 01:57:55,972 INFO cfg.MODEL.TEST.BOX_FILTER.LIMIT_RANGE: [0, -40, -3.0, 70.4, 40, 3.0]
2020-05-23 01:57:55,972 INFO cfg.TAG: pointpillar
2020-05-23 01:57:56,044 INFO Loading KITTI dataset
2020-05-23 01:57:56,331 INFO Total samples for KITTI dataset: 3769
2020-05-23 01:58:02,530 INFO ==> Loading parameters from checkpoint ./weight/pointpillar.pth to GPU
2020-05-23 01:58:03,157 INFO ==> Done (loaded 127/127)
2020-05-23 01:58:03,182 INFO *************** EPOCH no_number EVALUATION *****************
2020-05-23 02:02:46,837 INFO *************** Performance of EPOCH no_number *****************
2020-05-23 02:02:46,842 INFO Generate label finished(sec_per_example: 0.0752 second).
2020-05-23 02:02:46,844 INFO Average predicted number of objects(3769 samples): 25.871
2020-05-23 02:03:11,634 INFO Car [email protected], 0.70, 0.70:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Car [email protected], 0.70, 0.70:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Car [email protected], 0.50, 0.50:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Car [email protected], 0.50, 0.50:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Pedestrian [email protected], 0.50, 0.50:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Pedestrian [email protected], 0.50, 0.50:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Pedestrian [email protected], 0.25, 0.25:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Pedestrian [email protected], 0.25, 0.25:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Cyclist [email protected], 0.50, 0.50:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Cyclist [email protected], 0.50, 0.50:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Cyclist [email protected], 0.25, 0.25:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Cyclist [email protected], 0.25, 0.25:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
@Shaoqing26
I got the result same as author,maybe you used wrong model path,I met the same problem as you,And the reason is my model path is wrong。
from openpcdet.
Do you get the good result?
I fix like this:
#indices = this_coords[:, 1] * nz + this_coords[:, 2] * nx + this_coords[:, 3]
#indices = indices.type(torch.long)
indices = this_coords[:, 1].type(torch.long) * nz + this_coords[:, 2].type(torch.long) * nx + this_coords[:, 3].type(torch.long)
indices = indices.type(torch.long)
But i get the wrong result:
2020-05-23 01:57:55,972 INFO cfg.MODEL.TEST.BOX_FILTER.USE_IMAGE_AREA_FILTER: True
2020-05-23 01:57:55,972 INFO cfg.MODEL.TEST.BOX_FILTER.LIMIT_RANGE: [0, -40, -3.0, 70.4, 40, 3.0]
2020-05-23 01:57:55,972 INFO cfg.TAG: pointpillar
2020-05-23 01:57:56,044 INFO Loading KITTI dataset
2020-05-23 01:57:56,331 INFO Total samples for KITTI dataset: 3769
2020-05-23 01:58:02,530 INFO ==> Loading parameters from checkpoint ./weight/pointpillar.pth to GPU
2020-05-23 01:58:03,157 INFO ==> Done (loaded 127/127)
2020-05-23 01:58:03,182 INFO *************** EPOCH no_number EVALUATION *****************
2020-05-23 02:02:46,837 INFO *************** Performance of EPOCH no_number *****************
2020-05-23 02:02:46,842 INFO Generate label finished(sec_per_example: 0.0752 second).
2020-05-23 02:02:46,844 INFO Average predicted number of objects(3769 samples): 25.871
2020-05-23 02:03:11,634 INFO Car [email protected], 0.70, 0.70:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Car [email protected], 0.70, 0.70:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Car [email protected], 0.50, 0.50:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Car [email protected], 0.50, 0.50:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Pedestrian [email protected], 0.50, 0.50:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Pedestrian [email protected], 0.50, 0.50:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Pedestrian [email protected], 0.25, 0.25:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Pedestrian [email protected], 0.25, 0.25:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Cyclist [email protected], 0.50, 0.50:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Cyclist [email protected], 0.50, 0.50:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Cyclist [email protected], 0.25, 0.25:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Cyclist [email protected], 0.25, 0.25:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
@Shaoqing26I got the result same as author,maybe you used wrong model path,I met the same problem as you,And the reason is my model path is wrong。
Can you show me the model path and the command? If it's the wrong path,there will be a FileNotFoundErro。Thanks😭
from openpcdet.
This is my model path
and i run :
python3 test.py --cfg_file cfgs/pointpillar.yaml --batch_size 4 --ckpt pointpillar.pth
Whats the matter?I have spent plenty of time on looking at code but I haven't figure out the issue.
Thanks @Shaoqing26
from openpcdet.
This is my model path
and i run :
python3 test.py --cfg_file cfgs/pointpillar.yaml --batch_size 4 --ckpt pointpillar.pth
Whats the matter?I have spent plenty of time on looking at code but I haven't figure out the issue.
Thanks @Shaoqing26
Thanks to @chreisinger, his suggested code fix works for me as well
@lan-sky, are you sure that you correctly fix the code?
The file pcdet/models/rpn/pillar_scatter.py
correction (line 33) should be like this:
# indices = this_coords[:, 1] * nz + this_coords[:, 2] * nx + this_coords[:, 3]
indices = this_coords[:, 1].type(torch.long) * nz + this_coords[:, 2].type(torch.long) * nx + this_coords[:, 3].type(torch.long)
By the way, have you tried to launch another model, e.g. Part-A2
?
from openpcdet.
This is my model path
and i run :
python3 test.py --cfg_file cfgs/pointpillar.yaml --batch_size 4 --ckpt pointpillar.pth
Whats the matter?I have spent plenty of time on looking at code but I haven't figure out the issue.
Thanks @Shaoqing26Thanks to @chreisinger, his suggested code fix works for me as well
@lan-sky, are you sure that you correctly fix the code?
The filepcdet/models/rpn/pillar_scatter.py
correction (line 33) should be like this:# indices = this_coords[:, 1] * nz + this_coords[:, 2] * nx + this_coords[:, 3] indices = this_coords[:, 1].type(torch.long) * nz + this_coords[:, 2].type(torch.long) * nx + this_coords[:, 3].type(torch.long)By the way, have you tried to launch another model, e.g.
Part-A2
?
Yes, I think I fix the code same with you:
With this modification, the program can be tested, but I cant get the correct result
@Crowbar97 How can i solve this problem, THX
from openpcdet.
This is my model path
and i run :
python3 test.py --cfg_file cfgs/pointpillar.yaml --batch_size 4 --ckpt pointpillar.pth
Whats the matter?I have spent plenty of time on looking at code but I haven't figure out the issue.
Thanks @Shaoqing26Thanks to @chreisinger, his suggested code fix works for me as well
@lan-sky, are you sure that you correctly fix the code?
The filepcdet/models/rpn/pillar_scatter.py
correction (line 33) should be like this:# indices = this_coords[:, 1] * nz + this_coords[:, 2] * nx + this_coords[:, 3] indices = this_coords[:, 1].type(torch.long) * nz + this_coords[:, 2].type(torch.long) * nx + this_coords[:, 3].type(torch.long)By the way, have you tried to launch another model, e.g.
Part-A2
?Yes, I think I fix the code same with you:
With this modification, the program can be tested, but I cant get the correct result
@Crowbar97 How can i solve this problem, THX
Zero results can be due to next issues:
- Wrong / corrupted config
.yaml
file - Wrong
test.py
script arguments - Wrong
.pth
model - Wrong installed dependencies (make sure you install the
spconv v1.0
(commit 8da6f96
) instead of the latest one) - Corrupted dataset (or may be some mess in files and directories)
- ...
@lan-sky, so I think it's good idea to reinstall this repo from scratch and try once again
from openpcdet.
This is my model path
and i run :
python3 test.py --cfg_file cfgs/pointpillar.yaml --batch_size 4 --ckpt pointpillar.pth
Whats the matter?I have spent plenty of time on looking at code but I haven't figure out the issue.
Thanks @Shaoqing26Thanks to @chreisinger, his suggested code fix works for me as well
@lan-sky, are you sure that you correctly fix the code?
The filepcdet/models/rpn/pillar_scatter.py
correction (line 33) should be like this:# indices = this_coords[:, 1] * nz + this_coords[:, 2] * nx + this_coords[:, 3] indices = this_coords[:, 1].type(torch.long) * nz + this_coords[:, 2].type(torch.long) * nx + this_coords[:, 3].type(torch.long)By the way, have you tried to launch another model, e.g.
Part-A2
?Yes, I think I fix the code same with you:
With this modification, the program can be tested, but I cant get the correct result
@Crowbar97 How can i solve this problem, THXZero results can be due to next issues:
- Wrong / corrupted config
.yaml
file- Wrong
test.py
script arguments- Wrong
.pth
model- Wrong installed dependencies (make sure you install the
spconv v1.0
(commit 8da6f96
) instead of the latest one)- Corrupted dataset (or may be some mess in files and directories)
- ...
@lan-sky, so I think it's good idea to reinstall this repo from scratch and try once again
Ok,I will try to reinstall. BTY, I used the spconv v1.1 Because I saw author was update to spconv 1.1
Thank you for you advice @Crowbar97
from openpcdet.
Related Issues (20)
- Visualizing the results
- Error when running demo.py HOT 2
- About BaseBEVBackbone HOT 2
- How to get validation loss and display it in Tensorboard HOT 4
- Abnormal demo.py result using nuscenses dataset pre-trained models HOT 3
- Ensemble or test-time augmentation? HOT 2
- free(): double free detected in tcache 2 HOT 2
- questions about GETTING_STARTED.md HOT 2
- 'EasyDict' object has no attribute 'BACKUP_DB_INFO' HOT 3
- What should we set the overlap values for custom datasets? HOT 4
- TypeError: expected str, bytes or os.PathLike object, not NoneType HOT 1
- `pillar_features` IndexError: too many indices for tensor of dimension 1 HOT 1
- Question about adam_onecycle optimizer HOT 2
- PV RCNN++ pretained model HOT 1
- [Docs] A100算力加持!书生大模型实战营第3期全面升级,趣味闯关模式等你开启
- How to change source of pypi?
- Projecting 3D bounding boxes on 2D image
- Floating Point Exception HOT 5
- Training/Testing pv-rcnn using custom data
- Custom Dataset npy file demo.py
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from openpcdet.