Open vincentujs opened 5 years ago
my trainingdata have calib image_2 label_2 velodyne and my testing1 have calib image_2 velodyne
try adding the path to your project: export PYTHONPATH=$PYTHONPATH:'your path name'
Path to your project means, sorry I can't understand what you mean? Could you please explain to me to resolve this problem!
I had the same problem. But I found my datasets not strictly following this structure
PointRCNN ├── data │ ├── KITTI │ │ ├── ImageSets │ │ ├── object │ │ │ ├──training │ │ │ ├──calib & velodyne & label_2 & image_2 & (optional: planes) │ │ │ ├──testing │ │ │ ├──calib & velodyne & image_2
and I also found that files of my datasets were incomplete. I hope this is helpful.
thanks,I find the problem which is because of the wrong of the file path
thanks,I find the problem which is because of the wrong of the file path
老哥,能不能展开说说,啥错误的文件路径?我也遇到了和你一样的错误
检查一下数据集的路径是否正确,数据集的排布是否和作者的一样。还有那个plane这个文件也得下。
---Original--- From: @.> Date: Wed, Sep 22, 2021 15:43 PM To: @.>; Cc: @.**@.>; Subject: Re: [sshaoshuai/PointRCNN] assert os.path.exists(calib_file) when run eval_rcnn.py (#87)
thanks,I find the problem which is because of the wrong of the file path
老哥,能不能展开说说,啥错误的文件路径?我也遇到了和你一样的错误
— You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.
检查一下数据集的路径是否正确,数据集的排布是否和作者的一样。还有那个plane这个文件也得下。 … ---Original--- From: @.> Date: Wed, Sep 22, 2021 15:43 PM To: @.>; Cc: @.**@.>; Subject: Re: [sshaoshuai/PointRCNN] assert os.path.exists(calib_file) when run eval_rcnn.py (#87) thanks,I find the problem which is because of the wrong of the file path 老哥,能不能展开说说,啥错误的文件路径?我也遇到了和你一样的错误 — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.
我之前运行程序是会跑一段的,在448的时候突然停下来,在看到你的回复后,我仔细检查了两遍路径并下载了plane的文件,可是运行起来还是卡在了448的点,不知道老哥你是否能给些想法和建议: python eval_rcnn.py --cfg_file cfgs/default.yaml --ckpt PointRCNN.pth --batch_size 1 --eval_mode rcnn --set RPN.LOC_XZ_FINE False /usr/local/lib/python3.6/dist-packages/numba/cuda/envvars.py:17: NumbaWarning: Environment variables with the 'NUMBAPRO' prefix are deprecated and consequently ignored, found use of NUMBAPRO_NVVM=/usr/local/cuda-10.0/nvvm/lib64/libnvvm.so.
For more information about alternatives visit: ('https://numba.pydata.org/numba-doc/latest/cuda/overview.html', '#cudatoolkit-lookup') warnings.warn(errors.NumbaWarning(msg)) /usr/local/lib/python3.6/dist-packages/numba/cuda/envvars.py:17: NumbaWarning: Environment variables with the 'NUMBAPRO' prefix are deprecated and consequently ignored, found use of NUMBAPRO_LIBDEVICE=/usr/local/cuda-10.0/nvvm/libdevice/.
For more information about alternatives visit: ('https://numba.pydata.org/numba-doc/latest/cuda/overview.html', '#cudatoolkit-lookup') warnings.warn(errors.NumbaWarning(msg)) /home/wg/PointRCNN/tools/../lib/config.py:187: YAMLLoadWarning: calling yaml.load() without Loader=... is deprecated, as the default Loader is unsafe. Please read https://msg.pyyaml.org/load for full details. yaml_cfg = edict(yaml.load(f))
2021-09-22 04:35:14,630 INFO **Start logging**
2021-09-22 04:35:14,630 INFO cfg_file cfgs/default.yaml
2021-09-22 04:35:14,630 INFO eval_mode rcnn
2021-09-22 04:35:14,630 INFO eval_all False
2021-09-22 04:35:14,630 INFO test False
2021-09-22 04:35:14,630 INFO ckpt PointRCNN.pth
2021-09-22 04:35:14,630 INFO rpn_ckpt None
2021-09-22 04:35:14,630 INFO rcnn_ckpt None
2021-09-22 04:35:14,630 INFO batch_size 1
2021-09-22 04:35:14,630 INFO workers 4
2021-09-22 04:35:14,630 INFO extra_tag default
2021-09-22 04:35:14,630 INFO output_dir None
2021-09-22 04:35:14,630 INFO ckpt_dir None
2021-09-22 04:35:14,630 INFO save_result False
2021-09-22 04:35:14,630 INFO save_rpn_feature False
2021-09-22 04:35:14,630 INFO random_select True
2021-09-22 04:35:14,630 INFO start_epoch 0
2021-09-22 04:35:14,630 INFO rcnn_eval_roi_dir None
2021-09-22 04:35:14,630 INFO rcnn_eval_feature_dir None
2021-09-22 04:35:14,630 INFO set_cfgs ['RPN.LOC_XZ_FINE', 'False']
2021-09-22 04:35:14,630 INFO cfg.TAG: default
2021-09-22 04:35:14,630 INFO cfg.CLASSES: Car
2021-09-22 04:35:14,630 INFO cfg.INCLUDE_SIMILAR_TYPE: True
2021-09-22 04:35:14,630 INFO cfg.AUG_DATA: True
2021-09-22 04:35:14,630 INFO cfg.AUG_METHOD_LIST: ['rotation', 'scaling', 'flip']
2021-09-22 04:35:14,630 INFO cfg.AUG_METHOD_PROB: [1.0, 1.0, 0.5]
2021-09-22 04:35:14,630 INFO cfg.AUG_ROT_RANGE: 18
2021-09-22 04:35:14,630 INFO cfg.GT_AUG_ENABLED: True
2021-09-22 04:35:14,631 INFO cfg.GT_EXTRA_NUM: 15
2021-09-22 04:35:14,631 INFO cfg.GT_AUG_RAND_NUM: True
2021-09-22 04:35:14,631 INFO cfg.GT_AUG_APPLY_PROB: 1.0
2021-09-22 04:35:14,631 INFO cfg.GT_AUG_HARD_RATIO: 0.6
2021-09-22 04:35:14,631 INFO cfg.PC_REDUCE_BY_RANGE: True
2021-09-22 04:35:14,631 INFO cfg.PC_AREA_SCOPE: [[-40. 40. ]
[ -1. 3. ]
[ 0. 70.4]]
2021-09-22 04:35:14,631 INFO cfg.CLS_MEAN_SIZE: [[1.5256319 1.6285675 3.8831165]]
2021-09-22 04:35:14,631 INFO
cfg.RPN = edict()
2021-09-22 04:35:14,631 INFO cfg.RPN.ENABLED: True
2021-09-22 04:35:14,631 INFO cfg.RPN.FIXED: True
2021-09-22 04:35:14,631 INFO cfg.RPN.USE_INTENSITY: False
2021-09-22 04:35:14,631 INFO cfg.RPN.LOC_XZ_FINE: False
2021-09-22 04:35:14,631 INFO cfg.RPN.LOC_SCOPE: 3.0
2021-09-22 04:35:14,631 INFO cfg.RPN.LOC_BIN_SIZE: 0.5
2021-09-22 04:35:14,631 INFO cfg.RPN.NUM_HEAD_BIN: 12
2021-09-22 04:35:14,631 INFO cfg.RPN.BACKBONE: pointnet2_msg
2021-09-22 04:35:14,631 INFO cfg.RPN.USE_BN: True
2021-09-22 04:35:14,631 INFO cfg.RPN.NUM_POINTS: 16384
2021-09-22 04:35:14,631 INFO
cfg.RPN.SA_CONFIG = edict()
2021-09-22 04:35:14,631 INFO cfg.RPN.SA_CONFIG.NPOINTS: [4096, 1024, 256, 64]
2021-09-22 04:35:14,631 INFO cfg.RPN.SA_CONFIG.RADIUS: [[0.1, 0.5], [0.5, 1.0], [1.0, 2.0], [2.0, 4.0]]
2021-09-22 04:35:14,632 INFO cfg.RPN.SA_CONFIG.NSAMPLE: [[16, 32], [16, 32], [16, 32], [16, 32]]
2021-09-22 04:35:14,632 INFO cfg.RPN.SA_CONFIG.MLPS: [[[16, 16, 32], [32, 32, 64]], [[64, 64, 128], [64, 96, 128]], [[128, 196, 256], [128, 196, 256]], [[256, 256, 512], [256, 384, 512]]]
2021-09-22 04:35:14,632 INFO cfg.RPN.FP_MLPS: [[128, 128], [256, 256], [512, 512], [512, 512]]
2021-09-22 04:35:14,632 INFO cfg.RPN.CLS_FC: [128]
2021-09-22 04:35:14,632 INFO cfg.RPN.REG_FC: [128]
2021-09-22 04:35:14,632 INFO cfg.RPN.DP_RATIO: 0.5
2021-09-22 04:35:14,632 INFO cfg.RPN.LOSS_CLS: SigmoidFocalLoss
2021-09-22 04:35:14,632 INFO cfg.RPN.FG_WEIGHT: 15
2021-09-22 04:35:14,632 INFO cfg.RPN.FOCAL_ALPHA: [0.25, 0.75]
2021-09-22 04:35:14,632 INFO cfg.RPN.FOCAL_GAMMA: 2.0
2021-09-22 04:35:14,632 INFO cfg.RPN.REG_LOSS_WEIGHT: [1.0, 1.0, 1.0, 1.0]
2021-09-22 04:35:14,632 INFO cfg.RPN.LOSS_WEIGHT: [1.0, 1.0]
2021-09-22 04:35:14,632 INFO cfg.RPN.NMS_TYPE: normal
2021-09-22 04:35:14,632 INFO cfg.RPN.SCORE_THRESH: 0.3
2021-09-22 04:35:14,632 INFO
cfg.RCNN = edict()
2021-09-22 04:35:14,632 INFO cfg.RCNN.ENABLED: True
2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_RPN_FEATURES: True
2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_MASK: True
2021-09-22 04:35:14,632 INFO cfg.RCNN.MASK_TYPE: seg
2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_INTENSITY: False
2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_DEPTH: True
2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_SEG_SCORE: False
2021-09-22 04:35:14,632 INFO cfg.RCNN.ROI_SAMPLE_JIT: True
2021-09-22 04:35:14,632 INFO cfg.RCNN.ROI_FG_AUG_TIMES: 10
2021-09-22 04:35:14,632 INFO cfg.RCNN.REG_AUG_METHOD: multiple
2021-09-22 04:35:14,632 INFO cfg.RCNN.POOL_EXTRA_WIDTH: 1.0
2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_SCOPE: 1.5
2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_BIN_SIZE: 0.5
2021-09-22 04:35:14,632 INFO cfg.RCNN.NUM_HEAD_BIN: 9
2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_Y_BY_BIN: False
2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_Y_SCOPE: 0.5
2021-09-22 04:35:14,633 INFO cfg.RCNN.LOC_Y_BIN_SIZE: 0.25
2021-09-22 04:35:14,633 INFO cfg.RCNN.SIZE_RES_ON_ROI: False
2021-09-22 04:35:14,633 INFO cfg.RCNN.USE_BN: False
2021-09-22 04:35:14,633 INFO cfg.RCNN.DP_RATIO: 0.0
2021-09-22 04:35:14,633 INFO cfg.RCNN.BACKBONE: pointnet
2021-09-22 04:35:14,633 INFO cfg.RCNN.XYZ_UP_LAYER: [128, 128]
2021-09-22 04:35:14,633 INFO cfg.RCNN.NUM_POINTS: 512
2021-09-22 04:35:14,633 INFO
cfg.RCNN.SA_CONFIG = edict()
2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.NPOINTS: [128, 32, -1]
2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.RADIUS: [0.2, 0.4, 100]
2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.NSAMPLE: [64, 64, 64]
2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.MLPS: [[128, 128, 128], [128, 128, 256], [256, 256, 512]]
2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_FC: [256, 256]
2021-09-22 04:35:14,633 INFO cfg.RCNN.REG_FC: [256, 256]
2021-09-22 04:35:14,633 INFO cfg.RCNN.LOSS_CLS: BinaryCrossEntropy
2021-09-22 04:35:14,633 INFO cfg.RCNN.FOCAL_ALPHA: [0.25, 0.75]
2021-09-22 04:35:14,633 INFO cfg.RCNN.FOCAL_GAMMA: 2.0
2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_WEIGHT: [1. 1. 1.]
2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_FG_THRESH: 0.6
2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_BG_THRESH: 0.45
2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_BG_THRESH_LO: 0.05
2021-09-22 04:35:14,633 INFO cfg.RCNN.REG_FG_THRESH: 0.55
2021-09-22 04:35:14,633 INFO cfg.RCNN.FG_RATIO: 0.5
2021-09-22 04:35:14,633 INFO cfg.RCNN.ROI_PER_IMAGE: 64
2021-09-22 04:35:14,633 INFO cfg.RCNN.HARD_BG_RATIO: 0.8
2021-09-22 04:35:14,633 INFO cfg.RCNN.SCORE_THRESH: 0.3
2021-09-22 04:35:14,634 INFO cfg.RCNN.NMS_THRESH: 0.1
2021-09-22 04:35:14,634 INFO
cfg.TRAIN = edict()
2021-09-22 04:35:14,634 INFO cfg.TRAIN.SPLIT: train
2021-09-22 04:35:14,634 INFO cfg.TRAIN.VAL_SPLIT: smallval
2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR: 0.002
2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR_CLIP: 1e-05
2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR_DECAY: 0.5
2021-09-22 04:35:14,634 INFO cfg.TRAIN.DECAY_STEP_LIST: [100, 150, 180, 200]
2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR_WARMUP: True
2021-09-22 04:35:14,634 INFO cfg.TRAIN.WARMUP_MIN: 0.0002
2021-09-22 04:35:14,634 INFO cfg.TRAIN.WARMUP_EPOCH: 1
2021-09-22 04:35:14,634 INFO cfg.TRAIN.BN_MOMENTUM: 0.1
2021-09-22 04:35:14,634 INFO cfg.TRAIN.BN_DECAY: 0.5
2021-09-22 04:35:14,634 INFO cfg.TRAIN.BNM_CLIP: 0.01
2021-09-22 04:35:14,634 INFO cfg.TRAIN.BN_DECAY_STEP_LIST: [1000]
2021-09-22 04:35:14,634 INFO cfg.TRAIN.OPTIMIZER: adam_onecycle
2021-09-22 04:35:14,634 INFO cfg.TRAIN.WEIGHT_DECAY: 0.001
2021-09-22 04:35:14,634 INFO cfg.TRAIN.MOMENTUM: 0.9
2021-09-22 04:35:14,634 INFO cfg.TRAIN.MOMS: [0.95, 0.85]
2021-09-22 04:35:14,634 INFO cfg.TRAIN.DIV_FACTOR: 10.0
2021-09-22 04:35:14,634 INFO cfg.TRAIN.PCT_START: 0.4
2021-09-22 04:35:14,634 INFO cfg.TRAIN.GRAD_NORM_CLIP: 1.0
2021-09-22 04:35:14,634 INFO cfg.TRAIN.RPN_PRE_NMS_TOP_N: 9000
2021-09-22 04:35:14,634 INFO cfg.TRAIN.RPN_POST_NMS_TOP_N: 512
2021-09-22 04:35:14,634 INFO cfg.TRAIN.RPN_NMS_THRESH: 0.85
2021-09-22 04:35:14,635 INFO cfg.TRAIN.RPN_DISTANCE_BASED_PROPOSE: True
2021-09-22 04:35:14,635 INFO
cfg.TEST = edict()
2021-09-22 04:35:14,635 INFO cfg.TEST.SPLIT: val
2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_PRE_NMS_TOP_N: 9000
2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_POST_NMS_TOP_N: 100
2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_NMS_THRESH: 0.8
2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_DISTANCE_BASED_PROPOSE: True
2021-09-22 04:35:14,636 INFO Load testing samples from ../data/KITTI/object/training
2021-09-22 04:35:14,636 INFO Done: total test samples 3769
2021-09-22 04:35:16,962 INFO ==> Loading from checkpoint 'PointRCNN.pth'
2021-09-22 04:35:16,989 INFO ==> Done
2021-09-22 04:35:16,990 INFO ---- EPOCH no_number JOINT EVALUATION ----
2021-09-22 04:35:16,990 INFO ==> Output file: ../output/rcnn/default/eval/epoch_no_number/val
eval: 13%|▋ | 488/3769 [00:55<05:59, 9.13it/s, mode=EVAL, recall=1170/1769]Traceback (most recent call last):
File "eval_rcnn.py", line 902, in
eval: 13%|▋ | 488/3769 [00:55<06:12, 8.80it/s, mode=EVAL, recall=1170/1769]
我就说下我运行的流程吧。首先按照作者的要求,将需要的环境安装。之后在pointnet2文件夹里py setup.py。回到主文件夹下,sh 那个.sh文件。将数据集全部下载并且按照要求弄好位置。运行作者Quick demo下的代码,成功。之后按照作者的步骤,运行train rpn阶段。运行train rcnn阶段时,我发现,用a的方法即线上方法训练好了之后,运行inference的第一个没问题。但是如果用b的方法即离线方法训练之后,运行inference就有问题。总体就是这个了。我之前遇到这个问题时,主要是planes那个文件没有下载导致的。根据报错的结果,是那个文件不存在。如果你试了之后还是有问题,可以尝试一级一级找进去,把那个地址输出试试
---Original--- From: @.> Date: Wed, Sep 22, 2021 16:41 PM To: @.>; Cc: @.**@.>; Subject: Re: [sshaoshuai/PointRCNN] assert os.path.exists(calib_file) when run eval_rcnn.py (#87)
检查一下数据集的路径是否正确,数据集的排布是否和作者的一样。还有那个plane这个文件也得下。 … ---Original--- From: @.> Date: Wed, Sep 22, 2021 15:43 PM To: @.>; Cc: @.@.>; Subject: Re: [sshaoshuai/PointRCNN] assert os.path.exists(calib_file) when run eval_rcnn.py (#87) thanks,I find the problem which is because of the wrong of the file path 老哥,能不能展开说说,啥错误的文件路径?我也遇到了和你一样的错误 — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.
我之前运行程序是会跑一段的,在448的时候突然停下来,在看到你的回复后,我仔细检查了两遍路径并下载了plane的文件,可是运行起来还是卡在了448的点,不知道老哥你是否能给些想法和建议: python eval_rcnn.py --cfg_file cfgs/default.yaml --ckpt PointRCNN.pth --batch_size 1 --eval_mode rcnn --set RPN.LOC_XZ_FINE False /usr/local/lib/python3.6/dist-packages/numba/cuda/envvars.py:17: NumbaWarning: Environment variables with the 'NUMBAPRO' prefix are deprecated and consequently ignored, found use of NUMBAPRO_NVVM=/usr/local/cuda-10.0/nvvm/lib64/libnvvm.so.
For more information about alternatives visit: ('https://numba.pydata.org/numba-doc/latest/cuda/overview.html', '#cudatoolkit-lookup') warnings.warn(errors.NumbaWarning(msg)) /usr/local/lib/python3.6/dist-packages/numba/cuda/envvars.py:17: NumbaWarning: Environment variables with the 'NUMBAPRO' prefix are deprecated and consequently ignored, found use of NUMBAPRO_LIBDEVICE=/usr/local/cuda-10.0/nvvm/libdevice/.
For more information about alternatives visit: ('https://numba.pydata.org/numba-doc/latest/cuda/overview.html', '#cudatoolkit-lookup') warnings.warn(errors.NumbaWarning(msg)) /home/wg/PointRCNN/tools/../lib/config.py:187: YAMLLoadWarning: calling yaml.load() without Loader=... is deprecated, as the default Loader is unsafe. Please read https://msg.pyyaml.org/load for full details. yaml_cfg = edict(yaml.load(f))
2021-09-22 04:35:14,630 INFO Start logging 2021-09-22 04:35:14,630 INFO cfg_file cfgs/default.yaml 2021-09-22 04:35:14,630 INFO eval_mode rcnn 2021-09-22 04:35:14,630 INFO eval_all False 2021-09-22 04:35:14,630 INFO test False 2021-09-22 04:35:14,630 INFO ckpt PointRCNN.pth 2021-09-22 04:35:14,630 INFO rpn_ckpt None 2021-09-22 04:35:14,630 INFO rcnn_ckpt None 2021-09-22 04:35:14,630 INFO batch_size 1 2021-09-22 04:35:14,630 INFO workers 4 2021-09-22 04:35:14,630 INFO extra_tag default 2021-09-22 04:35:14,630 INFO output_dir None 2021-09-22 04:35:14,630 INFO ckpt_dir None 2021-09-22 04:35:14,630 INFO save_result False 2021-09-22 04:35:14,630 INFO save_rpn_feature False 2021-09-22 04:35:14,630 INFO random_select True 2021-09-22 04:35:14,630 INFO start_epoch 0 2021-09-22 04:35:14,630 INFO rcnn_eval_roi_dir None 2021-09-22 04:35:14,630 INFO rcnn_eval_feature_dir None 2021-09-22 04:35:14,630 INFO set_cfgs ['RPN.LOC_XZ_FINE', 'False'] 2021-09-22 04:35:14,630 INFO cfg.TAG: default 2021-09-22 04:35:14,630 INFO cfg.CLASSES: Car 2021-09-22 04:35:14,630 INFO cfg.INCLUDE_SIMILAR_TYPE: True 2021-09-22 04:35:14,630 INFO cfg.AUG_DATA: True 2021-09-22 04:35:14,630 INFO cfg.AUG_METHOD_LIST: ['rotation', 'scaling', 'flip'] 2021-09-22 04:35:14,630 INFO cfg.AUG_METHOD_PROB: [1.0, 1.0, 0.5] 2021-09-22 04:35:14,630 INFO cfg.AUG_ROT_RANGE: 18 2021-09-22 04:35:14,630 INFO cfg.GT_AUG_ENABLED: True 2021-09-22 04:35:14,631 INFO cfg.GT_EXTRA_NUM: 15 2021-09-22 04:35:14,631 INFO cfg.GT_AUG_RAND_NUM: True 2021-09-22 04:35:14,631 INFO cfg.GT_AUG_APPLY_PROB: 1.0 2021-09-22 04:35:14,631 INFO cfg.GT_AUG_HARD_RATIO: 0.6 2021-09-22 04:35:14,631 INFO cfg.PC_REDUCE_BY_RANGE: True 2021-09-22 04:35:14,631 INFO cfg.PC_AREA_SCOPE: [[-40. 40. ] [ -1. 3. ] [ 0. 70.4]] 2021-09-22 04:35:14,631 INFO cfg.CLS_MEAN_SIZE: [[1.5256319 1.6285675 3.8831165]] 2021-09-22 04:35:14,631 INFO cfg.RPN = edict() 2021-09-22 04:35:14,631 INFO cfg.RPN.ENABLED: True 2021-09-22 04:35:14,631 INFO cfg.RPN.FIXED: True 2021-09-22 04:35:14,631 INFO cfg.RPN.USE_INTENSITY: False 2021-09-22 04:35:14,631 INFO cfg.RPN.LOC_XZ_FINE: False 2021-09-22 04:35:14,631 INFO cfg.RPN.LOC_SCOPE: 3.0 2021-09-22 04:35:14,631 INFO cfg.RPN.LOC_BIN_SIZE: 0.5 2021-09-22 04:35:14,631 INFO cfg.RPN.NUM_HEAD_BIN: 12 2021-09-22 04:35:14,631 INFO cfg.RPN.BACKBONE: pointnet2_msg 2021-09-22 04:35:14,631 INFO cfg.RPN.USE_BN: True 2021-09-22 04:35:14,631 INFO cfg.RPN.NUM_POINTS: 16384 2021-09-22 04:35:14,631 INFO cfg.RPN.SA_CONFIG = edict() 2021-09-22 04:35:14,631 INFO cfg.RPN.SA_CONFIG.NPOINTS: [4096, 1024, 256, 64] 2021-09-22 04:35:14,631 INFO cfg.RPN.SA_CONFIG.RADIUS: [[0.1, 0.5], [0.5, 1.0], [1.0, 2.0], [2.0, 4.0]] 2021-09-22 04:35:14,632 INFO cfg.RPN.SA_CONFIG.NSAMPLE: [[16, 32], [16, 32], [16, 32], [16, 32]] 2021-09-22 04:35:14,632 INFO cfg.RPN.SA_CONFIG.MLPS: [[[16, 16, 32], [32, 32, 64]], [[64, 64, 128], [64, 96, 128]], [[128, 196, 256], [128, 196, 256]], [[256, 256, 512], [256, 384, 512]]] 2021-09-22 04:35:14,632 INFO cfg.RPN.FP_MLPS: [[128, 128], [256, 256], [512, 512], [512, 512]] 2021-09-22 04:35:14,632 INFO cfg.RPN.CLS_FC: [128] 2021-09-22 04:35:14,632 INFO cfg.RPN.REG_FC: [128] 2021-09-22 04:35:14,632 INFO cfg.RPN.DP_RATIO: 0.5 2021-09-22 04:35:14,632 INFO cfg.RPN.LOSS_CLS: SigmoidFocalLoss 2021-09-22 04:35:14,632 INFO cfg.RPN.FG_WEIGHT: 15 2021-09-22 04:35:14,632 INFO cfg.RPN.FOCAL_ALPHA: [0.25, 0.75] 2021-09-22 04:35:14,632 INFO cfg.RPN.FOCAL_GAMMA: 2.0 2021-09-22 04:35:14,632 INFO cfg.RPN.REG_LOSS_WEIGHT: [1.0, 1.0, 1.0, 1.0] 2021-09-22 04:35:14,632 INFO cfg.RPN.LOSS_WEIGHT: [1.0, 1.0] 2021-09-22 04:35:14,632 INFO cfg.RPN.NMS_TYPE: normal 2021-09-22 04:35:14,632 INFO cfg.RPN.SCORE_THRESH: 0.3 2021-09-22 04:35:14,632 INFO cfg.RCNN = edict() 2021-09-22 04:35:14,632 INFO cfg.RCNN.ENABLED: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_RPN_FEATURES: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_MASK: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.MASK_TYPE: seg 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_INTENSITY: False 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_DEPTH: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_SEG_SCORE: False 2021-09-22 04:35:14,632 INFO cfg.RCNN.ROI_SAMPLE_JIT: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.ROI_FG_AUG_TIMES: 10 2021-09-22 04:35:14,632 INFO cfg.RCNN.REG_AUG_METHOD: multiple 2021-09-22 04:35:14,632 INFO cfg.RCNN.POOL_EXTRA_WIDTH: 1.0 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_SCOPE: 1.5 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_BIN_SIZE: 0.5 2021-09-22 04:35:14,632 INFO cfg.RCNN.NUM_HEAD_BIN: 9 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_Y_BY_BIN: False 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_Y_SCOPE: 0.5 2021-09-22 04:35:14,633 INFO cfg.RCNN.LOC_Y_BIN_SIZE: 0.25 2021-09-22 04:35:14,633 INFO cfg.RCNN.SIZE_RES_ON_ROI: False 2021-09-22 04:35:14,633 INFO cfg.RCNN.USE_BN: False 2021-09-22 04:35:14,633 INFO cfg.RCNN.DP_RATIO: 0.0 2021-09-22 04:35:14,633 INFO cfg.RCNN.BACKBONE: pointnet 2021-09-22 04:35:14,633 INFO cfg.RCNN.XYZ_UP_LAYER: [128, 128] 2021-09-22 04:35:14,633 INFO cfg.RCNN.NUM_POINTS: 512 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG = edict() 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.NPOINTS: [128, 32, -1] 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.RADIUS: [0.2, 0.4, 100] 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.NSAMPLE: [64, 64, 64] 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.MLPS: [[128, 128, 128], [128, 128, 256], [256, 256, 512]] 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_FC: [256, 256] 2021-09-22 04:35:14,633 INFO cfg.RCNN.REG_FC: [256, 256] 2021-09-22 04:35:14,633 INFO cfg.RCNN.LOSS_CLS: BinaryCrossEntropy 2021-09-22 04:35:14,633 INFO cfg.RCNN.FOCAL_ALPHA: [0.25, 0.75] 2021-09-22 04:35:14,633 INFO cfg.RCNN.FOCAL_GAMMA: 2.0 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_WEIGHT: [1. 1. 1.] 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_FG_THRESH: 0.6 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_BG_THRESH: 0.45 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_BG_THRESH_LO: 0.05 2021-09-22 04:35:14,633 INFO cfg.RCNN.REG_FG_THRESH: 0.55 2021-09-22 04:35:14,633 INFO cfg.RCNN.FG_RATIO: 0.5 2021-09-22 04:35:14,633 INFO cfg.RCNN.ROI_PER_IMAGE: 64 2021-09-22 04:35:14,633 INFO cfg.RCNN.HARD_BG_RATIO: 0.8 2021-09-22 04:35:14,633 INFO cfg.RCNN.SCORE_THRESH: 0.3 2021-09-22 04:35:14,634 INFO cfg.RCNN.NMS_THRESH: 0.1 2021-09-22 04:35:14,634 INFO cfg.TRAIN = edict() 2021-09-22 04:35:14,634 INFO cfg.TRAIN.SPLIT: train 2021-09-22 04:35:14,634 INFO cfg.TRAIN.VAL_SPLIT: smallval 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR: 0.002 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR_CLIP: 1e-05 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR_DECAY: 0.5 2021-09-22 04:35:14,634 INFO cfg.TRAIN.DECAY_STEP_LIST: [100, 150, 180, 200] 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR_WARMUP: True 2021-09-22 04:35:14,634 INFO cfg.TRAIN.WARMUP_MIN: 0.0002 2021-09-22 04:35:14,634 INFO cfg.TRAIN.WARMUP_EPOCH: 1 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BN_MOMENTUM: 0.1 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BN_DECAY: 0.5 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BNM_CLIP: 0.01 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BN_DECAY_STEP_LIST: [1000] 2021-09-22 04:35:14,634 INFO cfg.TRAIN.OPTIMIZER: adam_onecycle 2021-09-22 04:35:14,634 INFO cfg.TRAIN.WEIGHT_DECAY: 0.001 2021-09-22 04:35:14,634 INFO cfg.TRAIN.MOMENTUM: 0.9 2021-09-22 04:35:14,634 INFO cfg.TRAIN.MOMS: [0.95, 0.85] 2021-09-22 04:35:14,634 INFO cfg.TRAIN.DIV_FACTOR: 10.0 2021-09-22 04:35:14,634 INFO cfg.TRAIN.PCT_START: 0.4 2021-09-22 04:35:14,634 INFO cfg.TRAIN.GRAD_NORM_CLIP: 1.0 2021-09-22 04:35:14,634 INFO cfg.TRAIN.RPN_PRE_NMS_TOP_N: 9000 2021-09-22 04:35:14,634 INFO cfg.TRAIN.RPN_POST_NMS_TOP_N: 512 2021-09-22 04:35:14,634 INFO cfg.TRAIN.RPN_NMS_THRESH: 0.85 2021-09-22 04:35:14,635 INFO cfg.TRAIN.RPN_DISTANCE_BASED_PROPOSE: True 2021-09-22 04:35:14,635 INFO cfg.TEST = edict() 2021-09-22 04:35:14,635 INFO cfg.TEST.SPLIT: val 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_PRE_NMS_TOP_N: 9000 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_POST_NMS_TOP_N: 100 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_NMS_THRESH: 0.8 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_DISTANCE_BASED_PROPOSE: True 2021-09-22 04:35:14,636 INFO Load testing samples from ../data/KITTI/object/training 2021-09-22 04:35:14,636 INFO Done: total test samples 3769 2021-09-22 04:35:16,962 INFO ==> Loading from checkpoint 'PointRCNN.pth' 2021-09-22 04:35:16,989 INFO ==> Done 2021-09-22 04:35:16,990 INFO ---- EPOCH no_number JOINT EVALUATION ---- 2021-09-22 04:35:16,990 INFO ==> Output file: ../output/rcnn/default/eval/epoch_no_number/val eval: 13%|▋ | 488/3769 [00:55<05:59, 9.13it/s, mode=EVAL, recall=1170/1769]Traceback (most recent call last): File "eval_rcnn.py", line 902, in eval_single_ckpt(root_result_dir) File "eval_rcnn.py", line 765, in eval_single_ckpt eval_one_epoch(model, test_loader, epoch_id, root_result_dir, logger) File "eval_rcnn.py", line 692, in eval_one_epoch ret_dict = eval_one_epoch_joint(model, dataloader, epoch_id, result_dir, logger) File "eval_rcnn.py", line 486, in eval_one_epoch_joint for data in dataloader: File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 582, in next return self._process_next_batch(batch) File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 608, in _process_next_batch raise batch.exc_type(batch.exc_msg) AssertionError: Traceback (most recent call last): File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/_utils/worker.py", line 99, in _worker_loop samples = collate_fn([dataset[i] for i in batch_indices]) File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/_utils/worker.py", line 99, in samples = collate_fn([dataset[i] for i in batch_indices]) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_rcnn_dataset.py", line 234, in getitem return self.get_rpn_sample(index) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_rcnn_dataset.py", line 249, in get_rpn_sample calib = self.get_calib(sample_id) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_rcnn_dataset.py", line 133, in get_calib return super().get_calib(idx % 10000) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_dataset.py", line 47, in get_calib assert os.path.exists(calib_file) AssertionError
eval: 13%|▋ | 488/3769 [00:55<06:12, 8.80it/s, mode=EVAL, recall=1170/1769]
— You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.
我就说下我运行的流程吧。首先按照作者的要求,将需要的环境安装。之后在pointnet2文件夹里py setup.py。回到主文件夹下,sh 那个.sh文件。将数据集全部下载并且按照要求弄好位置。运行作者Quick demo下的代码,成功。之后按照作者的步骤,运行train rpn阶段。运行train rcnn阶段时,我发现,用a的方法即线上方法训练好了之后,运行inference的第一个没问题。但是如果用b的方法即离线方法训练之后,运行inference就有问题。总体就是这个了。我之前遇到这个问题时,主要是planes那个文件没有下载导致的。根据报错的结果,是那个文件不存在。如果你试了之后还是有问题,可以尝试一级一级找进去,把那个地址输出试试 … ---Original--- From: @.> Date: Wed, Sep 22, 2021 16:41 PM To: @.>; Cc: @.**@.>; Subject: Re: [sshaoshuai/PointRCNN] assert os.path.exists(calib_file) when run eval_rcnn.py (#87) 检查一下数据集的路径是否正确,数据集的排布是否和作者的一样。还有那个plane这个文件也得下。 … ---Original--- From: @.> Date: Wed, Sep 22, 2021 15:43 PM To: @.>; Cc: @.@.>; Subject: Re: [sshaoshuai/PointRCNN] assert os.path.exists(calib_file) when run eval_rcnn.py (#87) thanks,I find the problem which is because of the wrong of the file path 老哥,能不能展开说说,啥错误的文件路径?我也遇到了和你一样的错误 — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android. 我之前运行程序是会跑一段的,在448的时候突然停下来,在看到你的回复后,我仔细检查了两遍路径并下载了plane的文件,可是运行起来还是卡在了448的点,不知道老哥你是否能给些想法和建议: python eval_rcnn.py --cfg_file cfgs/default.yaml --ckpt PointRCNN.pth --batch_size 1 --eval_mode rcnn --set RPN.LOC_XZ_FINE False /usr/local/lib/python3.6/dist-packages/numba/cuda/envvars.py:17: NumbaWarning: Environment variables with the 'NUMBAPRO' prefix are deprecated and consequently ignored, found use of NUMBAPRO_NVVM=/usr/local/cuda-10.0/nvvm/lib64/libnvvm.so. For more information about alternatives visit: ('https://numba.pydata.org/numba-doc/latest/cuda/overview.html', '#cudatoolkit-lookup') warnings.warn(errors.NumbaWarning(msg)) /usr/local/lib/python3.6/dist-packages/numba/cuda/envvars.py:17: NumbaWarning: Environment variables with the 'NUMBAPRO' prefix are deprecated and consequently ignored, found use of NUMBAPRO_LIBDEVICE=/usr/local/cuda-10.0/nvvm/libdevice/. For more information about alternatives visit: ('https://numba.pydata.org/numba-doc/latest/cuda/overview.html', '#cudatoolkit-lookup') warnings.warn(errors.NumbaWarning(msg)) /home/wg/PointRCNN/tools/../lib/config.py:187: YAMLLoadWarning: calling yaml.load() without Loader=... is deprecated, as the default Loader is unsafe. Please read https://msg.pyyaml.org/load for full details. yaml_cfg = edict(yaml.load(f)) 2021-09-22 04:35:14,630 INFO Start logging 2021-09-22 04:35:14,630 INFO cfg_file cfgs/default.yaml 2021-09-22 04:35:14,630 INFO eval_mode rcnn 2021-09-22 04:35:14,630 INFO eval_all False 2021-09-22 04:35:14,630 INFO test False 2021-09-22 04:35:14,630 INFO ckpt PointRCNN.pth 2021-09-22 04:35:14,630 INFO rpn_ckpt None 2021-09-22 04:35:14,630 INFO rcnn_ckpt None 2021-09-22 04:35:14,630 INFO batch_size 1 2021-09-22 04:35:14,630 INFO workers 4 2021-09-22 04:35:14,630 INFO extra_tag default 2021-09-22 04:35:14,630 INFO output_dir None 2021-09-22 04:35:14,630 INFO ckpt_dir None 2021-09-22 04:35:14,630 INFO save_result False 2021-09-22 04:35:14,630 INFO save_rpn_feature False 2021-09-22 04:35:14,630 INFO random_select True 2021-09-22 04:35:14,630 INFO start_epoch 0 2021-09-22 04:35:14,630 INFO rcnn_eval_roi_dir None 2021-09-22 04:35:14,630 INFO rcnn_eval_feature_dir None 2021-09-22 04:35:14,630 INFO set_cfgs ['RPN.LOC_XZ_FINE', 'False'] 2021-09-22 04:35:14,630 INFO cfg.TAG: default 2021-09-22 04:35:14,630 INFO cfg.CLASSES: Car 2021-09-22 04:35:14,630 INFO cfg.INCLUDE_SIMILAR_TYPE: True 2021-09-22 04:35:14,630 INFO cfg.AUG_DATA: True 2021-09-22 04:35:14,630 INFO cfg.AUG_METHOD_LIST: ['rotation', 'scaling', 'flip'] 2021-09-22 04:35:14,630 INFO cfg.AUG_METHOD_PROB: [1.0, 1.0, 0.5] 2021-09-22 04:35:14,630 INFO cfg.AUG_ROT_RANGE: 18 2021-09-22 04:35:14,630 INFO cfg.GT_AUG_ENABLED: True 2021-09-22 04:35:14,631 INFO cfg.GT_EXTRA_NUM: 15 2021-09-22 04:35:14,631 INFO cfg.GT_AUG_RAND_NUM: True 2021-09-22 04:35:14,631 INFO cfg.GT_AUG_APPLY_PROB: 1.0 2021-09-22 04:35:14,631 INFO cfg.GT_AUG_HARD_RATIO: 0.6 2021-09-22 04:35:14,631 INFO cfg.PC_REDUCE_BY_RANGE: True 2021-09-22 04:35:14,631 INFO cfg.PC_AREA_SCOPE: [[-40. 40. ] [ -1. 3. ] [ 0. 70.4]] 2021-09-22 04:35:14,631 INFO cfg.CLS_MEAN_SIZE: [[1.5256319 1.6285675 3.8831165]] 2021-09-22 04:35:14,631 INFO cfg.RPN = edict() 2021-09-22 04:35:14,631 INFO cfg.RPN.ENABLED: True 2021-09-22 04:35:14,631 INFO cfg.RPN.FIXED: True 2021-09-22 04:35:14,631 INFO cfg.RPN.USE_INTENSITY: False 2021-09-22 04:35:14,631 INFO cfg.RPN.LOC_XZ_FINE: False 2021-09-22 04:35:14,631 INFO cfg.RPN.LOC_SCOPE: 3.0 2021-09-22 04:35:14,631 INFO cfg.RPN.LOC_BIN_SIZE: 0.5 2021-09-22 04:35:14,631 INFO cfg.RPN.NUM_HEAD_BIN: 12 2021-09-22 04:35:14,631 INFO cfg.RPN.BACKBONE: pointnet2_msg 2021-09-22 04:35:14,631 INFO cfg.RPN.USE_BN: True 2021-09-22 04:35:14,631 INFO cfg.RPN.NUM_POINTS: 16384 2021-09-22 04:35:14,631 INFO cfg.RPN.SA_CONFIG = edict() 2021-09-22 04:35:14,631 INFO cfg.RPN.SA_CONFIG.NPOINTS: [4096, 1024, 256, 64] 2021-09-22 04:35:14,631 INFO cfg.RPN.SA_CONFIG.RADIUS: [[0.1, 0.5], [0.5, 1.0], [1.0, 2.0], [2.0, 4.0]] 2021-09-22 04:35:14,632 INFO cfg.RPN.SA_CONFIG.NSAMPLE: [[16, 32], [16, 32], [16, 32], [16, 32]] 2021-09-22 04:35:14,632 INFO cfg.RPN.SA_CONFIG.MLPS: [[[16, 16, 32], [32, 32, 64]], [[64, 64, 128], [64, 96, 128]], [[128, 196, 256], [128, 196, 256]], [[256, 256, 512], [256, 384, 512]]] 2021-09-22 04:35:14,632 INFO cfg.RPN.FP_MLPS: [[128, 128], [256, 256], [512, 512], [512, 512]] 2021-09-22 04:35:14,632 INFO cfg.RPN.CLS_FC: [128] 2021-09-22 04:35:14,632 INFO cfg.RPN.REG_FC: [128] 2021-09-22 04:35:14,632 INFO cfg.RPN.DP_RATIO: 0.5 2021-09-22 04:35:14,632 INFO cfg.RPN.LOSS_CLS: SigmoidFocalLoss 2021-09-22 04:35:14,632 INFO cfg.RPN.FG_WEIGHT: 15 2021-09-22 04:35:14,632 INFO cfg.RPN.FOCAL_ALPHA: [0.25, 0.75] 2021-09-22 04:35:14,632 INFO cfg.RPN.FOCAL_GAMMA: 2.0 2021-09-22 04:35:14,632 INFO cfg.RPN.REG_LOSS_WEIGHT: [1.0, 1.0, 1.0, 1.0] 2021-09-22 04:35:14,632 INFO cfg.RPN.LOSS_WEIGHT: [1.0, 1.0] 2021-09-22 04:35:14,632 INFO cfg.RPN.NMS_TYPE: normal 2021-09-22 04:35:14,632 INFO cfg.RPN.SCORE_THRESH: 0.3 2021-09-22 04:35:14,632 INFO cfg.RCNN = edict() 2021-09-22 04:35:14,632 INFO cfg.RCNN.ENABLED: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_RPN_FEATURES: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_MASK: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.MASK_TYPE: seg 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_INTENSITY: False 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_DEPTH: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_SEG_SCORE: False 2021-09-22 04:35:14,632 INFO cfg.RCNN.ROI_SAMPLE_JIT: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.ROI_FG_AUG_TIMES: 10 2021-09-22 04:35:14,632 INFO cfg.RCNN.REG_AUG_METHOD: multiple 2021-09-22 04:35:14,632 INFO cfg.RCNN.POOL_EXTRA_WIDTH: 1.0 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_SCOPE: 1.5 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_BIN_SIZE: 0.5 2021-09-22 04:35:14,632 INFO cfg.RCNN.NUM_HEAD_BIN: 9 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_Y_BY_BIN: False 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_Y_SCOPE: 0.5 2021-09-22 04:35:14,633 INFO cfg.RCNN.LOC_Y_BIN_SIZE: 0.25 2021-09-22 04:35:14,633 INFO cfg.RCNN.SIZE_RES_ON_ROI: False 2021-09-22 04:35:14,633 INFO cfg.RCNN.USE_BN: False 2021-09-22 04:35:14,633 INFO cfg.RCNN.DP_RATIO: 0.0 2021-09-22 04:35:14,633 INFO cfg.RCNN.BACKBONE: pointnet 2021-09-22 04:35:14,633 INFO cfg.RCNN.XYZ_UP_LAYER: [128, 128] 2021-09-22 04:35:14,633 INFO cfg.RCNN.NUM_POINTS: 512 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG = edict() 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.NPOINTS: [128, 32, -1] 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.RADIUS: [0.2, 0.4, 100] 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.NSAMPLE: [64, 64, 64] 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.MLPS: [[128, 128, 128], [128, 128, 256], [256, 256, 512]] 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_FC: [256, 256] 2021-09-22 04:35:14,633 INFO cfg.RCNN.REG_FC: [256, 256] 2021-09-22 04:35:14,633 INFO cfg.RCNN.LOSS_CLS: BinaryCrossEntropy 2021-09-22 04:35:14,633 INFO cfg.RCNN.FOCAL_ALPHA: [0.25, 0.75] 2021-09-22 04:35:14,633 INFO cfg.RCNN.FOCAL_GAMMA: 2.0 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_WEIGHT: [1. 1. 1.] 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_FG_THRESH: 0.6 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_BG_THRESH: 0.45 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_BG_THRESH_LO: 0.05 2021-09-22 04:35:14,633 INFO cfg.RCNN.REG_FG_THRESH: 0.55 2021-09-22 04:35:14,633 INFO cfg.RCNN.FG_RATIO: 0.5 2021-09-22 04:35:14,633 INFO cfg.RCNN.ROI_PER_IMAGE: 64 2021-09-22 04:35:14,633 INFO cfg.RCNN.HARD_BG_RATIO: 0.8 2021-09-22 04:35:14,633 INFO cfg.RCNN.SCORE_THRESH: 0.3 2021-09-22 04:35:14,634 INFO cfg.RCNN.NMS_THRESH: 0.1 2021-09-22 04:35:14,634 INFO cfg.TRAIN = edict() 2021-09-22 04:35:14,634 INFO cfg.TRAIN.SPLIT: train 2021-09-22 04:35:14,634 INFO cfg.TRAIN.VAL_SPLIT: smallval 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR: 0.002 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR_CLIP: 1e-05 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR_DECAY: 0.5 2021-09-22 04:35:14,634 INFO cfg.TRAIN.DECAY_STEP_LIST: [100, 150, 180, 200] 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR_WARMUP: True 2021-09-22 04:35:14,634 INFO cfg.TRAIN.WARMUP_MIN: 0.0002 2021-09-22 04:35:14,634 INFO cfg.TRAIN.WARMUP_EPOCH: 1 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BN_MOMENTUM: 0.1 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BN_DECAY: 0.5 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BNM_CLIP: 0.01 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BN_DECAY_STEP_LIST: [1000] 2021-09-22 04:35:14,634 INFO cfg.TRAIN.OPTIMIZER: adam_onecycle 2021-09-22 04:35:14,634 INFO cfg.TRAIN.WEIGHT_DECAY: 0.001 2021-09-22 04:35:14,634 INFO cfg.TRAIN.MOMENTUM: 0.9 2021-09-22 04:35:14,634 INFO cfg.TRAIN.MOMS: [0.95, 0.85] 2021-09-22 04:35:14,634 INFO cfg.TRAIN.DIV_FACTOR: 10.0 2021-09-22 04:35:14,634 INFO cfg.TRAIN.PCT_START: 0.4 2021-09-22 04:35:14,634 INFO cfg.TRAIN.GRAD_NORM_CLIP: 1.0 2021-09-22 04:35:14,634 INFO cfg.TRAIN.RPN_PRE_NMS_TOP_N: 9000 2021-09-22 04:35:14,634 INFO cfg.TRAIN.RPN_POST_NMS_TOP_N: 512 2021-09-22 04:35:14,634 INFO cfg.TRAIN.RPN_NMS_THRESH: 0.85 2021-09-22 04:35:14,635 INFO cfg.TRAIN.RPN_DISTANCE_BASED_PROPOSE: True 2021-09-22 04:35:14,635 INFO cfg.TEST = edict() 2021-09-22 04:35:14,635 INFO cfg.TEST.SPLIT: val 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_PRE_NMS_TOP_N: 9000 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_POST_NMS_TOP_N: 100 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_NMS_THRESH: 0.8 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_DISTANCE_BASED_PROPOSE: True 2021-09-22 04:35:14,636 INFO Load testing samples from ../data/KITTI/object/training 2021-09-22 04:35:14,636 INFO Done: total test samples 3769 2021-09-22 04:35:16,962 INFO ==> Loading from checkpoint 'PointRCNN.pth' 2021-09-22 04:35:16,989 INFO ==> Done 2021-09-22 04:35:16,990 INFO ---- EPOCH no_number JOINT EVALUATION ---- 2021-09-22 04:35:16,990 INFO ==> Output file: ../output/rcnn/default/eval/epoch_no_number/val eval: 13%|▋ | 488/3769 [00:55<05:59, 9.13it/s, mode=EVAL, recall=1170/1769]Traceback (most recent call last): File "eval_rcnn.py", line 902, in eval_single_ckpt(root_result_dir) File "eval_rcnn.py", line 765, in eval_single_ckpt eval_one_epoch(model, test_loader, epoch_id, root_result_dir, logger) File "eval_rcnn.py", line 692, in eval_one_epoch ret_dict = eval_one_epoch_joint(model, dataloader, epoch_id, result_dir, logger) File "eval_rcnn.py", line 486, in eval_one_epoch_joint for data in dataloader: File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 582, in next return self._process_next_batch(batch) File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 608, in _process_next_batch raise batch.exc_type(batch.exc_msg) AssertionError: Traceback (most recent call last): File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/_utils/worker.py", line 99, in _worker_loop samples = collate_fn([dataset[i] for i in batch_indices]) File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/_utils/worker.py", line 99, in samples = collate_fn([dataset[i] for i in batch_indices]) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_rcnn_dataset.py", line 234, in getitem return self.get_rpn_sample(index) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_rcnn_dataset.py", line 249, in get_rpn_sample calib = self.get_calib(sample_id) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_rcnn_dataset.py", line 133, in get_calib return super().get_calib(idx % 10000) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_dataset.py", line 47, in get_calib assert os.path.exists(calib_file) AssertionError eval: 13%|▋ | 488/3769 [00:55<06:12, 8.80it/s, mode=EVAL, recall=1170/1769] — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.
好的,谢谢!
I had the same problem. But I found my datasets not strictly following this structure
PointRCNN ├── data │ ├── KITTI │ │ ├── ImageSets │ │ ├── object │ │ │ ├──training │ │ │ ├──calib & velodyne & label_2 & image_2 & (optional: planes) │ │ │ ├──testing │ │ │ ├──calib & velodyne & image_2
and I also found that files of my datasets were incomplete. I hope this is helpful.
Hello, do you mean that I must download the whole dataset(about 38GB) within it? I have no more free space so I cut 5/6 from it. And the process stoped when 13% every time...
检查一下数据集的路径是否正确,数据集的排布是否和作者的一样。还有那个plane这个文件也得下。 … ---Original--- From: @.**> Date: Wed, Sep 22, 2021 15:43 PM To: @.**>; Cc: @.**@.**>; Subject: Re: [sshaoshuai/PointRCNN] assert os.path.exists(calib_file) when run eval_rcnn.py (#87) thanks,I find the problem which is because of the wrong of the file path 老哥,能不能展开说说,啥错误的文件路径?我也遇到了和你一样的错误 — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.
我之前运行程序是会跑一段的,在448的时候突然停下来,在看到你的回复后,我仔细检查了两遍路径并下载了plane的文件,可是运行起来还是卡在了448的点,不知道老哥你是否能给些想法和建议: python eval_rcnn.py --cfg_file cfgs/default.yaml --ckpt PointRCNN.pth --batch_size 1 --eval_mode rcnn --set RPN.LOC_XZ_FINE False /usr/local/lib/python3.6/dist-packages/numba/cuda/envvars.py:17: NumbaWarning: Environment variables with the 'NUMBAPRO' prefix are deprecated and consequently ignored, found use of NUMBAPRO_NVVM=/usr/local/cuda-10.0/nvvm/lib64/libnvvm.so.
For more information about alternatives visit: ('https://numba.pydata.org/numba-doc/latest/cuda/overview.html', '#cudatoolkit-lookup') warnings.warn(errors.NumbaWarning(msg)) /usr/local/lib/python3.6/dist-packages/numba/cuda/envvars.py:17: NumbaWarning: Environment variables with the 'NUMBAPRO' prefix are deprecated and consequently ignored, found use of NUMBAPRO_LIBDEVICE=/usr/local/cuda-10.0/nvvm/libdevice/.
For more information about alternatives visit: ('https://numba.pydata.org/numba-doc/latest/cuda/overview.html', '#cudatoolkit-lookup') warnings.warn(errors.NumbaWarning(msg)) /home/wg/PointRCNN/tools/../lib/config.py:187: YAMLLoadWarning: calling yaml.load() without Loader=... is deprecated, as the default Loader is unsafe. Please read https://msg.pyyaml.org/load for full details. yaml_cfg = edict(yaml.load(f))
2021-09-22 04:35:14,630 INFO **Start logging** 2021-09-22 04:35:14,630 INFO cfg_file cfgs/default.yaml 2021-09-22 04:35:14,630 INFO eval_mode rcnn 2021-09-22 04:35:14,630 INFO eval_all False 2021-09-22 04:35:14,630 INFO test False 2021-09-22 04:35:14,630 INFO ckpt PointRCNN.pth 2021-09-22 04:35:14,630 INFO rpn_ckpt None 2021-09-22 04:35:14,630 INFO rcnn_ckpt None 2021-09-22 04:35:14,630 INFO batch_size 1 2021-09-22 04:35:14,630 INFO workers 4 2021-09-22 04:35:14,630 INFO extra_tag default 2021-09-22 04:35:14,630 INFO output_dir None 2021-09-22 04:35:14,630 INFO ckpt_dir None 2021-09-22 04:35:14,630 INFO save_result False 2021-09-22 04:35:14,630 INFO save_rpn_feature False 2021-09-22 04:35:14,630 INFO random_select True 2021-09-22 04:35:14,630 INFO start_epoch 0 2021-09-22 04:35:14,630 INFO rcnn_eval_roi_dir None 2021-09-22 04:35:14,630 INFO rcnn_eval_feature_dir None 2021-09-22 04:35:14,630 INFO set_cfgs ['RPN.LOC_XZ_FINE', 'False'] 2021-09-22 04:35:14,630 INFO cfg.TAG: default 2021-09-22 04:35:14,630 INFO cfg.CLASSES: Car 2021-09-22 04:35:14,630 INFO cfg.INCLUDE_SIMILAR_TYPE: True 2021-09-22 04:35:14,630 INFO cfg.AUG_DATA: True 2021-09-22 04:35:14,630 INFO cfg.AUG_METHOD_LIST: ['rotation', 'scaling', 'flip'] 2021-09-22 04:35:14,630 INFO cfg.AUG_METHOD_PROB: [1.0, 1.0, 0.5] 2021-09-22 04:35:14,630 INFO cfg.AUG_ROT_RANGE: 18 2021-09-22 04:35:14,630 INFO cfg.GT_AUG_ENABLED: True 2021-09-22 04:35:14,631 INFO cfg.GT_EXTRA_NUM: 15 2021-09-22 04:35:14,631 INFO cfg.GT_AUG_RAND_NUM: True 2021-09-22 04:35:14,631 INFO cfg.GT_AUG_APPLY_PROB: 1.0 2021-09-22 04:35:14,631 INFO cfg.GT_AUG_HARD_RATIO: 0.6 2021-09-22 04:35:14,631 INFO cfg.PC_REDUCE_BY_RANGE: True 2021-09-22 04:35:14,631 INFO cfg.PC_AREA_SCOPE: [[-40. 40. ] [ -1. 3. ] [ 0. 70.4]] 2021-09-22 04:35:14,631 INFO cfg.CLS_MEAN_SIZE: [[1.5256319 1.6285675 3.8831165]] 2021-09-22 04:35:14,631 INFO cfg.RPN = edict() 2021-09-22 04:35:14,631 INFO cfg.RPN.ENABLED: True 2021-09-22 04:35:14,631 INFO cfg.RPN.FIXED: True 2021-09-22 04:35:14,631 INFO cfg.RPN.USE_INTENSITY: False 2021-09-22 04:35:14,631 INFO cfg.RPN.LOC_XZ_FINE: False 2021-09-22 04:35:14,631 INFO cfg.RPN.LOC_SCOPE: 3.0 2021-09-22 04:35:14,631 INFO cfg.RPN.LOC_BIN_SIZE: 0.5 2021-09-22 04:35:14,631 INFO cfg.RPN.NUM_HEAD_BIN: 12 2021-09-22 04:35:14,631 INFO cfg.RPN.BACKBONE: pointnet2_msg 2021-09-22 04:35:14,631 INFO cfg.RPN.USE_BN: True 2021-09-22 04:35:14,631 INFO cfg.RPN.NUM_POINTS: 16384 2021-09-22 04:35:14,631 INFO cfg.RPN.SA_CONFIG = edict() 2021-09-22 04:35:14,631 INFO cfg.RPN.SA_CONFIG.NPOINTS: [4096, 1024, 256, 64] 2021-09-22 04:35:14,631 INFO cfg.RPN.SA_CONFIG.RADIUS: [[0.1, 0.5], [0.5, 1.0], [1.0, 2.0], [2.0, 4.0]] 2021-09-22 04:35:14,632 INFO cfg.RPN.SA_CONFIG.NSAMPLE: [[16, 32], [16, 32], [16, 32], [16, 32]] 2021-09-22 04:35:14,632 INFO cfg.RPN.SA_CONFIG.MLPS: [[[16, 16, 32], [32, 32, 64]], [[64, 64, 128], [64, 96, 128]], [[128, 196, 256], [128, 196, 256]], [[256, 256, 512], [256, 384, 512]]] 2021-09-22 04:35:14,632 INFO cfg.RPN.FP_MLPS: [[128, 128], [256, 256], [512, 512], [512, 512]] 2021-09-22 04:35:14,632 INFO cfg.RPN.CLS_FC: [128] 2021-09-22 04:35:14,632 INFO cfg.RPN.REG_FC: [128] 2021-09-22 04:35:14,632 INFO cfg.RPN.DP_RATIO: 0.5 2021-09-22 04:35:14,632 INFO cfg.RPN.LOSS_CLS: SigmoidFocalLoss 2021-09-22 04:35:14,632 INFO cfg.RPN.FG_WEIGHT: 15 2021-09-22 04:35:14,632 INFO cfg.RPN.FOCAL_ALPHA: [0.25, 0.75] 2021-09-22 04:35:14,632 INFO cfg.RPN.FOCAL_GAMMA: 2.0 2021-09-22 04:35:14,632 INFO cfg.RPN.REG_LOSS_WEIGHT: [1.0, 1.0, 1.0, 1.0] 2021-09-22 04:35:14,632 INFO cfg.RPN.LOSS_WEIGHT: [1.0, 1.0] 2021-09-22 04:35:14,632 INFO cfg.RPN.NMS_TYPE: normal 2021-09-22 04:35:14,632 INFO cfg.RPN.SCORE_THRESH: 0.3 2021-09-22 04:35:14,632 INFO cfg.RCNN = edict() 2021-09-22 04:35:14,632 INFO cfg.RCNN.ENABLED: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_RPN_FEATURES: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_MASK: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.MASK_TYPE: seg 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_INTENSITY: False 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_DEPTH: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_SEG_SCORE: False 2021-09-22 04:35:14,632 INFO cfg.RCNN.ROI_SAMPLE_JIT: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.ROI_FG_AUG_TIMES: 10 2021-09-22 04:35:14,632 INFO cfg.RCNN.REG_AUG_METHOD: multiple 2021-09-22 04:35:14,632 INFO cfg.RCNN.POOL_EXTRA_WIDTH: 1.0 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_SCOPE: 1.5 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_BIN_SIZE: 0.5 2021-09-22 04:35:14,632 INFO cfg.RCNN.NUM_HEAD_BIN: 9 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_Y_BY_BIN: False 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_Y_SCOPE: 0.5 2021-09-22 04:35:14,633 INFO cfg.RCNN.LOC_Y_BIN_SIZE: 0.25 2021-09-22 04:35:14,633 INFO cfg.RCNN.SIZE_RES_ON_ROI: False 2021-09-22 04:35:14,633 INFO cfg.RCNN.USE_BN: False 2021-09-22 04:35:14,633 INFO cfg.RCNN.DP_RATIO: 0.0 2021-09-22 04:35:14,633 INFO cfg.RCNN.BACKBONE: pointnet 2021-09-22 04:35:14,633 INFO cfg.RCNN.XYZ_UP_LAYER: [128, 128] 2021-09-22 04:35:14,633 INFO cfg.RCNN.NUM_POINTS: 512 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG = edict() 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.NPOINTS: [128, 32, -1] 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.RADIUS: [0.2, 0.4, 100] 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.NSAMPLE: [64, 64, 64] 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.MLPS: [[128, 128, 128], [128, 128, 256], [256, 256, 512]] 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_FC: [256, 256] 2021-09-22 04:35:14,633 INFO cfg.RCNN.REG_FC: [256, 256] 2021-09-22 04:35:14,633 INFO cfg.RCNN.LOSS_CLS: BinaryCrossEntropy 2021-09-22 04:35:14,633 INFO cfg.RCNN.FOCAL_ALPHA: [0.25, 0.75] 2021-09-22 04:35:14,633 INFO cfg.RCNN.FOCAL_GAMMA: 2.0 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_WEIGHT: [1. 1. 1.] 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_FG_THRESH: 0.6 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_BG_THRESH: 0.45 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_BG_THRESH_LO: 0.05 2021-09-22 04:35:14,633 INFO cfg.RCNN.REG_FG_THRESH: 0.55 2021-09-22 04:35:14,633 INFO cfg.RCNN.FG_RATIO: 0.5 2021-09-22 04:35:14,633 INFO cfg.RCNN.ROI_PER_IMAGE: 64 2021-09-22 04:35:14,633 INFO cfg.RCNN.HARD_BG_RATIO: 0.8 2021-09-22 04:35:14,633 INFO cfg.RCNN.SCORE_THRESH: 0.3 2021-09-22 04:35:14,634 INFO cfg.RCNN.NMS_THRESH: 0.1 2021-09-22 04:35:14,634 INFO cfg.TRAIN = edict() 2021-09-22 04:35:14,634 INFO cfg.TRAIN.SPLIT: train 2021-09-22 04:35:14,634 INFO cfg.TRAIN.VAL_SPLIT: smallval 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR: 0.002 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR_CLIP: 1e-05 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR_DECAY: 0.5 2021-09-22 04:35:14,634 INFO cfg.TRAIN.DECAY_STEP_LIST: [100, 150, 180, 200] 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR_WARMUP: True 2021-09-22 04:35:14,634 INFO cfg.TRAIN.WARMUP_MIN: 0.0002 2021-09-22 04:35:14,634 INFO cfg.TRAIN.WARMUP_EPOCH: 1 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BN_MOMENTUM: 0.1 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BN_DECAY: 0.5 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BNM_CLIP: 0.01 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BN_DECAY_STEP_LIST: [1000] 2021-09-22 04:35:14,634 INFO cfg.TRAIN.OPTIMIZER: adam_onecycle 2021-09-22 04:35:14,634 INFO cfg.TRAIN.WEIGHT_DECAY: 0.001 2021-09-22 04:35:14,634 INFO cfg.TRAIN.MOMENTUM: 0.9 2021-09-22 04:35:14,634 INFO cfg.TRAIN.MOMS: [0.95, 0.85] 2021-09-22 04:35:14,634 INFO cfg.TRAIN.DIV_FACTOR: 10.0 2021-09-22 04:35:14,634 INFO cfg.TRAIN.PCT_START: 0.4 2021-09-22 04:35:14,634 INFO cfg.TRAIN.GRAD_NORM_CLIP: 1.0 2021-09-22 04:35:14,634 INFO cfg.TRAIN.RPN_PRE_NMS_TOP_N: 9000 2021-09-22 04:35:14,634 INFO cfg.TRAIN.RPN_POST_NMS_TOP_N: 512 2021-09-22 04:35:14,634 INFO cfg.TRAIN.RPN_NMS_THRESH: 0.85 2021-09-22 04:35:14,635 INFO cfg.TRAIN.RPN_DISTANCE_BASED_PROPOSE: True 2021-09-22 04:35:14,635 INFO cfg.TEST = edict() 2021-09-22 04:35:14,635 INFO cfg.TEST.SPLIT: val 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_PRE_NMS_TOP_N: 9000 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_POST_NMS_TOP_N: 100 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_NMS_THRESH: 0.8 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_DISTANCE_BASED_PROPOSE: True 2021-09-22 04:35:14,636 INFO Load testing samples from ../data/KITTI/object/training 2021-09-22 04:35:14,636 INFO Done: total test samples 3769 2021-09-22 04:35:16,962 INFO ==> Loading from checkpoint 'PointRCNN.pth' 2021-09-22 04:35:16,989 INFO ==> Done 2021-09-22 04:35:16,990 INFO ---- EPOCH no_number JOINT EVALUATION ---- 2021-09-22 04:35:16,990 INFO ==> Output file: ../output/rcnn/default/eval/epoch_no_number/val eval: 13%|▋ | 488/3769 [00:55<05:59, 9.13it/s, mode=EVAL, recall=1170/1769]Traceback (most recent call last): File "eval_rcnn.py", line 902, in eval_single_ckpt(root_result_dir) File "eval_rcnn.py", line 765, in eval_single_ckpt eval_one_epoch(model, test_loader, epoch_id, root_result_dir, logger) File "eval_rcnn.py", line 692, in eval_one_epoch ret_dict = eval_one_epoch_joint(model, dataloader, epoch_id, result_dir, logger) File "eval_rcnn.py", line 486, in eval_one_epoch_joint for data in dataloader: File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 582, in next return self._process_next_batch(batch) File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 608, in _process_next_batch raise batch.exc_type(batch.exc_msg) AssertionError: Traceback (most recent call last): File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/_utils/worker.py", line 99, in _worker_loop samples = collate_fn([dataset[i] for i in batch_indices]) File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/_utils/worker.py", line 99, in samples = collate_fn([dataset[i] for i in batch_indices]) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_rcnn_dataset.py", line 234, in getitem return self.get_rpn_sample(index) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_rcnn_dataset.py", line 249, in get_rpn_sample calib = self.get_calib(sample_id) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_rcnn_dataset.py", line 133, in get_calib return super().get_calib(idx % 10000) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_dataset.py", line 47, in get_calib assert os.path.exists(calib_file) AssertionError
eval: 13%|▋ | 488/3769 [00:55<06:12, 8.80it/s, mode=EVAL, recall=1170/1769]
Hello? I found that I have the same problem with you! My process stopped when 448 13% too! Then I think that probably is not because the lack or wrong file path of dataset, what other problem??
检查一下数据集的路径是否正确,数据集的排布是否和作者的一样。还有那个plane这个文件也得下。 … ---Original--- From: @.**> Date: Wed, Sep 22, 2021 15:43 PM To: @.**>; Cc: @.**@.**>; Subject: Re: [sshaoshuai/PointRCNN] assert os.path.exists(calib_file) when run eval_rcnn.py (#87) thanks,I find the problem which is because of the wrong of the file path 老哥,能不能展开说说,啥错误的文件路径?我也遇到了和你一样的错误 — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.
我之前运行程序是会跑一段的,在448的时候突然停下来,在看到你的回复后,我仔细检查了两遍路径并下载了plane的文件,可是运行起来还是卡在了448的点,不知道老哥你是否能给些想法和建议: python eval_rcnn.py --cfg_file cfgs/default.yaml --ckpt PointRCNN.pth --batch_size 1 --eval_mode rcnn --set RPN.LOC_XZ_FINE False /usr/local/lib/python3.6/dist-packages/numba/cuda/envvars.py:17: NumbaWarning: Environment variables with the 'NUMBAPRO' prefix are deprecated and consequently ignored, found use of NUMBAPRO_NVVM=/usr/local/cuda-10.0/nvvm/lib64/libnvvm.so. For more information about alternatives visit: ('https://numba.pydata.org/numba-doc/latest/cuda/overview.html', '#cudatoolkit-lookup') warnings.warn(errors.NumbaWarning(msg)) /usr/local/lib/python3.6/dist-packages/numba/cuda/envvars.py:17: NumbaWarning: Environment variables with the 'NUMBAPRO' prefix are deprecated and consequently ignored, found use of NUMBAPRO_LIBDEVICE=/usr/local/cuda-10.0/nvvm/libdevice/. For more information about alternatives visit: ('https://numba.pydata.org/numba-doc/latest/cuda/overview.html', '#cudatoolkit-lookup') warnings.warn(errors.NumbaWarning(msg)) /home/wg/PointRCNN/tools/../lib/config.py:187: YAMLLoadWarning: calling yaml.load() without Loader=... is deprecated, as the default Loader is unsafe. Please read https://msg.pyyaml.org/load for full details. yaml_cfg = edict(yaml.load(f)) 2021-09-22 04:35:14,630 INFO **Start logging** 2021-09-22 04:35:14,630 INFO cfg_file cfgs/default.yaml 2021-09-22 04:35:14,630 INFO eval_mode rcnn 2021-09-22 04:35:14,630 INFO eval_all False 2021-09-22 04:35:14,630 INFO test False 2021-09-22 04:35:14,630 INFO ckpt PointRCNN.pth 2021-09-22 04:35:14,630 INFO rpn_ckpt None 2021-09-22 04:35:14,630 INFO rcnn_ckpt None 2021-09-22 04:35:14,630 INFO batch_size 1 2021-09-22 04:35:14,630 INFO workers 4 2021-09-22 04:35:14,630 INFO extra_tag default 2021-09-22 04:35:14,630 INFO output_dir None 2021-09-22 04:35:14,630 INFO ckpt_dir None 2021-09-22 04:35:14,630 INFO save_result False 2021-09-22 04:35:14,630 INFO save_rpn_feature False 2021-09-22 04:35:14,630 INFO random_select True 2021-09-22 04:35:14,630 INFO start_epoch 0 2021-09-22 04:35:14,630 INFO rcnn_eval_roi_dir None 2021-09-22 04:35:14,630 INFO rcnn_eval_feature_dir None 2021-09-22 04:35:14,630 INFO set_cfgs ['RPN.LOC_XZ_FINE', 'False'] 2021-09-22 04:35:14,630 INFO cfg.TAG: default 2021-09-22 04:35:14,630 INFO cfg.CLASSES: Car 2021-09-22 04:35:14,630 INFO cfg.INCLUDE_SIMILAR_TYPE: True 2021-09-22 04:35:14,630 INFO cfg.AUG_DATA: True 2021-09-22 04:35:14,630 INFO cfg.AUG_METHOD_LIST: ['rotation', 'scaling', 'flip'] 2021-09-22 04:35:14,630 INFO cfg.AUG_METHOD_PROB: [1.0, 1.0, 0.5] 2021-09-22 04:35:14,630 INFO cfg.AUG_ROT_RANGE: 18 2021-09-22 04:35:14,630 INFO cfg.GT_AUG_ENABLED: True 2021-09-22 04:35:14,631 INFO cfg.GT_EXTRA_NUM: 15 2021-09-22 04:35:14,631 INFO cfg.GT_AUG_RAND_NUM: True 2021-09-22 04:35:14,631 INFO cfg.GT_AUG_APPLY_PROB: 1.0 2021-09-22 04:35:14,631 INFO cfg.GT_AUG_HARD_RATIO: 0.6 2021-09-22 04:35:14,631 INFO cfg.PC_REDUCE_BY_RANGE: True 2021-09-22 04:35:14,631 INFO cfg.PC_AREA_SCOPE: [[-40. 40. ] [ -1. 3. ] [ 0. 70.4]] 2021-09-22 04:35:14,631 INFO cfg.CLS_MEAN_SIZE: [[1.5256319 1.6285675 3.8831165]] 2021-09-22 04:35:14,631 INFO cfg.RPN = edict() 2021-09-22 04:35:14,631 INFO cfg.RPN.ENABLED: True 2021-09-22 04:35:14,631 INFO cfg.RPN.FIXED: True 2021-09-22 04:35:14,631 INFO cfg.RPN.USE_INTENSITY: False 2021-09-22 04:35:14,631 INFO cfg.RPN.LOC_XZ_FINE: False 2021-09-22 04:35:14,631 INFO cfg.RPN.LOC_SCOPE: 3.0 2021-09-22 04:35:14,631 INFO cfg.RPN.LOC_BIN_SIZE: 0.5 2021-09-22 04:35:14,631 INFO cfg.RPN.NUM_HEAD_BIN: 12 2021-09-22 04:35:14,631 INFO cfg.RPN.BACKBONE: pointnet2_msg 2021-09-22 04:35:14,631 INFO cfg.RPN.USE_BN: True 2021-09-22 04:35:14,631 INFO cfg.RPN.NUM_POINTS: 16384 2021-09-22 04:35:14,631 INFO cfg.RPN.SA_CONFIG = edict() 2021-09-22 04:35:14,631 INFO cfg.RPN.SA_CONFIG.NPOINTS: [4096, 1024, 256, 64] 2021-09-22 04:35:14,631 INFO cfg.RPN.SA_CONFIG.RADIUS: [[0.1, 0.5], [0.5, 1.0], [1.0, 2.0], [2.0, 4.0]] 2021-09-22 04:35:14,632 INFO cfg.RPN.SA_CONFIG.NSAMPLE: [[16, 32], [16, 32], [16, 32], [16, 32]] 2021-09-22 04:35:14,632 INFO cfg.RPN.SA_CONFIG.MLPS: [[[16, 16, 32], [32, 32, 64]], [[64, 64, 128], [64, 96, 128]], [[128, 196, 256], [128, 196, 256]], [[256, 256, 512], [256, 384, 512]]] 2021-09-22 04:35:14,632 INFO cfg.RPN.FP_MLPS: [[128, 128], [256, 256], [512, 512], [512, 512]] 2021-09-22 04:35:14,632 INFO cfg.RPN.CLS_FC: [128] 2021-09-22 04:35:14,632 INFO cfg.RPN.REG_FC: [128] 2021-09-22 04:35:14,632 INFO cfg.RPN.DP_RATIO: 0.5 2021-09-22 04:35:14,632 INFO cfg.RPN.LOSS_CLS: SigmoidFocalLoss 2021-09-22 04:35:14,632 INFO cfg.RPN.FG_WEIGHT: 15 2021-09-22 04:35:14,632 INFO cfg.RPN.FOCAL_ALPHA: [0.25, 0.75] 2021-09-22 04:35:14,632 INFO cfg.RPN.FOCAL_GAMMA: 2.0 2021-09-22 04:35:14,632 INFO cfg.RPN.REG_LOSS_WEIGHT: [1.0, 1.0, 1.0, 1.0] 2021-09-22 04:35:14,632 INFO cfg.RPN.LOSS_WEIGHT: [1.0, 1.0] 2021-09-22 04:35:14,632 INFO cfg.RPN.NMS_TYPE: normal 2021-09-22 04:35:14,632 INFO cfg.RPN.SCORE_THRESH: 0.3 2021-09-22 04:35:14,632 INFO cfg.RCNN = edict() 2021-09-22 04:35:14,632 INFO cfg.RCNN.ENABLED: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_RPN_FEATURES: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_MASK: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.MASK_TYPE: seg 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_INTENSITY: False 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_DEPTH: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_SEG_SCORE: False 2021-09-22 04:35:14,632 INFO cfg.RCNN.ROI_SAMPLE_JIT: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.ROI_FG_AUG_TIMES: 10 2021-09-22 04:35:14,632 INFO cfg.RCNN.REG_AUG_METHOD: multiple 2021-09-22 04:35:14,632 INFO cfg.RCNN.POOL_EXTRA_WIDTH: 1.0 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_SCOPE: 1.5 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_BIN_SIZE: 0.5 2021-09-22 04:35:14,632 INFO cfg.RCNN.NUM_HEAD_BIN: 9 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_Y_BY_BIN: False 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_Y_SCOPE: 0.5 2021-09-22 04:35:14,633 INFO cfg.RCNN.LOC_Y_BIN_SIZE: 0.25 2021-09-22 04:35:14,633 INFO cfg.RCNN.SIZE_RES_ON_ROI: False 2021-09-22 04:35:14,633 INFO cfg.RCNN.USE_BN: False 2021-09-22 04:35:14,633 INFO cfg.RCNN.DP_RATIO: 0.0 2021-09-22 04:35:14,633 INFO cfg.RCNN.BACKBONE: pointnet 2021-09-22 04:35:14,633 INFO cfg.RCNN.XYZ_UP_LAYER: [128, 128] 2021-09-22 04:35:14,633 INFO cfg.RCNN.NUM_POINTS: 512 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG = edict() 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.NPOINTS: [128, 32, -1] 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.RADIUS: [0.2, 0.4, 100] 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.NSAMPLE: [64, 64, 64] 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.MLPS: [[128, 128, 128], [128, 128, 256], [256, 256, 512]] 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_FC: [256, 256] 2021-09-22 04:35:14,633 INFO cfg.RCNN.REG_FC: [256, 256] 2021-09-22 04:35:14,633 INFO cfg.RCNN.LOSS_CLS: BinaryCrossEntropy 2021-09-22 04:35:14,633 INFO cfg.RCNN.FOCAL_ALPHA: [0.25, 0.75] 2021-09-22 04:35:14,633 INFO cfg.RCNN.FOCAL_GAMMA: 2.0 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_WEIGHT: [1. 1. 1.] 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_FG_THRESH: 0.6 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_BG_THRESH: 0.45 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_BG_THRESH_LO: 0.05 2021-09-22 04:35:14,633 INFO cfg.RCNN.REG_FG_THRESH: 0.55 2021-09-22 04:35:14,633 INFO cfg.RCNN.FG_RATIO: 0.5 2021-09-22 04:35:14,633 INFO cfg.RCNN.ROI_PER_IMAGE: 64 2021-09-22 04:35:14,633 INFO cfg.RCNN.HARD_BG_RATIO: 0.8 2021-09-22 04:35:14,633 INFO cfg.RCNN.SCORE_THRESH: 0.3 2021-09-22 04:35:14,634 INFO cfg.RCNN.NMS_THRESH: 0.1 2021-09-22 04:35:14,634 INFO cfg.TRAIN = edict() 2021-09-22 04:35:14,634 INFO cfg.TRAIN.SPLIT: train 2021-09-22 04:35:14,634 INFO cfg.TRAIN.VAL_SPLIT: smallval 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR: 0.002 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR_CLIP: 1e-05 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR_DECAY: 0.5 2021-09-22 04:35:14,634 INFO cfg.TRAIN.DECAY_STEP_LIST: [100, 150, 180, 200] 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR_WARMUP: True 2021-09-22 04:35:14,634 INFO cfg.TRAIN.WARMUP_MIN: 0.0002 2021-09-22 04:35:14,634 INFO cfg.TRAIN.WARMUP_EPOCH: 1 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BN_MOMENTUM: 0.1 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BN_DECAY: 0.5 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BNM_CLIP: 0.01 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BN_DECAY_STEP_LIST: [1000] 2021-09-22 04:35:14,634 INFO cfg.TRAIN.OPTIMIZER: adam_onecycle 2021-09-22 04:35:14,634 INFO cfg.TRAIN.WEIGHT_DECAY: 0.001 2021-09-22 04:35:14,634 INFO cfg.TRAIN.MOMENTUM: 0.9 2021-09-22 04:35:14,634 INFO cfg.TRAIN.MOMS: [0.95, 0.85] 2021-09-22 04:35:14,634 INFO cfg.TRAIN.DIV_FACTOR: 10.0 2021-09-22 04:35:14,634 INFO cfg.TRAIN.PCT_START: 0.4 2021-09-22 04:35:14,634 INFO cfg.TRAIN.GRAD_NORM_CLIP: 1.0 2021-09-22 04:35:14,634 INFO cfg.TRAIN.RPN_PRE_NMS_TOP_N: 9000 2021-09-22 04:35:14,634 INFO cfg.TRAIN.RPN_POST_NMS_TOP_N: 512 2021-09-22 04:35:14,634 INFO cfg.TRAIN.RPN_NMS_THRESH: 0.85 2021-09-22 04:35:14,635 INFO cfg.TRAIN.RPN_DISTANCE_BASED_PROPOSE: True 2021-09-22 04:35:14,635 INFO cfg.TEST = edict() 2021-09-22 04:35:14,635 INFO cfg.TEST.SPLIT: val 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_PRE_NMS_TOP_N: 9000 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_POST_NMS_TOP_N: 100 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_NMS_THRESH: 0.8 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_DISTANCE_BASED_PROPOSE: True 2021-09-22 04:35:14,636 INFO Load testing samples from ../data/KITTI/object/training 2021-09-22 04:35:14,636 INFO Done: total test samples 3769 2021-09-22 04:35:16,962 INFO ==> Loading from checkpoint 'PointRCNN.pth' 2021-09-22 04:35:16,989 INFO ==> Done 2021-09-22 04:35:16,990 INFO ---- EPOCH no_number JOINT EVALUATION ---- 2021-09-22 04:35:16,990 INFO ==> Output file: ../output/rcnn/default/eval/epoch_no_number/val eval: 13%|▋ | 488/3769 [00:55<05:59, 9.13it/s, mode=EVAL, recall=1170/1769]Traceback (most recent call last): File "eval_rcnn.py", line 902, in eval_single_ckpt(root_result_dir) File "eval_rcnn.py", line 765, in eval_single_ckpt eval_one_epoch(model, test_loader, epoch_id, root_result_dir, logger) File "eval_rcnn.py", line 692, in eval_one_epoch ret_dict = eval_one_epoch_joint(model, dataloader, epoch_id, result_dir, logger) File "eval_rcnn.py", line 486, in eval_one_epoch_joint for data in dataloader: File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 582, in next return self._process_next_batch(batch) File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 608, in _process_next_batch raise batch.exc_type(batch.exc_msg) AssertionError: Traceback (most recent call last): File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/_utils/worker.py", line 99, in _worker_loop samples = collate_fn([dataset[i] for i in batch_indices]) File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/_utils/worker.py", line 99, in samples = collate_fn([dataset[i] for i in batch_indices]) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_rcnn_dataset.py", line 234, in getitem return self.get_rpn_sample(index) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_rcnn_dataset.py", line 249, in get_rpn_sample calib = self.get_calib(sample_id) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_rcnn_dataset.py", line 133, in get_calib return super().get_calib(idx % 10000) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_dataset.py", line 47, in get_calib assert os.path.exists(calib_file) AssertionError eval: 13%|▋ | 488/3769 [00:55<06:12, 8.80it/s, mode=EVAL, recall=1170/1769]
Hello? I found that I have the same problem with you! My process stopped when 448 13% too! Then I think that probably is not because the lack or wrong file path of dataset, what other problem??
嗯...我的问题就是因为数据集排列的路径不对。你可以仔细检查一下是否已经下载了planes数据集并且数据集排列的顺序没有错误。我并没有遇到其他问题
检查一下数据集的路径是否正确,数据集的排布是否和作者的一样。还有那个plane这个文件也得下。 … ---Original--- From: @.**> Date: Wed, Sep 22, 2021 15:43 PM To: @.**>; Cc: @.**@.**>; Subject: Re: [sshaoshuai/PointRCNN] assert os.path.exists(calib_file) when run eval_rcnn.py (#87) thanks,I find the problem which is because of the wrong of the file path 老哥,能不能展开说说,啥错误的文件路径?我也遇到了和你一样的错误 — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.
我之前运行程序是会跑一段的,在448的时候突然停下来,在看到你的回复后,我仔细检查了两遍路径并下载了plane的文件,可是运行起来还是卡在了448的点,不知道老哥你是否能给些想法和建议: python eval_rcnn.py --cfg_file cfgs/default.yaml --ckpt PointRCNN.pth --batch_size 1 --eval_mode rcnn --set RPN.LOC_XZ_FINE False /usr/local/lib/python3.6/dist-packages/numba/cuda/envvars.py:17: NumbaWarning: Environment variables with the 'NUMBAPRO' prefix are deprecated and consequently ignored, found use of NUMBAPRO_NVVM=/usr/local/cuda-10.0/nvvm/lib64/libnvvm.so. For more information about alternatives visit: ('https://numba.pydata.org/numba-doc/latest/cuda/overview.html', '#cudatoolkit-lookup') warnings.warn(errors.NumbaWarning(msg)) /usr/local/lib/python3.6/dist-packages/numba/cuda/envvars.py:17: NumbaWarning: Environment variables with the 'NUMBAPRO' prefix are deprecated and consequently ignored, found use of NUMBAPRO_LIBDEVICE=/usr/local/cuda-10.0/nvvm/libdevice/. For more information about alternatives visit: ('https://numba.pydata.org/numba-doc/latest/cuda/overview.html', '#cudatoolkit-lookup') warnings.warn(errors.NumbaWarning(msg)) /home/wg/PointRCNN/tools/../lib/config.py:187: YAMLLoadWarning: calling yaml.load() without Loader=... is deprecated, as the default Loader is unsafe. Please read https://msg.pyyaml.org/load for full details. yaml_cfg = edict(yaml.load(f)) 2021-09-22 04:35:14,630 INFO **Start logging** 2021-09-22 04:35:14,630 INFO cfg_file cfgs/default.yaml 2021-09-22 04:35:14,630 INFO eval_mode rcnn 2021-09-22 04:35:14,630 INFO eval_all False 2021-09-22 04:35:14,630 INFO test False 2021-09-22 04:35:14,630 INFO ckpt PointRCNN.pth 2021-09-22 04:35:14,630 INFO rpn_ckpt None 2021-09-22 04:35:14,630 INFO rcnn_ckpt None 2021-09-22 04:35:14,630 INFO batch_size 1 2021-09-22 04:35:14,630 INFO workers 4 2021-09-22 04:35:14,630 INFO extra_tag default 2021-09-22 04:35:14,630 INFO output_dir None 2021-09-22 04:35:14,630 INFO ckpt_dir None 2021-09-22 04:35:14,630 INFO save_result False 2021-09-22 04:35:14,630 INFO save_rpn_feature False 2021-09-22 04:35:14,630 INFO random_select True 2021-09-22 04:35:14,630 INFO start_epoch 0 2021-09-22 04:35:14,630 INFO rcnn_eval_roi_dir None 2021-09-22 04:35:14,630 INFO rcnn_eval_feature_dir None 2021-09-22 04:35:14,630 INFO set_cfgs ['RPN.LOC_XZ_FINE', 'False'] 2021-09-22 04:35:14,630 INFO cfg.TAG: default 2021-09-22 04:35:14,630 INFO cfg.CLASSES: Car 2021-09-22 04:35:14,630 INFO cfg.INCLUDE_SIMILAR_TYPE: True 2021-09-22 04:35:14,630 INFO cfg.AUG_DATA: True 2021-09-22 04:35:14,630 INFO cfg.AUG_METHOD_LIST: ['rotation', 'scaling', 'flip'] 2021-09-22 04:35:14,630 INFO cfg.AUG_METHOD_PROB: [1.0, 1.0, 0.5] 2021-09-22 04:35:14,630 INFO cfg.AUG_ROT_RANGE: 18 2021-09-22 04:35:14,630 INFO cfg.GT_AUG_ENABLED: True 2021-09-22 04:35:14,631 INFO cfg.GT_EXTRA_NUM: 15 2021-09-22 04:35:14,631 INFO cfg.GT_AUG_RAND_NUM: True 2021-09-22 04:35:14,631 INFO cfg.GT_AUG_APPLY_PROB: 1.0 2021-09-22 04:35:14,631 INFO cfg.GT_AUG_HARD_RATIO: 0.6 2021-09-22 04:35:14,631 INFO cfg.PC_REDUCE_BY_RANGE: True 2021-09-22 04:35:14,631 INFO cfg.PC_AREA_SCOPE: [[-40. 40. ] [ -1. 3. ] [ 0. 70.4]] 2021-09-22 04:35:14,631 INFO cfg.CLS_MEAN_SIZE: [[1.5256319 1.6285675 3.8831165]] 2021-09-22 04:35:14,631 INFO cfg.RPN = edict() 2021-09-22 04:35:14,631 INFO cfg.RPN.ENABLED: True 2021-09-22 04:35:14,631 INFO cfg.RPN.FIXED: True 2021-09-22 04:35:14,631 INFO cfg.RPN.USE_INTENSITY: False 2021-09-22 04:35:14,631 INFO cfg.RPN.LOC_XZ_FINE: False 2021-09-22 04:35:14,631 INFO cfg.RPN.LOC_SCOPE: 3.0 2021-09-22 04:35:14,631 INFO cfg.RPN.LOC_BIN_SIZE: 0.5 2021-09-22 04:35:14,631 INFO cfg.RPN.NUM_HEAD_BIN: 12 2021-09-22 04:35:14,631 INFO cfg.RPN.BACKBONE: pointnet2_msg 2021-09-22 04:35:14,631 INFO cfg.RPN.USE_BN: True 2021-09-22 04:35:14,631 INFO cfg.RPN.NUM_POINTS: 16384 2021-09-22 04:35:14,631 INFO cfg.RPN.SA_CONFIG = edict() 2021-09-22 04:35:14,631 INFO cfg.RPN.SA_CONFIG.NPOINTS: [4096, 1024, 256, 64] 2021-09-22 04:35:14,631 INFO cfg.RPN.SA_CONFIG.RADIUS: [[0.1, 0.5], [0.5, 1.0], [1.0, 2.0], [2.0, 4.0]] 2021-09-22 04:35:14,632 INFO cfg.RPN.SA_CONFIG.NSAMPLE: [[16, 32], [16, 32], [16, 32], [16, 32]] 2021-09-22 04:35:14,632 INFO cfg.RPN.SA_CONFIG.MLPS: [[[16, 16, 32], [32, 32, 64]], [[64, 64, 128], [64, 96, 128]], [[128, 196, 256], [128, 196, 256]], [[256, 256, 512], [256, 384, 512]]] 2021-09-22 04:35:14,632 INFO cfg.RPN.FP_MLPS: [[128, 128], [256, 256], [512, 512], [512, 512]] 2021-09-22 04:35:14,632 INFO cfg.RPN.CLS_FC: [128] 2021-09-22 04:35:14,632 INFO cfg.RPN.REG_FC: [128] 2021-09-22 04:35:14,632 INFO cfg.RPN.DP_RATIO: 0.5 2021-09-22 04:35:14,632 INFO cfg.RPN.LOSS_CLS: SigmoidFocalLoss 2021-09-22 04:35:14,632 INFO cfg.RPN.FG_WEIGHT: 15 2021-09-22 04:35:14,632 INFO cfg.RPN.FOCAL_ALPHA: [0.25, 0.75] 2021-09-22 04:35:14,632 INFO cfg.RPN.FOCAL_GAMMA: 2.0 2021-09-22 04:35:14,632 INFO cfg.RPN.REG_LOSS_WEIGHT: [1.0, 1.0, 1.0, 1.0] 2021-09-22 04:35:14,632 INFO cfg.RPN.LOSS_WEIGHT: [1.0, 1.0] 2021-09-22 04:35:14,632 INFO cfg.RPN.NMS_TYPE: normal 2021-09-22 04:35:14,632 INFO cfg.RPN.SCORE_THRESH: 0.3 2021-09-22 04:35:14,632 INFO cfg.RCNN = edict() 2021-09-22 04:35:14,632 INFO cfg.RCNN.ENABLED: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_RPN_FEATURES: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_MASK: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.MASK_TYPE: seg 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_INTENSITY: False 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_DEPTH: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_SEG_SCORE: False 2021-09-22 04:35:14,632 INFO cfg.RCNN.ROI_SAMPLE_JIT: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.ROI_FG_AUG_TIMES: 10 2021-09-22 04:35:14,632 INFO cfg.RCNN.REG_AUG_METHOD: multiple 2021-09-22 04:35:14,632 INFO cfg.RCNN.POOL_EXTRA_WIDTH: 1.0 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_SCOPE: 1.5 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_BIN_SIZE: 0.5 2021-09-22 04:35:14,632 INFO cfg.RCNN.NUM_HEAD_BIN: 9 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_Y_BY_BIN: False 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_Y_SCOPE: 0.5 2021-09-22 04:35:14,633 INFO cfg.RCNN.LOC_Y_BIN_SIZE: 0.25 2021-09-22 04:35:14,633 INFO cfg.RCNN.SIZE_RES_ON_ROI: False 2021-09-22 04:35:14,633 INFO cfg.RCNN.USE_BN: False 2021-09-22 04:35:14,633 INFO cfg.RCNN.DP_RATIO: 0.0 2021-09-22 04:35:14,633 INFO cfg.RCNN.BACKBONE: pointnet 2021-09-22 04:35:14,633 INFO cfg.RCNN.XYZ_UP_LAYER: [128, 128] 2021-09-22 04:35:14,633 INFO cfg.RCNN.NUM_POINTS: 512 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG = edict() 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.NPOINTS: [128, 32, -1] 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.RADIUS: [0.2, 0.4, 100] 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.NSAMPLE: [64, 64, 64] 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.MLPS: [[128, 128, 128], [128, 128, 256], [256, 256, 512]] 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_FC: [256, 256] 2021-09-22 04:35:14,633 INFO cfg.RCNN.REG_FC: [256, 256] 2021-09-22 04:35:14,633 INFO cfg.RCNN.LOSS_CLS: BinaryCrossEntropy 2021-09-22 04:35:14,633 INFO cfg.RCNN.FOCAL_ALPHA: [0.25, 0.75] 2021-09-22 04:35:14,633 INFO cfg.RCNN.FOCAL_GAMMA: 2.0 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_WEIGHT: [1. 1. 1.] 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_FG_THRESH: 0.6 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_BG_THRESH: 0.45 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_BG_THRESH_LO: 0.05 2021-09-22 04:35:14,633 INFO cfg.RCNN.REG_FG_THRESH: 0.55 2021-09-22 04:35:14,633 INFO cfg.RCNN.FG_RATIO: 0.5 2021-09-22 04:35:14,633 INFO cfg.RCNN.ROI_PER_IMAGE: 64 2021-09-22 04:35:14,633 INFO cfg.RCNN.HARD_BG_RATIO: 0.8 2021-09-22 04:35:14,633 INFO cfg.RCNN.SCORE_THRESH: 0.3 2021-09-22 04:35:14,634 INFO cfg.RCNN.NMS_THRESH: 0.1 2021-09-22 04:35:14,634 INFO cfg.TRAIN = edict() 2021-09-22 04:35:14,634 INFO cfg.TRAIN.SPLIT: train 2021-09-22 04:35:14,634 INFO cfg.TRAIN.VAL_SPLIT: smallval 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR: 0.002 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR_CLIP: 1e-05 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR_DECAY: 0.5 2021-09-22 04:35:14,634 INFO cfg.TRAIN.DECAY_STEP_LIST: [100, 150, 180, 200] 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR_WARMUP: True 2021-09-22 04:35:14,634 INFO cfg.TRAIN.WARMUP_MIN: 0.0002 2021-09-22 04:35:14,634 INFO cfg.TRAIN.WARMUP_EPOCH: 1 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BN_MOMENTUM: 0.1 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BN_DECAY: 0.5 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BNM_CLIP: 0.01 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BN_DECAY_STEP_LIST: [1000] 2021-09-22 04:35:14,634 INFO cfg.TRAIN.OPTIMIZER: adam_onecycle 2021-09-22 04:35:14,634 INFO cfg.TRAIN.WEIGHT_DECAY: 0.001 2021-09-22 04:35:14,634 INFO cfg.TRAIN.MOMENTUM: 0.9 2021-09-22 04:35:14,634 INFO cfg.TRAIN.MOMS: [0.95, 0.85] 2021-09-22 04:35:14,634 INFO cfg.TRAIN.DIV_FACTOR: 10.0 2021-09-22 04:35:14,634 INFO cfg.TRAIN.PCT_START: 0.4 2021-09-22 04:35:14,634 INFO cfg.TRAIN.GRAD_NORM_CLIP: 1.0 2021-09-22 04:35:14,634 INFO cfg.TRAIN.RPN_PRE_NMS_TOP_N: 9000 2021-09-22 04:35:14,634 INFO cfg.TRAIN.RPN_POST_NMS_TOP_N: 512 2021-09-22 04:35:14,634 INFO cfg.TRAIN.RPN_NMS_THRESH: 0.85 2021-09-22 04:35:14,635 INFO cfg.TRAIN.RPN_DISTANCE_BASED_PROPOSE: True 2021-09-22 04:35:14,635 INFO cfg.TEST = edict() 2021-09-22 04:35:14,635 INFO cfg.TEST.SPLIT: val 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_PRE_NMS_TOP_N: 9000 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_POST_NMS_TOP_N: 100 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_NMS_THRESH: 0.8 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_DISTANCE_BASED_PROPOSE: True 2021-09-22 04:35:14,636 INFO Load testing samples from ../data/KITTI/object/training 2021-09-22 04:35:14,636 INFO Done: total test samples 3769 2021-09-22 04:35:16,962 INFO ==> Loading from checkpoint 'PointRCNN.pth' 2021-09-22 04:35:16,989 INFO ==> Done 2021-09-22 04:35:16,990 INFO ---- EPOCH no_number JOINT EVALUATION ---- 2021-09-22 04:35:16,990 INFO ==> Output file: ../output/rcnn/default/eval/epoch_no_number/val eval: 13%|▋ | 488/3769 [00:55<05:59, 9.13it/s, mode=EVAL, recall=1170/1769]Traceback (most recent call last): File "eval_rcnn.py", line 902, in eval_single_ckpt(root_result_dir) File "eval_rcnn.py", line 765, in eval_single_ckpt eval_one_epoch(model, test_loader, epoch_id, root_result_dir, logger) File "eval_rcnn.py", line 692, in eval_one_epoch ret_dict = eval_one_epoch_joint(model, dataloader, epoch_id, result_dir, logger) File "eval_rcnn.py", line 486, in eval_one_epoch_joint for data in dataloader: File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 582, in next return self._process_next_batch(batch) File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 608, in _process_next_batch raise batch.exc_type(batch.exc_msg) AssertionError: Traceback (most recent call last): File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/_utils/worker.py", line 99, in _worker_loop samples = collate_fn([dataset[i] for i in batch_indices]) File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/_utils/worker.py", line 99, in samples = collate_fn([dataset[i] for i in batch_indices]) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_rcnn_dataset.py", line 234, in getitem return self.get_rpn_sample(index) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_rcnn_dataset.py", line 249, in get_rpn_sample calib = self.get_calib(sample_id) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_rcnn_dataset.py", line 133, in get_calib return super().get_calib(idx % 10000) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_dataset.py", line 47, in get_calib assert os.path.exists(calib_file) AssertionError eval: 13%|▋ | 488/3769 [00:55<06:12, 8.80it/s, mode=EVAL, recall=1170/1769]
Hello? I found that I have the same problem with you! My process stopped when 448 13% too! Then I think that probably is not because the lack or wrong file path of dataset, what other problem??
嗯...我的问题就是因为数据集排列的路径不对。你可以仔细检查一下是否已经下载了planes数据集并且数据集排列的顺序没有错误。我并没有遇到其他问题
好的感谢!我还有一个问题,那个完整数据集太大了我只移动了1~1000进去,可能是因为这个原因造成的吗。planes我已经下载了。
检查一下数据集的路径是否正确,数据集的排布是否和作者的一样。还有那个plane这个文件也得下。 … ---Original--- From: @.**> Date: Wed, Sep 22, 2021 15:43 PM To: @.**>; Cc: @.**@.**>; Subject: Re: [sshaoshuai/PointRCNN] assert os.path.exists(calib_file) when run eval_rcnn.py (#87) thanks,I find the problem which is because of the wrong of the file path 老哥,能不能展开说说,啥错误的文件路径?我也遇到了和你一样的错误 — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.
我之前运行程序是会跑一段的,在448的时候突然停下来,在看到你的回复后,我仔细检查了两遍路径并下载了plane的文件,可是运行起来还是卡在了448的点,不知道老哥你是否能给些想法和建议: python eval_rcnn.py --cfg_file cfgs/default.yaml --ckpt PointRCNN.pth --batch_size 1 --eval_mode rcnn --set RPN.LOC_XZ_FINE False /usr/local/lib/python3.6/dist-packages/numba/cuda/envvars.py:17: NumbaWarning: Environment variables with the 'NUMBAPRO' prefix are deprecated and consequently ignored, found use of NUMBAPRO_NVVM=/usr/local/cuda-10.0/nvvm/lib64/libnvvm.so. For more information about alternatives visit: ('https://numba.pydata.org/numba-doc/latest/cuda/overview.html', '#cudatoolkit-lookup') warnings.warn(errors.NumbaWarning(msg)) /usr/local/lib/python3.6/dist-packages/numba/cuda/envvars.py:17: NumbaWarning: Environment variables with the 'NUMBAPRO' prefix are deprecated and consequently ignored, found use of NUMBAPRO_LIBDEVICE=/usr/local/cuda-10.0/nvvm/libdevice/. For more information about alternatives visit: ('https://numba.pydata.org/numba-doc/latest/cuda/overview.html', '#cudatoolkit-lookup') warnings.warn(errors.NumbaWarning(msg)) /home/wg/PointRCNN/tools/../lib/config.py:187: YAMLLoadWarning: calling yaml.load() without Loader=... is deprecated, as the default Loader is unsafe. Please read https://msg.pyyaml.org/load for full details. yaml_cfg = edict(yaml.load(f)) 2021-09-22 04:35:14,630 INFO **Start logging** 2021-09-22 04:35:14,630 INFO cfg_file cfgs/default.yaml 2021-09-22 04:35:14,630 INFO eval_mode rcnn 2021-09-22 04:35:14,630 INFO eval_all False 2021-09-22 04:35:14,630 INFO test False 2021-09-22 04:35:14,630 INFO ckpt PointRCNN.pth 2021-09-22 04:35:14,630 INFO rpn_ckpt None 2021-09-22 04:35:14,630 INFO rcnn_ckpt None 2021-09-22 04:35:14,630 INFO batch_size 1 2021-09-22 04:35:14,630 INFO workers 4 2021-09-22 04:35:14,630 INFO extra_tag default 2021-09-22 04:35:14,630 INFO output_dir None 2021-09-22 04:35:14,630 INFO ckpt_dir None 2021-09-22 04:35:14,630 INFO save_result False 2021-09-22 04:35:14,630 INFO save_rpn_feature False 2021-09-22 04:35:14,630 INFO random_select True 2021-09-22 04:35:14,630 INFO start_epoch 0 2021-09-22 04:35:14,630 INFO rcnn_eval_roi_dir None 2021-09-22 04:35:14,630 INFO rcnn_eval_feature_dir None 2021-09-22 04:35:14,630 INFO set_cfgs ['RPN.LOC_XZ_FINE', 'False'] 2021-09-22 04:35:14,630 INFO cfg.TAG: default 2021-09-22 04:35:14,630 INFO cfg.CLASSES: Car 2021-09-22 04:35:14,630 INFO cfg.INCLUDE_SIMILAR_TYPE: True 2021-09-22 04:35:14,630 INFO cfg.AUG_DATA: True 2021-09-22 04:35:14,630 INFO cfg.AUG_METHOD_LIST: ['rotation', 'scaling', 'flip'] 2021-09-22 04:35:14,630 INFO cfg.AUG_METHOD_PROB: [1.0, 1.0, 0.5] 2021-09-22 04:35:14,630 INFO cfg.AUG_ROT_RANGE: 18 2021-09-22 04:35:14,630 INFO cfg.GT_AUG_ENABLED: True 2021-09-22 04:35:14,631 INFO cfg.GT_EXTRA_NUM: 15 2021-09-22 04:35:14,631 INFO cfg.GT_AUG_RAND_NUM: True 2021-09-22 04:35:14,631 INFO cfg.GT_AUG_APPLY_PROB: 1.0 2021-09-22 04:35:14,631 INFO cfg.GT_AUG_HARD_RATIO: 0.6 2021-09-22 04:35:14,631 INFO cfg.PC_REDUCE_BY_RANGE: True 2021-09-22 04:35:14,631 INFO cfg.PC_AREA_SCOPE: [[-40. 40. ] [ -1. 3. ] [ 0. 70.4]] 2021-09-22 04:35:14,631 INFO cfg.CLS_MEAN_SIZE: [[1.5256319 1.6285675 3.8831165]] 2021-09-22 04:35:14,631 INFO cfg.RPN = edict() 2021-09-22 04:35:14,631 INFO cfg.RPN.ENABLED: True 2021-09-22 04:35:14,631 INFO cfg.RPN.FIXED: True 2021-09-22 04:35:14,631 INFO cfg.RPN.USE_INTENSITY: False 2021-09-22 04:35:14,631 INFO cfg.RPN.LOC_XZ_FINE: False 2021-09-22 04:35:14,631 INFO cfg.RPN.LOC_SCOPE: 3.0 2021-09-22 04:35:14,631 INFO cfg.RPN.LOC_BIN_SIZE: 0.5 2021-09-22 04:35:14,631 INFO cfg.RPN.NUM_HEAD_BIN: 12 2021-09-22 04:35:14,631 INFO cfg.RPN.BACKBONE: pointnet2_msg 2021-09-22 04:35:14,631 INFO cfg.RPN.USE_BN: True 2021-09-22 04:35:14,631 INFO cfg.RPN.NUM_POINTS: 16384 2021-09-22 04:35:14,631 INFO cfg.RPN.SA_CONFIG = edict() 2021-09-22 04:35:14,631 INFO cfg.RPN.SA_CONFIG.NPOINTS: [4096, 1024, 256, 64] 2021-09-22 04:35:14,631 INFO cfg.RPN.SA_CONFIG.RADIUS: [[0.1, 0.5], [0.5, 1.0], [1.0, 2.0], [2.0, 4.0]] 2021-09-22 04:35:14,632 INFO cfg.RPN.SA_CONFIG.NSAMPLE: [[16, 32], [16, 32], [16, 32], [16, 32]] 2021-09-22 04:35:14,632 INFO cfg.RPN.SA_CONFIG.MLPS: [[[16, 16, 32], [32, 32, 64]], [[64, 64, 128], [64, 96, 128]], [[128, 196, 256], [128, 196, 256]], [[256, 256, 512], [256, 384, 512]]] 2021-09-22 04:35:14,632 INFO cfg.RPN.FP_MLPS: [[128, 128], [256, 256], [512, 512], [512, 512]] 2021-09-22 04:35:14,632 INFO cfg.RPN.CLS_FC: [128] 2021-09-22 04:35:14,632 INFO cfg.RPN.REG_FC: [128] 2021-09-22 04:35:14,632 INFO cfg.RPN.DP_RATIO: 0.5 2021-09-22 04:35:14,632 INFO cfg.RPN.LOSS_CLS: SigmoidFocalLoss 2021-09-22 04:35:14,632 INFO cfg.RPN.FG_WEIGHT: 15 2021-09-22 04:35:14,632 INFO cfg.RPN.FOCAL_ALPHA: [0.25, 0.75] 2021-09-22 04:35:14,632 INFO cfg.RPN.FOCAL_GAMMA: 2.0 2021-09-22 04:35:14,632 INFO cfg.RPN.REG_LOSS_WEIGHT: [1.0, 1.0, 1.0, 1.0] 2021-09-22 04:35:14,632 INFO cfg.RPN.LOSS_WEIGHT: [1.0, 1.0] 2021-09-22 04:35:14,632 INFO cfg.RPN.NMS_TYPE: normal 2021-09-22 04:35:14,632 INFO cfg.RPN.SCORE_THRESH: 0.3 2021-09-22 04:35:14,632 INFO cfg.RCNN = edict() 2021-09-22 04:35:14,632 INFO cfg.RCNN.ENABLED: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_RPN_FEATURES: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_MASK: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.MASK_TYPE: seg 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_INTENSITY: False 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_DEPTH: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_SEG_SCORE: False 2021-09-22 04:35:14,632 INFO cfg.RCNN.ROI_SAMPLE_JIT: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.ROI_FG_AUG_TIMES: 10 2021-09-22 04:35:14,632 INFO cfg.RCNN.REG_AUG_METHOD: multiple 2021-09-22 04:35:14,632 INFO cfg.RCNN.POOL_EXTRA_WIDTH: 1.0 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_SCOPE: 1.5 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_BIN_SIZE: 0.5 2021-09-22 04:35:14,632 INFO cfg.RCNN.NUM_HEAD_BIN: 9 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_Y_BY_BIN: False 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_Y_SCOPE: 0.5 2021-09-22 04:35:14,633 INFO cfg.RCNN.LOC_Y_BIN_SIZE: 0.25 2021-09-22 04:35:14,633 INFO cfg.RCNN.SIZE_RES_ON_ROI: False 2021-09-22 04:35:14,633 INFO cfg.RCNN.USE_BN: False 2021-09-22 04:35:14,633 INFO cfg.RCNN.DP_RATIO: 0.0 2021-09-22 04:35:14,633 INFO cfg.RCNN.BACKBONE: pointnet 2021-09-22 04:35:14,633 INFO cfg.RCNN.XYZ_UP_LAYER: [128, 128] 2021-09-22 04:35:14,633 INFO cfg.RCNN.NUM_POINTS: 512 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG = edict() 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.NPOINTS: [128, 32, -1] 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.RADIUS: [0.2, 0.4, 100] 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.NSAMPLE: [64, 64, 64] 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.MLPS: [[128, 128, 128], [128, 128, 256], [256, 256, 512]] 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_FC: [256, 256] 2021-09-22 04:35:14,633 INFO cfg.RCNN.REG_FC: [256, 256] 2021-09-22 04:35:14,633 INFO cfg.RCNN.LOSS_CLS: BinaryCrossEntropy 2021-09-22 04:35:14,633 INFO cfg.RCNN.FOCAL_ALPHA: [0.25, 0.75] 2021-09-22 04:35:14,633 INFO cfg.RCNN.FOCAL_GAMMA: 2.0 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_WEIGHT: [1. 1. 1.] 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_FG_THRESH: 0.6 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_BG_THRESH: 0.45 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_BG_THRESH_LO: 0.05 2021-09-22 04:35:14,633 INFO cfg.RCNN.REG_FG_THRESH: 0.55 2021-09-22 04:35:14,633 INFO cfg.RCNN.FG_RATIO: 0.5 2021-09-22 04:35:14,633 INFO cfg.RCNN.ROI_PER_IMAGE: 64 2021-09-22 04:35:14,633 INFO cfg.RCNN.HARD_BG_RATIO: 0.8 2021-09-22 04:35:14,633 INFO cfg.RCNN.SCORE_THRESH: 0.3 2021-09-22 04:35:14,634 INFO cfg.RCNN.NMS_THRESH: 0.1 2021-09-22 04:35:14,634 INFO cfg.TRAIN = edict() 2021-09-22 04:35:14,634 INFO cfg.TRAIN.SPLIT: train 2021-09-22 04:35:14,634 INFO cfg.TRAIN.VAL_SPLIT: smallval 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR: 0.002 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR_CLIP: 1e-05 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR_DECAY: 0.5 2021-09-22 04:35:14,634 INFO cfg.TRAIN.DECAY_STEP_LIST: [100, 150, 180, 200] 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR_WARMUP: True 2021-09-22 04:35:14,634 INFO cfg.TRAIN.WARMUP_MIN: 0.0002 2021-09-22 04:35:14,634 INFO cfg.TRAIN.WARMUP_EPOCH: 1 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BN_MOMENTUM: 0.1 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BN_DECAY: 0.5 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BNM_CLIP: 0.01 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BN_DECAY_STEP_LIST: [1000] 2021-09-22 04:35:14,634 INFO cfg.TRAIN.OPTIMIZER: adam_onecycle 2021-09-22 04:35:14,634 INFO cfg.TRAIN.WEIGHT_DECAY: 0.001 2021-09-22 04:35:14,634 INFO cfg.TRAIN.MOMENTUM: 0.9 2021-09-22 04:35:14,634 INFO cfg.TRAIN.MOMS: [0.95, 0.85] 2021-09-22 04:35:14,634 INFO cfg.TRAIN.DIV_FACTOR: 10.0 2021-09-22 04:35:14,634 INFO cfg.TRAIN.PCT_START: 0.4 2021-09-22 04:35:14,634 INFO cfg.TRAIN.GRAD_NORM_CLIP: 1.0 2021-09-22 04:35:14,634 INFO cfg.TRAIN.RPN_PRE_NMS_TOP_N: 9000 2021-09-22 04:35:14,634 INFO cfg.TRAIN.RPN_POST_NMS_TOP_N: 512 2021-09-22 04:35:14,634 INFO cfg.TRAIN.RPN_NMS_THRESH: 0.85 2021-09-22 04:35:14,635 INFO cfg.TRAIN.RPN_DISTANCE_BASED_PROPOSE: True 2021-09-22 04:35:14,635 INFO cfg.TEST = edict() 2021-09-22 04:35:14,635 INFO cfg.TEST.SPLIT: val 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_PRE_NMS_TOP_N: 9000 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_POST_NMS_TOP_N: 100 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_NMS_THRESH: 0.8 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_DISTANCE_BASED_PROPOSE: True 2021-09-22 04:35:14,636 INFO Load testing samples from ../data/KITTI/object/training 2021-09-22 04:35:14,636 INFO Done: total test samples 3769 2021-09-22 04:35:16,962 INFO ==> Loading from checkpoint 'PointRCNN.pth' 2021-09-22 04:35:16,989 INFO ==> Done 2021-09-22 04:35:16,990 INFO ---- EPOCH no_number JOINT EVALUATION ---- 2021-09-22 04:35:16,990 INFO ==> Output file: ../output/rcnn/default/eval/epoch_no_number/val eval: 13%|▋ | 488/3769 [00:55<05:59, 9.13it/s, mode=EVAL, recall=1170/1769]Traceback (most recent call last): File "eval_rcnn.py", line 902, in eval_single_ckpt(root_result_dir) File "eval_rcnn.py", line 765, in eval_single_ckpt eval_one_epoch(model, test_loader, epoch_id, root_result_dir, logger) File "eval_rcnn.py", line 692, in eval_one_epoch ret_dict = eval_one_epoch_joint(model, dataloader, epoch_id, result_dir, logger) File "eval_rcnn.py", line 486, in eval_one_epoch_joint for data in dataloader: File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 582, in next return self._process_next_batch(batch) File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 608, in _process_next_batch raise batch.exc_type(batch.exc_msg) AssertionError: Traceback (most recent call last): File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/_utils/worker.py", line 99, in _worker_loop samples = collate_fn([dataset[i] for i in batch_indices]) File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/_utils/worker.py", line 99, in samples = collate_fn([dataset[i] for i in batch_indices]) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_rcnn_dataset.py", line 234, in getitem return self.get_rpn_sample(index) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_rcnn_dataset.py", line 249, in get_rpn_sample calib = self.get_calib(sample_id) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_rcnn_dataset.py", line 133, in get_calib return super().get_calib(idx % 10000) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_dataset.py", line 47, in get_calib assert os.path.exists(calib_file) AssertionError eval: 13%|▋ | 488/3769 [00:55<06:12, 8.80it/s, mode=EVAL, recall=1170/1769]
Hello? I found that I have the same problem with you! My process stopped when 448 13% too! Then I think that probably is not because the lack or wrong file path of dataset, what other problem??
嗯...我的问题就是因为数据集排列的路径不对。你可以仔细检查一下是否已经下载了planes数据集并且数据集排列的顺序没有错误。我并没有遇到其他问题
好的感谢!我还有一个问题,那个完整数据集太大了我只移动了1~1000进去,可能是因为这个原因造成的吗。planes我已经下载了。
一般来说对应的那几个部分都只放1~1000份数据进去应该是没有问题的。pointRCNN我没有试过把其中一部分放进去会是怎么样的,我就是单纯跑了一遍。如果你现阶段的目的单纯是复现的话,建议还是完整的来。
检查一下数据集的路径是否正确,数据集的排布是否和作者的一样。还有那个plane这个文件也得下。 … ---Original--- From: @.**> Date: Wed, Sep 22, 2021 15:43 PM To: @.**>; Cc: @.**@.**>; Subject: Re: [sshaoshuai/PointRCNN] assert os.path.exists(calib_file) when run eval_rcnn.py (#87) thanks,I find the problem which is because of the wrong of the file path 老哥,能不能展开说说,啥错误的文件路径?我也遇到了和你一样的错误 — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.
我之前运行程序是会跑一段的,在448的时候突然停下来,在看到你的回复后,我仔细检查了两遍路径并下载了plane的文件,可是运行起来还是卡在了448的点,不知道老哥你是否能给些想法和建议: python eval_rcnn.py --cfg_file cfgs/default.yaml --ckpt PointRCNN.pth --batch_size 1 --eval_mode rcnn --set RPN.LOC_XZ_FINE False /usr/local/lib/python3.6/dist-packages/numba/cuda/envvars.py:17: NumbaWarning: Environment variables with the 'NUMBAPRO' prefix are deprecated and consequently ignored, found use of NUMBAPRO_NVVM=/usr/local/cuda-10.0/nvvm/lib64/libnvvm.so. For more information about alternatives visit: ('https://numba.pydata.org/numba-doc/latest/cuda/overview.html', '#cudatoolkit-lookup') warnings.warn(errors.NumbaWarning(msg)) /usr/local/lib/python3.6/dist-packages/numba/cuda/envvars.py:17: NumbaWarning: Environment variables with the 'NUMBAPRO' prefix are deprecated and consequently ignored, found use of NUMBAPRO_LIBDEVICE=/usr/local/cuda-10.0/nvvm/libdevice/. For more information about alternatives visit: ('https://numba.pydata.org/numba-doc/latest/cuda/overview.html', '#cudatoolkit-lookup') warnings.warn(errors.NumbaWarning(msg)) /home/wg/PointRCNN/tools/../lib/config.py:187: YAMLLoadWarning: calling yaml.load() without Loader=... is deprecated, as the default Loader is unsafe. Please read https://msg.pyyaml.org/load for full details. yaml_cfg = edict(yaml.load(f)) 2021-09-22 04:35:14,630 INFO **Start logging** 2021-09-22 04:35:14,630 INFO cfg_file cfgs/default.yaml 2021-09-22 04:35:14,630 INFO eval_mode rcnn 2021-09-22 04:35:14,630 INFO eval_all False 2021-09-22 04:35:14,630 INFO test False 2021-09-22 04:35:14,630 INFO ckpt PointRCNN.pth 2021-09-22 04:35:14,630 INFO rpn_ckpt None 2021-09-22 04:35:14,630 INFO rcnn_ckpt None 2021-09-22 04:35:14,630 INFO batch_size 1 2021-09-22 04:35:14,630 INFO workers 4 2021-09-22 04:35:14,630 INFO extra_tag default 2021-09-22 04:35:14,630 INFO output_dir None 2021-09-22 04:35:14,630 INFO ckpt_dir None 2021-09-22 04:35:14,630 INFO save_result False 2021-09-22 04:35:14,630 INFO save_rpn_feature False 2021-09-22 04:35:14,630 INFO random_select True 2021-09-22 04:35:14,630 INFO start_epoch 0 2021-09-22 04:35:14,630 INFO rcnn_eval_roi_dir None 2021-09-22 04:35:14,630 INFO rcnn_eval_feature_dir None 2021-09-22 04:35:14,630 INFO set_cfgs ['RPN.LOC_XZ_FINE', 'False'] 2021-09-22 04:35:14,630 INFO cfg.TAG: default 2021-09-22 04:35:14,630 INFO cfg.CLASSES: Car 2021-09-22 04:35:14,630 INFO cfg.INCLUDE_SIMILAR_TYPE: True 2021-09-22 04:35:14,630 INFO cfg.AUG_DATA: True 2021-09-22 04:35:14,630 INFO cfg.AUG_METHOD_LIST: ['rotation', 'scaling', 'flip'] 2021-09-22 04:35:14,630 INFO cfg.AUG_METHOD_PROB: [1.0, 1.0, 0.5] 2021-09-22 04:35:14,630 INFO cfg.AUG_ROT_RANGE: 18 2021-09-22 04:35:14,630 INFO cfg.GT_AUG_ENABLED: True 2021-09-22 04:35:14,631 INFO cfg.GT_EXTRA_NUM: 15 2021-09-22 04:35:14,631 INFO cfg.GT_AUG_RAND_NUM: True 2021-09-22 04:35:14,631 INFO cfg.GT_AUG_APPLY_PROB: 1.0 2021-09-22 04:35:14,631 INFO cfg.GT_AUG_HARD_RATIO: 0.6 2021-09-22 04:35:14,631 INFO cfg.PC_REDUCE_BY_RANGE: True 2021-09-22 04:35:14,631 INFO cfg.PC_AREA_SCOPE: [[-40. 40. ] [ -1. 3. ] [ 0. 70.4]] 2021-09-22 04:35:14,631 INFO cfg.CLS_MEAN_SIZE: [[1.5256319 1.6285675 3.8831165]] 2021-09-22 04:35:14,631 INFO cfg.RPN = edict() 2021-09-22 04:35:14,631 INFO cfg.RPN.ENABLED: True 2021-09-22 04:35:14,631 INFO cfg.RPN.FIXED: True 2021-09-22 04:35:14,631 INFO cfg.RPN.USE_INTENSITY: False 2021-09-22 04:35:14,631 INFO cfg.RPN.LOC_XZ_FINE: False 2021-09-22 04:35:14,631 INFO cfg.RPN.LOC_SCOPE: 3.0 2021-09-22 04:35:14,631 INFO cfg.RPN.LOC_BIN_SIZE: 0.5 2021-09-22 04:35:14,631 INFO cfg.RPN.NUM_HEAD_BIN: 12 2021-09-22 04:35:14,631 INFO cfg.RPN.BACKBONE: pointnet2_msg 2021-09-22 04:35:14,631 INFO cfg.RPN.USE_BN: True 2021-09-22 04:35:14,631 INFO cfg.RPN.NUM_POINTS: 16384 2021-09-22 04:35:14,631 INFO cfg.RPN.SA_CONFIG = edict() 2021-09-22 04:35:14,631 INFO cfg.RPN.SA_CONFIG.NPOINTS: [4096, 1024, 256, 64] 2021-09-22 04:35:14,631 INFO cfg.RPN.SA_CONFIG.RADIUS: [[0.1, 0.5], [0.5, 1.0], [1.0, 2.0], [2.0, 4.0]] 2021-09-22 04:35:14,632 INFO cfg.RPN.SA_CONFIG.NSAMPLE: [[16, 32], [16, 32], [16, 32], [16, 32]] 2021-09-22 04:35:14,632 INFO cfg.RPN.SA_CONFIG.MLPS: [[[16, 16, 32], [32, 32, 64]], [[64, 64, 128], [64, 96, 128]], [[128, 196, 256], [128, 196, 256]], [[256, 256, 512], [256, 384, 512]]] 2021-09-22 04:35:14,632 INFO cfg.RPN.FP_MLPS: [[128, 128], [256, 256], [512, 512], [512, 512]] 2021-09-22 04:35:14,632 INFO cfg.RPN.CLS_FC: [128] 2021-09-22 04:35:14,632 INFO cfg.RPN.REG_FC: [128] 2021-09-22 04:35:14,632 INFO cfg.RPN.DP_RATIO: 0.5 2021-09-22 04:35:14,632 INFO cfg.RPN.LOSS_CLS: SigmoidFocalLoss 2021-09-22 04:35:14,632 INFO cfg.RPN.FG_WEIGHT: 15 2021-09-22 04:35:14,632 INFO cfg.RPN.FOCAL_ALPHA: [0.25, 0.75] 2021-09-22 04:35:14,632 INFO cfg.RPN.FOCAL_GAMMA: 2.0 2021-09-22 04:35:14,632 INFO cfg.RPN.REG_LOSS_WEIGHT: [1.0, 1.0, 1.0, 1.0] 2021-09-22 04:35:14,632 INFO cfg.RPN.LOSS_WEIGHT: [1.0, 1.0] 2021-09-22 04:35:14,632 INFO cfg.RPN.NMS_TYPE: normal 2021-09-22 04:35:14,632 INFO cfg.RPN.SCORE_THRESH: 0.3 2021-09-22 04:35:14,632 INFO cfg.RCNN = edict() 2021-09-22 04:35:14,632 INFO cfg.RCNN.ENABLED: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_RPN_FEATURES: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_MASK: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.MASK_TYPE: seg 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_INTENSITY: False 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_DEPTH: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_SEG_SCORE: False 2021-09-22 04:35:14,632 INFO cfg.RCNN.ROI_SAMPLE_JIT: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.ROI_FG_AUG_TIMES: 10 2021-09-22 04:35:14,632 INFO cfg.RCNN.REG_AUG_METHOD: multiple 2021-09-22 04:35:14,632 INFO cfg.RCNN.POOL_EXTRA_WIDTH: 1.0 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_SCOPE: 1.5 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_BIN_SIZE: 0.5 2021-09-22 04:35:14,632 INFO cfg.RCNN.NUM_HEAD_BIN: 9 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_Y_BY_BIN: False 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_Y_SCOPE: 0.5 2021-09-22 04:35:14,633 INFO cfg.RCNN.LOC_Y_BIN_SIZE: 0.25 2021-09-22 04:35:14,633 INFO cfg.RCNN.SIZE_RES_ON_ROI: False 2021-09-22 04:35:14,633 INFO cfg.RCNN.USE_BN: False 2021-09-22 04:35:14,633 INFO cfg.RCNN.DP_RATIO: 0.0 2021-09-22 04:35:14,633 INFO cfg.RCNN.BACKBONE: pointnet 2021-09-22 04:35:14,633 INFO cfg.RCNN.XYZ_UP_LAYER: [128, 128] 2021-09-22 04:35:14,633 INFO cfg.RCNN.NUM_POINTS: 512 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG = edict() 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.NPOINTS: [128, 32, -1] 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.RADIUS: [0.2, 0.4, 100] 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.NSAMPLE: [64, 64, 64] 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.MLPS: [[128, 128, 128], [128, 128, 256], [256, 256, 512]] 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_FC: [256, 256] 2021-09-22 04:35:14,633 INFO cfg.RCNN.REG_FC: [256, 256] 2021-09-22 04:35:14,633 INFO cfg.RCNN.LOSS_CLS: BinaryCrossEntropy 2021-09-22 04:35:14,633 INFO cfg.RCNN.FOCAL_ALPHA: [0.25, 0.75] 2021-09-22 04:35:14,633 INFO cfg.RCNN.FOCAL_GAMMA: 2.0 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_WEIGHT: [1. 1. 1.] 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_FG_THRESH: 0.6 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_BG_THRESH: 0.45 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_BG_THRESH_LO: 0.05 2021-09-22 04:35:14,633 INFO cfg.RCNN.REG_FG_THRESH: 0.55 2021-09-22 04:35:14,633 INFO cfg.RCNN.FG_RATIO: 0.5 2021-09-22 04:35:14,633 INFO cfg.RCNN.ROI_PER_IMAGE: 64 2021-09-22 04:35:14,633 INFO cfg.RCNN.HARD_BG_RATIO: 0.8 2021-09-22 04:35:14,633 INFO cfg.RCNN.SCORE_THRESH: 0.3 2021-09-22 04:35:14,634 INFO cfg.RCNN.NMS_THRESH: 0.1 2021-09-22 04:35:14,634 INFO cfg.TRAIN = edict() 2021-09-22 04:35:14,634 INFO cfg.TRAIN.SPLIT: train 2021-09-22 04:35:14,634 INFO cfg.TRAIN.VAL_SPLIT: smallval 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR: 0.002 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR_CLIP: 1e-05 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR_DECAY: 0.5 2021-09-22 04:35:14,634 INFO cfg.TRAIN.DECAY_STEP_LIST: [100, 150, 180, 200] 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR_WARMUP: True 2021-09-22 04:35:14,634 INFO cfg.TRAIN.WARMUP_MIN: 0.0002 2021-09-22 04:35:14,634 INFO cfg.TRAIN.WARMUP_EPOCH: 1 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BN_MOMENTUM: 0.1 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BN_DECAY: 0.5 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BNM_CLIP: 0.01 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BN_DECAY_STEP_LIST: [1000] 2021-09-22 04:35:14,634 INFO cfg.TRAIN.OPTIMIZER: adam_onecycle 2021-09-22 04:35:14,634 INFO cfg.TRAIN.WEIGHT_DECAY: 0.001 2021-09-22 04:35:14,634 INFO cfg.TRAIN.MOMENTUM: 0.9 2021-09-22 04:35:14,634 INFO cfg.TRAIN.MOMS: [0.95, 0.85] 2021-09-22 04:35:14,634 INFO cfg.TRAIN.DIV_FACTOR: 10.0 2021-09-22 04:35:14,634 INFO cfg.TRAIN.PCT_START: 0.4 2021-09-22 04:35:14,634 INFO cfg.TRAIN.GRAD_NORM_CLIP: 1.0 2021-09-22 04:35:14,634 INFO cfg.TRAIN.RPN_PRE_NMS_TOP_N: 9000 2021-09-22 04:35:14,634 INFO cfg.TRAIN.RPN_POST_NMS_TOP_N: 512 2021-09-22 04:35:14,634 INFO cfg.TRAIN.RPN_NMS_THRESH: 0.85 2021-09-22 04:35:14,635 INFO cfg.TRAIN.RPN_DISTANCE_BASED_PROPOSE: True 2021-09-22 04:35:14,635 INFO cfg.TEST = edict() 2021-09-22 04:35:14,635 INFO cfg.TEST.SPLIT: val 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_PRE_NMS_TOP_N: 9000 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_POST_NMS_TOP_N: 100 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_NMS_THRESH: 0.8 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_DISTANCE_BASED_PROPOSE: True 2021-09-22 04:35:14,636 INFO Load testing samples from ../data/KITTI/object/training 2021-09-22 04:35:14,636 INFO Done: total test samples 3769 2021-09-22 04:35:16,962 INFO ==> Loading from checkpoint 'PointRCNN.pth' 2021-09-22 04:35:16,989 INFO ==> Done 2021-09-22 04:35:16,990 INFO ---- EPOCH no_number JOINT EVALUATION ---- 2021-09-22 04:35:16,990 INFO ==> Output file: ../output/rcnn/default/eval/epoch_no_number/val eval: 13%|▋ | 488/3769 [00:55<05:59, 9.13it/s, mode=EVAL, recall=1170/1769]Traceback (most recent call last): File "eval_rcnn.py", line 902, in eval_single_ckpt(root_result_dir) File "eval_rcnn.py", line 765, in eval_single_ckpt eval_one_epoch(model, test_loader, epoch_id, root_result_dir, logger) File "eval_rcnn.py", line 692, in eval_one_epoch ret_dict = eval_one_epoch_joint(model, dataloader, epoch_id, result_dir, logger) File "eval_rcnn.py", line 486, in eval_one_epoch_joint for data in dataloader: File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 582, in next return self._process_next_batch(batch) File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 608, in _process_next_batch raise batch.exc_type(batch.exc_msg) AssertionError: Traceback (most recent call last): File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/_utils/worker.py", line 99, in _worker_loop samples = collate_fn([dataset[i] for i in batch_indices]) File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/_utils/worker.py", line 99, in samples = collate_fn([dataset[i] for i in batch_indices]) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_rcnn_dataset.py", line 234, in getitem return self.get_rpn_sample(index) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_rcnn_dataset.py", line 249, in get_rpn_sample calib = self.get_calib(sample_id) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_rcnn_dataset.py", line 133, in get_calib return super().get_calib(idx % 10000) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_dataset.py", line 47, in get_calib assert os.path.exists(calib_file) AssertionError eval: 13%|▋ | 488/3769 [00:55<06:12, 8.80it/s, mode=EVAL, recall=1170/1769]
Hello? I found that I have the same problem with you! My process stopped when 448 13% too! Then I think that probably is not because the lack or wrong file path of dataset, what other problem??
嗯...我的问题就是因为数据集排列的路径不对。你可以仔细检查一下是否已经下载了planes数据集并且数据集排列的顺序没有错误。我并没有遇到其他问题
好的感谢!我还有一个问题,那个完整数据集太大了我只移动了1~1000进去,可能是因为这个原因造成的吗。planes我已经下载了。
一般来说对应的那几个部分都只放1~1000份数据进去应该是没有问题的。pointRCNN我没有试过把其中一部分放进去会是怎么样的,我就是单纯跑了一遍。如果你现阶段的目的单纯是复现的话,建议还是完整的来。
感谢老哥,已经跑通了
检查一下数据集的路径是否正确,数据集的排布是否和作者的一样。还有那个plane这个文件也得下。 … ---Original--- From: @.**> Date: Wed, Sep 22, 2021 15:43 PM To: @.**>; Cc: @.**@.**>; Subject: Re: [sshaoshuai/PointRCNN] assert os.path.exists(calib_file) when run eval_rcnn.py (#87) thanks,I find the problem which is because of the wrong of the file path 老哥,能不能展开说说,啥错误的文件路径?我也遇到了和你一样的错误 — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.
我之前运行程序是会跑一段的,在448的时候突然停下来,在看到你的回复后,我仔细检查了两遍路径并下载了plane的文件,可是运行起来还是卡在了448的点,不知道老哥你是否能给些想法和建议: python eval_rcnn.py --cfg_file cfgs/default.yaml --ckpt PointRCNN.pth --batch_size 1 --eval_mode rcnn --set RPN.LOC_XZ_FINE False /usr/local/lib/python3.6/dist-packages/numba/cuda/envvars.py:17: NumbaWarning: Environment variables with the 'NUMBAPRO' prefix are deprecated and consequently ignored, found use of NUMBAPRO_NVVM=/usr/local/cuda-10.0/nvvm/lib64/libnvvm.so. For more information about alternatives visit: ('https://numba.pydata.org/numba-doc/latest/cuda/overview.html', '#cudatoolkit-lookup') warnings.warn(errors.NumbaWarning(msg)) /usr/local/lib/python3.6/dist-packages/numba/cuda/envvars.py:17: NumbaWarning: Environment variables with the 'NUMBAPRO' prefix are deprecated and consequently ignored, found use of NUMBAPRO_LIBDEVICE=/usr/local/cuda-10.0/nvvm/libdevice/. For more information about alternatives visit: ('https://numba.pydata.org/numba-doc/latest/cuda/overview.html', '#cudatoolkit-lookup') warnings.warn(errors.NumbaWarning(msg)) /home/wg/PointRCNN/tools/../lib/config.py:187: YAMLLoadWarning: calling yaml.load() without Loader=... is deprecated, as the default Loader is unsafe. Please read https://msg.pyyaml.org/load for full details. yaml_cfg = edict(yaml.load(f)) 2021-09-22 04:35:14,630 INFO **Start logging** 2021-09-22 04:35:14,630 INFO cfg_file cfgs/default.yaml 2021-09-22 04:35:14,630 INFO eval_mode rcnn 2021-09-22 04:35:14,630 INFO eval_all False 2021-09-22 04:35:14,630 INFO test False 2021-09-22 04:35:14,630 INFO ckpt PointRCNN.pth 2021-09-22 04:35:14,630 INFO rpn_ckpt None 2021-09-22 04:35:14,630 INFO rcnn_ckpt None 2021-09-22 04:35:14,630 INFO batch_size 1 2021-09-22 04:35:14,630 INFO workers 4 2021-09-22 04:35:14,630 INFO extra_tag default 2021-09-22 04:35:14,630 INFO output_dir None 2021-09-22 04:35:14,630 INFO ckpt_dir None 2021-09-22 04:35:14,630 INFO save_result False 2021-09-22 04:35:14,630 INFO save_rpn_feature False 2021-09-22 04:35:14,630 INFO random_select True 2021-09-22 04:35:14,630 INFO start_epoch 0 2021-09-22 04:35:14,630 INFO rcnn_eval_roi_dir None 2021-09-22 04:35:14,630 INFO rcnn_eval_feature_dir None 2021-09-22 04:35:14,630 INFO set_cfgs ['RPN.LOC_XZ_FINE', 'False'] 2021-09-22 04:35:14,630 INFO cfg.TAG: default 2021-09-22 04:35:14,630 INFO cfg.CLASSES: Car 2021-09-22 04:35:14,630 INFO cfg.INCLUDE_SIMILAR_TYPE: True 2021-09-22 04:35:14,630 INFO cfg.AUG_DATA: True 2021-09-22 04:35:14,630 INFO cfg.AUG_METHOD_LIST: ['rotation', 'scaling', 'flip'] 2021-09-22 04:35:14,630 INFO cfg.AUG_METHOD_PROB: [1.0, 1.0, 0.5] 2021-09-22 04:35:14,630 INFO cfg.AUG_ROT_RANGE: 18 2021-09-22 04:35:14,630 INFO cfg.GT_AUG_ENABLED: True 2021-09-22 04:35:14,631 INFO cfg.GT_EXTRA_NUM: 15 2021-09-22 04:35:14,631 INFO cfg.GT_AUG_RAND_NUM: True 2021-09-22 04:35:14,631 INFO cfg.GT_AUG_APPLY_PROB: 1.0 2021-09-22 04:35:14,631 INFO cfg.GT_AUG_HARD_RATIO: 0.6 2021-09-22 04:35:14,631 INFO cfg.PC_REDUCE_BY_RANGE: True 2021-09-22 04:35:14,631 INFO cfg.PC_AREA_SCOPE: [[-40. 40. ] [ -1. 3. ] [ 0. 70.4]] 2021-09-22 04:35:14,631 INFO cfg.CLS_MEAN_SIZE: [[1.5256319 1.6285675 3.8831165]] 2021-09-22 04:35:14,631 INFO cfg.RPN = edict() 2021-09-22 04:35:14,631 INFO cfg.RPN.ENABLED: True 2021-09-22 04:35:14,631 INFO cfg.RPN.FIXED: True 2021-09-22 04:35:14,631 INFO cfg.RPN.USE_INTENSITY: False 2021-09-22 04:35:14,631 INFO cfg.RPN.LOC_XZ_FINE: False 2021-09-22 04:35:14,631 INFO cfg.RPN.LOC_SCOPE: 3.0 2021-09-22 04:35:14,631 INFO cfg.RPN.LOC_BIN_SIZE: 0.5 2021-09-22 04:35:14,631 INFO cfg.RPN.NUM_HEAD_BIN: 12 2021-09-22 04:35:14,631 INFO cfg.RPN.BACKBONE: pointnet2_msg 2021-09-22 04:35:14,631 INFO cfg.RPN.USE_BN: True 2021-09-22 04:35:14,631 INFO cfg.RPN.NUM_POINTS: 16384 2021-09-22 04:35:14,631 INFO cfg.RPN.SA_CONFIG = edict() 2021-09-22 04:35:14,631 INFO cfg.RPN.SA_CONFIG.NPOINTS: [4096, 1024, 256, 64] 2021-09-22 04:35:14,631 INFO cfg.RPN.SA_CONFIG.RADIUS: [[0.1, 0.5], [0.5, 1.0], [1.0, 2.0], [2.0, 4.0]] 2021-09-22 04:35:14,632 INFO cfg.RPN.SA_CONFIG.NSAMPLE: [[16, 32], [16, 32], [16, 32], [16, 32]] 2021-09-22 04:35:14,632 INFO cfg.RPN.SA_CONFIG.MLPS: [[[16, 16, 32], [32, 32, 64]], [[64, 64, 128], [64, 96, 128]], [[128, 196, 256], [128, 196, 256]], [[256, 256, 512], [256, 384, 512]]] 2021-09-22 04:35:14,632 INFO cfg.RPN.FP_MLPS: [[128, 128], [256, 256], [512, 512], [512, 512]] 2021-09-22 04:35:14,632 INFO cfg.RPN.CLS_FC: [128] 2021-09-22 04:35:14,632 INFO cfg.RPN.REG_FC: [128] 2021-09-22 04:35:14,632 INFO cfg.RPN.DP_RATIO: 0.5 2021-09-22 04:35:14,632 INFO cfg.RPN.LOSS_CLS: SigmoidFocalLoss 2021-09-22 04:35:14,632 INFO cfg.RPN.FG_WEIGHT: 15 2021-09-22 04:35:14,632 INFO cfg.RPN.FOCAL_ALPHA: [0.25, 0.75] 2021-09-22 04:35:14,632 INFO cfg.RPN.FOCAL_GAMMA: 2.0 2021-09-22 04:35:14,632 INFO cfg.RPN.REG_LOSS_WEIGHT: [1.0, 1.0, 1.0, 1.0] 2021-09-22 04:35:14,632 INFO cfg.RPN.LOSS_WEIGHT: [1.0, 1.0] 2021-09-22 04:35:14,632 INFO cfg.RPN.NMS_TYPE: normal 2021-09-22 04:35:14,632 INFO cfg.RPN.SCORE_THRESH: 0.3 2021-09-22 04:35:14,632 INFO cfg.RCNN = edict() 2021-09-22 04:35:14,632 INFO cfg.RCNN.ENABLED: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_RPN_FEATURES: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_MASK: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.MASK_TYPE: seg 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_INTENSITY: False 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_DEPTH: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_SEG_SCORE: False 2021-09-22 04:35:14,632 INFO cfg.RCNN.ROI_SAMPLE_JIT: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.ROI_FG_AUG_TIMES: 10 2021-09-22 04:35:14,632 INFO cfg.RCNN.REG_AUG_METHOD: multiple 2021-09-22 04:35:14,632 INFO cfg.RCNN.POOL_EXTRA_WIDTH: 1.0 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_SCOPE: 1.5 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_BIN_SIZE: 0.5 2021-09-22 04:35:14,632 INFO cfg.RCNN.NUM_HEAD_BIN: 9 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_Y_BY_BIN: False 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_Y_SCOPE: 0.5 2021-09-22 04:35:14,633 INFO cfg.RCNN.LOC_Y_BIN_SIZE: 0.25 2021-09-22 04:35:14,633 INFO cfg.RCNN.SIZE_RES_ON_ROI: False 2021-09-22 04:35:14,633 INFO cfg.RCNN.USE_BN: False 2021-09-22 04:35:14,633 INFO cfg.RCNN.DP_RATIO: 0.0 2021-09-22 04:35:14,633 INFO cfg.RCNN.BACKBONE: pointnet 2021-09-22 04:35:14,633 INFO cfg.RCNN.XYZ_UP_LAYER: [128, 128] 2021-09-22 04:35:14,633 INFO cfg.RCNN.NUM_POINTS: 512 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG = edict() 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.NPOINTS: [128, 32, -1] 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.RADIUS: [0.2, 0.4, 100] 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.NSAMPLE: [64, 64, 64] 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.MLPS: [[128, 128, 128], [128, 128, 256], [256, 256, 512]] 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_FC: [256, 256] 2021-09-22 04:35:14,633 INFO cfg.RCNN.REG_FC: [256, 256] 2021-09-22 04:35:14,633 INFO cfg.RCNN.LOSS_CLS: BinaryCrossEntropy 2021-09-22 04:35:14,633 INFO cfg.RCNN.FOCAL_ALPHA: [0.25, 0.75] 2021-09-22 04:35:14,633 INFO cfg.RCNN.FOCAL_GAMMA: 2.0 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_WEIGHT: [1. 1. 1.] 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_FG_THRESH: 0.6 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_BG_THRESH: 0.45 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_BG_THRESH_LO: 0.05 2021-09-22 04:35:14,633 INFO cfg.RCNN.REG_FG_THRESH: 0.55 2021-09-22 04:35:14,633 INFO cfg.RCNN.FG_RATIO: 0.5 2021-09-22 04:35:14,633 INFO cfg.RCNN.ROI_PER_IMAGE: 64 2021-09-22 04:35:14,633 INFO cfg.RCNN.HARD_BG_RATIO: 0.8 2021-09-22 04:35:14,633 INFO cfg.RCNN.SCORE_THRESH: 0.3 2021-09-22 04:35:14,634 INFO cfg.RCNN.NMS_THRESH: 0.1 2021-09-22 04:35:14,634 INFO cfg.TRAIN = edict() 2021-09-22 04:35:14,634 INFO cfg.TRAIN.SPLIT: train 2021-09-22 04:35:14,634 INFO cfg.TRAIN.VAL_SPLIT: smallval 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR: 0.002 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR_CLIP: 1e-05 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR_DECAY: 0.5 2021-09-22 04:35:14,634 INFO cfg.TRAIN.DECAY_STEP_LIST: [100, 150, 180, 200] 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR_WARMUP: True 2021-09-22 04:35:14,634 INFO cfg.TRAIN.WARMUP_MIN: 0.0002 2021-09-22 04:35:14,634 INFO cfg.TRAIN.WARMUP_EPOCH: 1 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BN_MOMENTUM: 0.1 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BN_DECAY: 0.5 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BNM_CLIP: 0.01 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BN_DECAY_STEP_LIST: [1000] 2021-09-22 04:35:14,634 INFO cfg.TRAIN.OPTIMIZER: adam_onecycle 2021-09-22 04:35:14,634 INFO cfg.TRAIN.WEIGHT_DECAY: 0.001 2021-09-22 04:35:14,634 INFO cfg.TRAIN.MOMENTUM: 0.9 2021-09-22 04:35:14,634 INFO cfg.TRAIN.MOMS: [0.95, 0.85] 2021-09-22 04:35:14,634 INFO cfg.TRAIN.DIV_FACTOR: 10.0 2021-09-22 04:35:14,634 INFO cfg.TRAIN.PCT_START: 0.4 2021-09-22 04:35:14,634 INFO cfg.TRAIN.GRAD_NORM_CLIP: 1.0 2021-09-22 04:35:14,634 INFO cfg.TRAIN.RPN_PRE_NMS_TOP_N: 9000 2021-09-22 04:35:14,634 INFO cfg.TRAIN.RPN_POST_NMS_TOP_N: 512 2021-09-22 04:35:14,634 INFO cfg.TRAIN.RPN_NMS_THRESH: 0.85 2021-09-22 04:35:14,635 INFO cfg.TRAIN.RPN_DISTANCE_BASED_PROPOSE: True 2021-09-22 04:35:14,635 INFO cfg.TEST = edict() 2021-09-22 04:35:14,635 INFO cfg.TEST.SPLIT: val 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_PRE_NMS_TOP_N: 9000 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_POST_NMS_TOP_N: 100 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_NMS_THRESH: 0.8 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_DISTANCE_BASED_PROPOSE: True 2021-09-22 04:35:14,636 INFO Load testing samples from ../data/KITTI/object/training 2021-09-22 04:35:14,636 INFO Done: total test samples 3769 2021-09-22 04:35:16,962 INFO ==> Loading from checkpoint 'PointRCNN.pth' 2021-09-22 04:35:16,989 INFO ==> Done 2021-09-22 04:35:16,990 INFO ---- EPOCH no_number JOINT EVALUATION ---- 2021-09-22 04:35:16,990 INFO ==> Output file: ../output/rcnn/default/eval/epoch_no_number/val eval: 13%|▋ | 488/3769 [00:55<05:59, 9.13it/s, mode=EVAL, recall=1170/1769]Traceback (most recent call last): File "eval_rcnn.py", line 902, in eval_single_ckpt(root_result_dir) File "eval_rcnn.py", line 765, in eval_single_ckpt eval_one_epoch(model, test_loader, epoch_id, root_result_dir, logger) File "eval_rcnn.py", line 692, in eval_one_epoch ret_dict = eval_one_epoch_joint(model, dataloader, epoch_id, result_dir, logger) File "eval_rcnn.py", line 486, in eval_one_epoch_joint for data in dataloader: File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 582, in next return self._process_next_batch(batch) File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 608, in _process_next_batch raise batch.exc_type(batch.exc_msg) AssertionError: Traceback (most recent call last): File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/_utils/worker.py", line 99, in _worker_loop samples = collate_fn([dataset[i] for i in batch_indices]) File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/_utils/worker.py", line 99, in samples = collate_fn([dataset[i] for i in batch_indices]) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_rcnn_dataset.py", line 234, in getitem return self.get_rpn_sample(index) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_rcnn_dataset.py", line 249, in get_rpn_sample calib = self.get_calib(sample_id) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_rcnn_dataset.py", line 133, in get_calib return super().get_calib(idx % 10000) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_dataset.py", line 47, in get_calib assert os.path.exists(calib_file) AssertionError eval: 13%|▋ | 488/3769 [00:55<06:12, 8.80it/s, mode=EVAL, recall=1170/1769]
Hello? I found that I have the same problem with you! My process stopped when 448 13% too! Then I think that probably is not because the lack or wrong file path of dataset, what other problem??
嗯...我的问题就是因为数据集排列的路径不对。你可以仔细检查一下是否已经下载了planes数据集并且数据集排列的顺序没有错误。我并没有遇到其他问题
好的感谢!我还有一个问题,那个完整数据集太大了我只移动了1~1000进去,可能是因为这个原因造成的吗。planes我已经下载了。
一般来说对应的那几个部分都只放1~1000份数据进去应该是没有问题的。pointRCNN我没有试过把其中一部分放进去会是怎么样的,我就是单纯跑了一遍。如果你现阶段的目的单纯是复现的话,建议还是完整的来。
感谢老哥,已经跑通了
ok
检查一下数据集的路径是否正确,数据集的排布是否和作者的一样。还有那个plane这个文件也得下。 … ---Original--- From: @.**> Date: Wed, Sep 22, 2021 15:43 PM To: @.**>; Cc: @.**@.**>; Subject: Re: [sshaoshuai/PointRCNN] assert os.path.exists(calib_file) when run eval_rcnn.py (#87) thanks,I find the problem which is because of the wrong of the file path 老哥,能不能展开说说,啥错误的文件路径?我也遇到了和你一样的错误 — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.
我之前运行程序是会跑一段的,在448的时候突然停下来,在看到你的回复后,我仔细检查了两遍路径并下载了plane的文件,可是运行起来还是卡在了448的点,不知道老哥你是否能给些想法和建议: python eval_rcnn.py --cfg_file cfgs/default.yaml --ckpt PointRCNN.pth --batch_size 1 --eval_mode rcnn --set RPN.LOC_XZ_FINE False /usr/local/lib/python3.6/dist-packages/numba/cuda/envvars.py:17: NumbaWarning: Environment variables with the 'NUMBAPRO' prefix are deprecated and consequently ignored, found use of NUMBAPRO_NVVM=/usr/local/cuda-10.0/nvvm/lib64/libnvvm.so. For more information about alternatives visit: ('https://numba.pydata.org/numba-doc/latest/cuda/overview.html', '#cudatoolkit-lookup') warnings.warn(errors.NumbaWarning(msg)) /usr/local/lib/python3.6/dist-packages/numba/cuda/envvars.py:17: NumbaWarning: Environment variables with the 'NUMBAPRO' prefix are deprecated and consequently ignored, found use of NUMBAPRO_LIBDEVICE=/usr/local/cuda-10.0/nvvm/libdevice/. For more information about alternatives visit: ('https://numba.pydata.org/numba-doc/latest/cuda/overview.html', '#cudatoolkit-lookup') warnings.warn(errors.NumbaWarning(msg)) /home/wg/PointRCNN/tools/../lib/config.py:187: YAMLLoadWarning: calling yaml.load() without Loader=... is deprecated, as the default Loader is unsafe. Please read https://msg.pyyaml.org/load for full details. yaml_cfg = edict(yaml.load(f)) 2021-09-22 04:35:14,630 INFO **Start logging** 2021-09-22 04:35:14,630 INFO cfg_file cfgs/default.yaml 2021-09-22 04:35:14,630 INFO eval_mode rcnn 2021-09-22 04:35:14,630 INFO eval_all False 2021-09-22 04:35:14,630 INFO test False 2021-09-22 04:35:14,630 INFO ckpt PointRCNN.pth 2021-09-22 04:35:14,630 INFO rpn_ckpt None 2021-09-22 04:35:14,630 INFO rcnn_ckpt None 2021-09-22 04:35:14,630 INFO batch_size 1 2021-09-22 04:35:14,630 INFO workers 4 2021-09-22 04:35:14,630 INFO extra_tag default 2021-09-22 04:35:14,630 INFO output_dir None 2021-09-22 04:35:14,630 INFO ckpt_dir None 2021-09-22 04:35:14,630 INFO save_result False 2021-09-22 04:35:14,630 INFO save_rpn_feature False 2021-09-22 04:35:14,630 INFO random_select True 2021-09-22 04:35:14,630 INFO start_epoch 0 2021-09-22 04:35:14,630 INFO rcnn_eval_roi_dir None 2021-09-22 04:35:14,630 INFO rcnn_eval_feature_dir None 2021-09-22 04:35:14,630 INFO set_cfgs ['RPN.LOC_XZ_FINE', 'False'] 2021-09-22 04:35:14,630 INFO cfg.TAG: default 2021-09-22 04:35:14,630 INFO cfg.CLASSES: Car 2021-09-22 04:35:14,630 INFO cfg.INCLUDE_SIMILAR_TYPE: True 2021-09-22 04:35:14,630 INFO cfg.AUG_DATA: True 2021-09-22 04:35:14,630 INFO cfg.AUG_METHOD_LIST: ['rotation', 'scaling', 'flip'] 2021-09-22 04:35:14,630 INFO cfg.AUG_METHOD_PROB: [1.0, 1.0, 0.5] 2021-09-22 04:35:14,630 INFO cfg.AUG_ROT_RANGE: 18 2021-09-22 04:35:14,630 INFO cfg.GT_AUG_ENABLED: True 2021-09-22 04:35:14,631 INFO cfg.GT_EXTRA_NUM: 15 2021-09-22 04:35:14,631 INFO cfg.GT_AUG_RAND_NUM: True 2021-09-22 04:35:14,631 INFO cfg.GT_AUG_APPLY_PROB: 1.0 2021-09-22 04:35:14,631 INFO cfg.GT_AUG_HARD_RATIO: 0.6 2021-09-22 04:35:14,631 INFO cfg.PC_REDUCE_BY_RANGE: True 2021-09-22 04:35:14,631 INFO cfg.PC_AREA_SCOPE: [[-40. 40. ] [ -1. 3. ] [ 0. 70.4]] 2021-09-22 04:35:14,631 INFO cfg.CLS_MEAN_SIZE: [[1.5256319 1.6285675 3.8831165]] 2021-09-22 04:35:14,631 INFO cfg.RPN = edict() 2021-09-22 04:35:14,631 INFO cfg.RPN.ENABLED: True 2021-09-22 04:35:14,631 INFO cfg.RPN.FIXED: True 2021-09-22 04:35:14,631 INFO cfg.RPN.USE_INTENSITY: False 2021-09-22 04:35:14,631 INFO cfg.RPN.LOC_XZ_FINE: False 2021-09-22 04:35:14,631 INFO cfg.RPN.LOC_SCOPE: 3.0 2021-09-22 04:35:14,631 INFO cfg.RPN.LOC_BIN_SIZE: 0.5 2021-09-22 04:35:14,631 INFO cfg.RPN.NUM_HEAD_BIN: 12 2021-09-22 04:35:14,631 INFO cfg.RPN.BACKBONE: pointnet2_msg 2021-09-22 04:35:14,631 INFO cfg.RPN.USE_BN: True 2021-09-22 04:35:14,631 INFO cfg.RPN.NUM_POINTS: 16384 2021-09-22 04:35:14,631 INFO cfg.RPN.SA_CONFIG = edict() 2021-09-22 04:35:14,631 INFO cfg.RPN.SA_CONFIG.NPOINTS: [4096, 1024, 256, 64] 2021-09-22 04:35:14,631 INFO cfg.RPN.SA_CONFIG.RADIUS: [[0.1, 0.5], [0.5, 1.0], [1.0, 2.0], [2.0, 4.0]] 2021-09-22 04:35:14,632 INFO cfg.RPN.SA_CONFIG.NSAMPLE: [[16, 32], [16, 32], [16, 32], [16, 32]] 2021-09-22 04:35:14,632 INFO cfg.RPN.SA_CONFIG.MLPS: [[[16, 16, 32], [32, 32, 64]], [[64, 64, 128], [64, 96, 128]], [[128, 196, 256], [128, 196, 256]], [[256, 256, 512], [256, 384, 512]]] 2021-09-22 04:35:14,632 INFO cfg.RPN.FP_MLPS: [[128, 128], [256, 256], [512, 512], [512, 512]] 2021-09-22 04:35:14,632 INFO cfg.RPN.CLS_FC: [128] 2021-09-22 04:35:14,632 INFO cfg.RPN.REG_FC: [128] 2021-09-22 04:35:14,632 INFO cfg.RPN.DP_RATIO: 0.5 2021-09-22 04:35:14,632 INFO cfg.RPN.LOSS_CLS: SigmoidFocalLoss 2021-09-22 04:35:14,632 INFO cfg.RPN.FG_WEIGHT: 15 2021-09-22 04:35:14,632 INFO cfg.RPN.FOCAL_ALPHA: [0.25, 0.75] 2021-09-22 04:35:14,632 INFO cfg.RPN.FOCAL_GAMMA: 2.0 2021-09-22 04:35:14,632 INFO cfg.RPN.REG_LOSS_WEIGHT: [1.0, 1.0, 1.0, 1.0] 2021-09-22 04:35:14,632 INFO cfg.RPN.LOSS_WEIGHT: [1.0, 1.0] 2021-09-22 04:35:14,632 INFO cfg.RPN.NMS_TYPE: normal 2021-09-22 04:35:14,632 INFO cfg.RPN.SCORE_THRESH: 0.3 2021-09-22 04:35:14,632 INFO cfg.RCNN = edict() 2021-09-22 04:35:14,632 INFO cfg.RCNN.ENABLED: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_RPN_FEATURES: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_MASK: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.MASK_TYPE: seg 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_INTENSITY: False 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_DEPTH: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.USE_SEG_SCORE: False 2021-09-22 04:35:14,632 INFO cfg.RCNN.ROI_SAMPLE_JIT: True 2021-09-22 04:35:14,632 INFO cfg.RCNN.ROI_FG_AUG_TIMES: 10 2021-09-22 04:35:14,632 INFO cfg.RCNN.REG_AUG_METHOD: multiple 2021-09-22 04:35:14,632 INFO cfg.RCNN.POOL_EXTRA_WIDTH: 1.0 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_SCOPE: 1.5 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_BIN_SIZE: 0.5 2021-09-22 04:35:14,632 INFO cfg.RCNN.NUM_HEAD_BIN: 9 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_Y_BY_BIN: False 2021-09-22 04:35:14,632 INFO cfg.RCNN.LOC_Y_SCOPE: 0.5 2021-09-22 04:35:14,633 INFO cfg.RCNN.LOC_Y_BIN_SIZE: 0.25 2021-09-22 04:35:14,633 INFO cfg.RCNN.SIZE_RES_ON_ROI: False 2021-09-22 04:35:14,633 INFO cfg.RCNN.USE_BN: False 2021-09-22 04:35:14,633 INFO cfg.RCNN.DP_RATIO: 0.0 2021-09-22 04:35:14,633 INFO cfg.RCNN.BACKBONE: pointnet 2021-09-22 04:35:14,633 INFO cfg.RCNN.XYZ_UP_LAYER: [128, 128] 2021-09-22 04:35:14,633 INFO cfg.RCNN.NUM_POINTS: 512 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG = edict() 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.NPOINTS: [128, 32, -1] 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.RADIUS: [0.2, 0.4, 100] 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.NSAMPLE: [64, 64, 64] 2021-09-22 04:35:14,633 INFO cfg.RCNN.SA_CONFIG.MLPS: [[128, 128, 128], [128, 128, 256], [256, 256, 512]] 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_FC: [256, 256] 2021-09-22 04:35:14,633 INFO cfg.RCNN.REG_FC: [256, 256] 2021-09-22 04:35:14,633 INFO cfg.RCNN.LOSS_CLS: BinaryCrossEntropy 2021-09-22 04:35:14,633 INFO cfg.RCNN.FOCAL_ALPHA: [0.25, 0.75] 2021-09-22 04:35:14,633 INFO cfg.RCNN.FOCAL_GAMMA: 2.0 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_WEIGHT: [1. 1. 1.] 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_FG_THRESH: 0.6 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_BG_THRESH: 0.45 2021-09-22 04:35:14,633 INFO cfg.RCNN.CLS_BG_THRESH_LO: 0.05 2021-09-22 04:35:14,633 INFO cfg.RCNN.REG_FG_THRESH: 0.55 2021-09-22 04:35:14,633 INFO cfg.RCNN.FG_RATIO: 0.5 2021-09-22 04:35:14,633 INFO cfg.RCNN.ROI_PER_IMAGE: 64 2021-09-22 04:35:14,633 INFO cfg.RCNN.HARD_BG_RATIO: 0.8 2021-09-22 04:35:14,633 INFO cfg.RCNN.SCORE_THRESH: 0.3 2021-09-22 04:35:14,634 INFO cfg.RCNN.NMS_THRESH: 0.1 2021-09-22 04:35:14,634 INFO cfg.TRAIN = edict() 2021-09-22 04:35:14,634 INFO cfg.TRAIN.SPLIT: train 2021-09-22 04:35:14,634 INFO cfg.TRAIN.VAL_SPLIT: smallval 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR: 0.002 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR_CLIP: 1e-05 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR_DECAY: 0.5 2021-09-22 04:35:14,634 INFO cfg.TRAIN.DECAY_STEP_LIST: [100, 150, 180, 200] 2021-09-22 04:35:14,634 INFO cfg.TRAIN.LR_WARMUP: True 2021-09-22 04:35:14,634 INFO cfg.TRAIN.WARMUP_MIN: 0.0002 2021-09-22 04:35:14,634 INFO cfg.TRAIN.WARMUP_EPOCH: 1 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BN_MOMENTUM: 0.1 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BN_DECAY: 0.5 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BNM_CLIP: 0.01 2021-09-22 04:35:14,634 INFO cfg.TRAIN.BN_DECAY_STEP_LIST: [1000] 2021-09-22 04:35:14,634 INFO cfg.TRAIN.OPTIMIZER: adam_onecycle 2021-09-22 04:35:14,634 INFO cfg.TRAIN.WEIGHT_DECAY: 0.001 2021-09-22 04:35:14,634 INFO cfg.TRAIN.MOMENTUM: 0.9 2021-09-22 04:35:14,634 INFO cfg.TRAIN.MOMS: [0.95, 0.85] 2021-09-22 04:35:14,634 INFO cfg.TRAIN.DIV_FACTOR: 10.0 2021-09-22 04:35:14,634 INFO cfg.TRAIN.PCT_START: 0.4 2021-09-22 04:35:14,634 INFO cfg.TRAIN.GRAD_NORM_CLIP: 1.0 2021-09-22 04:35:14,634 INFO cfg.TRAIN.RPN_PRE_NMS_TOP_N: 9000 2021-09-22 04:35:14,634 INFO cfg.TRAIN.RPN_POST_NMS_TOP_N: 512 2021-09-22 04:35:14,634 INFO cfg.TRAIN.RPN_NMS_THRESH: 0.85 2021-09-22 04:35:14,635 INFO cfg.TRAIN.RPN_DISTANCE_BASED_PROPOSE: True 2021-09-22 04:35:14,635 INFO cfg.TEST = edict() 2021-09-22 04:35:14,635 INFO cfg.TEST.SPLIT: val 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_PRE_NMS_TOP_N: 9000 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_POST_NMS_TOP_N: 100 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_NMS_THRESH: 0.8 2021-09-22 04:35:14,635 INFO cfg.TEST.RPN_DISTANCE_BASED_PROPOSE: True 2021-09-22 04:35:14,636 INFO Load testing samples from ../data/KITTI/object/training 2021-09-22 04:35:14,636 INFO Done: total test samples 3769 2021-09-22 04:35:16,962 INFO ==> Loading from checkpoint 'PointRCNN.pth' 2021-09-22 04:35:16,989 INFO ==> Done 2021-09-22 04:35:16,990 INFO ---- EPOCH no_number JOINT EVALUATION ---- 2021-09-22 04:35:16,990 INFO ==> Output file: ../output/rcnn/default/eval/epoch_no_number/val eval: 13%|▋ | 488/3769 [00:55<05:59, 9.13it/s, mode=EVAL, recall=1170/1769]Traceback (most recent call last): File "eval_rcnn.py", line 902, in eval_single_ckpt(root_result_dir) File "eval_rcnn.py", line 765, in eval_single_ckpt eval_one_epoch(model, test_loader, epoch_id, root_result_dir, logger) File "eval_rcnn.py", line 692, in eval_one_epoch ret_dict = eval_one_epoch_joint(model, dataloader, epoch_id, result_dir, logger) File "eval_rcnn.py", line 486, in eval_one_epoch_joint for data in dataloader: File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 582, in next return self._process_next_batch(batch) File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 608, in _process_next_batch raise batch.exc_type(batch.exc_msg) AssertionError: Traceback (most recent call last): File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/_utils/worker.py", line 99, in _worker_loop samples = collate_fn([dataset[i] for i in batch_indices]) File "/home/wg/.local/lib/python3.6/site-packages/torch/utils/data/_utils/worker.py", line 99, in samples = collate_fn([dataset[i] for i in batch_indices]) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_rcnn_dataset.py", line 234, in getitem return self.get_rpn_sample(index) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_rcnn_dataset.py", line 249, in get_rpn_sample calib = self.get_calib(sample_id) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_rcnn_dataset.py", line 133, in get_calib return super().get_calib(idx % 10000) File "/home/wg/PointRCNN/tools/../lib/datasets/kitti_dataset.py", line 47, in get_calib assert os.path.exists(calib_file) AssertionError eval: 13%|▋ | 488/3769 [00:55<06:12, 8.80it/s, mode=EVAL, recall=1170/1769]
Hello? I found that I have the same problem with you! My process stopped when 448 13% too! Then I think that probably is not because the lack or wrong file path of dataset, what other problem??
嗯...我的问题就是因为数据集排列的路径不对。你可以仔细检查一下是否已经下载了planes数据集并且数据集排列的顺序没有错误。我并没有遇到其他问题
好的感谢!我还有一个问题,那个完整数据集太大了我只移动了1~1000进去,可能是因为这个原因造成的吗。planes我已经下载了。
一般来说对应的那几个部分都只放1~1000份数据进去应该是没有问题的。pointRCNN我没有试过把其中一部分放进去会是怎么样的,我就是单纯跑了一遍。如果你现阶段的目的单纯是复现的话,建议还是完整的来。
感谢老哥,已经跑通了
ok
再问一下,你做可视化了嘛
你指的是数据集的雷达图?
Can anyone help it got stuck at 7481 on test.txt.....it works fine on val and trainval.txt.
File "/tools/../lib/datasets/kitti_dataset.py", line 154, in get_calib assert os.path.exists(calib_file) AssertionError
eval: 100%|███▉| 7481/7518 [15:03<00:04, 8.28it/s, mode=EVAL, recall=2113/4487]
2019-08-08 16:50:28,702 INFO **Start logging** 2019-08-08 16:50:28,702 INFO cfg_file cfgs/default.yaml 2019-08-08 16:50:28,702 INFO eval_mode rcnn 2019-08-08 16:50:28,702 INFO eval_all False 2019-08-08 16:50:28,703 INFO test False 2019-08-08 16:50:28,703 INFO ckpt ../PointRCNN.pth 2019-08-08 16:50:28,703 INFO rpn_ckpt None 2019-08-08 16:50:28,703 INFO rcnn_ckpt None 2019-08-08 16:50:28,703 INFO batch_size 1 2019-08-08 16:50:28,703 INFO workers 4 2019-08-08 16:50:28,703 INFO extra_tag default 2019-08-08 16:50:28,703 INFO output_dir None 2019-08-08 16:50:28,703 INFO ckpt_dir None 2019-08-08 16:50:28,703 INFO save_result False 2019-08-08 16:50:28,703 INFO save_rpn_feature False 2019-08-08 16:50:28,703 INFO random_select True 2019-08-08 16:50:28,703 INFO start_epoch 0 2019-08-08 16:50:28,703 INFO rcnn_eval_roi_dir None 2019-08-08 16:50:28,703 INFO rcnn_eval_feature_dir None 2019-08-08 16:50:28,703 INFO set_cfgs ['RPN.LOC_XZ_FINE', 'False'] 2019-08-08 16:50:28,703 INFO cfg.TAG: default 2019-08-08 16:50:28,703 INFO cfg.CLASSES: Car 2019-08-08 16:50:28,703 INFO cfg.INCLUDE_SIMILAR_TYPE: True 2019-08-08 16:50:28,703 INFO cfg.AUG_DATA: True 2019-08-08 16:50:28,703 INFO cfg.AUG_METHOD_LIST: ['rotation', 'scaling', 'flip'] 2019-08-08 16:50:28,703 INFO cfg.AUG_METHOD_PROB: [1.0, 1.0, 0.5] 2019-08-08 16:50:28,703 INFO cfg.AUG_ROT_RANGE: 18 2019-08-08 16:50:28,703 INFO cfg.GT_AUG_ENABLED: True 2019-08-08 16:50:28,703 INFO cfg.GT_EXTRA_NUM: 15 2019-08-08 16:50:28,704 INFO cfg.GT_AUG_RAND_NUM: True 2019-08-08 16:50:28,704 INFO cfg.GT_AUG_APPLY_PROB: 1.0 2019-08-08 16:50:28,704 INFO cfg.GT_AUG_HARD_RATIO: 0.6 2019-08-08 16:50:28,704 INFO cfg.PC_REDUCE_BY_RANGE: True 2019-08-08 16:50:28,704 INFO cfg.PC_AREA_SCOPE: [[-40. 40. ] [ -1. 3. ] [ 0. 70.4]] 2019-08-08 16:50:28,704 INFO cfg.CLS_MEAN_SIZE: [[1.5256319 1.6285675 3.8831165]] 2019-08-08 16:50:28,704 INFO
eval_single_ckpt(root_result_dir)
File "eval_rcnn.py", line 765, in eval_single_ckpt
eval_one_epoch(model, test_loader, epoch_id, root_result_dir, logger)
File "eval_rcnn.py", line 692, in eval_one_epoch
ret_dict = eval_one_epoch_joint(model, dataloader, epoch_id, result_dir, logger)
File "eval_rcnn.py", line 486, in eval_one_epoch_joint
for data in dataloader:
File "/home/vincent/anaconda3/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 582, in next
return self._process_next_batch(batch)
File "/home/vincent/anaconda3/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 608, in _process_next_batch
raise batch.exc_type(batch.exc_msg)
AssertionError: Traceback (most recent call last):
File "/home/vincent/anaconda3/lib/python3.7/site-packages/torch/utils/data/_utils/worker.py", line 99, in _worker_loop
samples = collate_fn([dataset[i] for i in batch_indices])
File "/home/vincent/anaconda3/lib/python3.7/site-packages/torch/utils/data/_utils/worker.py", line 99, in
samples = collate_fn([dataset[i] for i in batch_indices])
File "/home/vincent/PointRCNN/tools/../lib/datasets/kitti_rcnn_dataset.py", line 234, in getitem
return self.get_rpn_sample(index)
File "/home/vincent/PointRCNN/tools/../lib/datasets/kitti_rcnn_dataset.py", line 249, in get_rpn_sample
calib = self.get_calib(sample_id)
File "/home/vincent/PointRCNN/tools/../lib/datasets/kitti_rcnn_dataset.py", line 133, in get_calib
return super().get_calib(idx % 10000)
File "/home/vincent/PointRCNN/tools/../lib/datasets/kitti_dataset.py", line 47, in get_calib
assert os.path.exists(calib_file)
AssertionError
cfg.RPN = edict() 2019-08-08 16:50:28,704 INFO cfg.RPN.ENABLED: True 2019-08-08 16:50:28,704 INFO cfg.RPN.FIXED: True 2019-08-08 16:50:28,704 INFO cfg.RPN.USE_INTENSITY: False 2019-08-08 16:50:28,704 INFO cfg.RPN.LOC_XZ_FINE: False 2019-08-08 16:50:28,704 INFO cfg.RPN.LOC_SCOPE: 3.0 2019-08-08 16:50:28,704 INFO cfg.RPN.LOC_BIN_SIZE: 0.5 2019-08-08 16:50:28,704 INFO cfg.RPN.NUM_HEAD_BIN: 12 2019-08-08 16:50:28,704 INFO cfg.RPN.BACKBONE: pointnet2_msg 2019-08-08 16:50:28,704 INFO cfg.RPN.USE_BN: True 2019-08-08 16:50:28,705 INFO cfg.RPN.NUM_POINTS: 16384 2019-08-08 16:50:28,705 INFO
cfg.RPN.SA_CONFIG = edict() 2019-08-08 16:50:28,705 INFO cfg.RPN.SA_CONFIG.NPOINTS: [4096, 1024, 256, 64] 2019-08-08 16:50:28,705 INFO cfg.RPN.SA_CONFIG.RADIUS: [[0.1, 0.5], [0.5, 1.0], [1.0, 2.0], [2.0, 4.0]] 2019-08-08 16:50:28,705 INFO cfg.RPN.SA_CONFIG.NSAMPLE: [[16, 32], [16, 32], [16, 32], [16, 32]] 2019-08-08 16:50:28,705 INFO cfg.RPN.SA_CONFIG.MLPS: [[[16, 16, 32], [32, 32, 64]], [[64, 64, 128], [64, 96, 128]], [[128, 196, 256], [128, 196, 256]], [[256, 256, 512], [256, 384, 512]]] 2019-08-08 16:50:28,705 INFO cfg.RPN.FP_MLPS: [[128, 128], [256, 256], [512, 512], [512, 512]] 2019-08-08 16:50:28,705 INFO cfg.RPN.CLS_FC: [128] 2019-08-08 16:50:28,705 INFO cfg.RPN.REG_FC: [128] 2019-08-08 16:50:28,705 INFO cfg.RPN.DP_RATIO: 0.5 2019-08-08 16:50:28,705 INFO cfg.RPN.LOSS_CLS: SigmoidFocalLoss 2019-08-08 16:50:28,705 INFO cfg.RPN.FG_WEIGHT: 15 2019-08-08 16:50:28,705 INFO cfg.RPN.FOCAL_ALPHA: [0.25, 0.75] 2019-08-08 16:50:28,705 INFO cfg.RPN.FOCAL_GAMMA: 2.0 2019-08-08 16:50:28,705 INFO cfg.RPN.REG_LOSS_WEIGHT: [1.0, 1.0, 1.0, 1.0] 2019-08-08 16:50:28,705 INFO cfg.RPN.LOSS_WEIGHT: [1.0, 1.0] 2019-08-08 16:50:28,705 INFO cfg.RPN.NMS_TYPE: normal 2019-08-08 16:50:28,705 INFO cfg.RPN.SCORE_THRESH: 0.3 2019-08-08 16:50:28,705 INFO
cfg.RCNN = edict() 2019-08-08 16:50:28,705 INFO cfg.RCNN.ENABLED: True 2019-08-08 16:50:28,705 INFO cfg.RCNN.USE_RPN_FEATURES: True 2019-08-08 16:50:28,705 INFO cfg.RCNN.USE_MASK: True 2019-08-08 16:50:28,705 INFO cfg.RCNN.MASK_TYPE: seg 2019-08-08 16:50:28,705 INFO cfg.RCNN.USE_INTENSITY: False 2019-08-08 16:50:28,706 INFO cfg.RCNN.USE_DEPTH: True 2019-08-08 16:50:28,706 INFO cfg.RCNN.USE_SEG_SCORE: False 2019-08-08 16:50:28,706 INFO cfg.RCNN.ROI_SAMPLE_JIT: True 2019-08-08 16:50:28,706 INFO cfg.RCNN.ROI_FG_AUG_TIMES: 10 2019-08-08 16:50:28,706 INFO cfg.RCNN.REG_AUG_METHOD: multiple 2019-08-08 16:50:28,706 INFO cfg.RCNN.POOL_EXTRA_WIDTH: 1.0 2019-08-08 16:50:28,706 INFO cfg.RCNN.LOC_SCOPE: 1.5 2019-08-08 16:50:28,706 INFO cfg.RCNN.LOC_BIN_SIZE: 0.5 2019-08-08 16:50:28,706 INFO cfg.RCNN.NUM_HEAD_BIN: 9 2019-08-08 16:50:28,706 INFO cfg.RCNN.LOC_Y_BY_BIN: False 2019-08-08 16:50:28,706 INFO cfg.RCNN.LOC_Y_SCOPE: 0.5 2019-08-08 16:50:28,706 INFO cfg.RCNN.LOC_Y_BIN_SIZE: 0.25 2019-08-08 16:50:28,706 INFO cfg.RCNN.SIZE_RES_ON_ROI: False 2019-08-08 16:50:28,706 INFO cfg.RCNN.USE_BN: False 2019-08-08 16:50:28,706 INFO cfg.RCNN.DP_RATIO: 0.0 2019-08-08 16:50:28,706 INFO cfg.RCNN.BACKBONE: pointnet 2019-08-08 16:50:28,706 INFO cfg.RCNN.XYZ_UP_LAYER: [128, 128] 2019-08-08 16:50:28,706 INFO cfg.RCNN.NUM_POINTS: 512 2019-08-08 16:50:28,706 INFO
cfg.RCNN.SA_CONFIG = edict() 2019-08-08 16:50:28,706 INFO cfg.RCNN.SA_CONFIG.NPOINTS: [128, 32, -1] 2019-08-08 16:50:28,706 INFO cfg.RCNN.SA_CONFIG.RADIUS: [0.2, 0.4, 100] 2019-08-08 16:50:28,706 INFO cfg.RCNN.SA_CONFIG.NSAMPLE: [64, 64, 64] 2019-08-08 16:50:28,706 INFO cfg.RCNN.SA_CONFIG.MLPS: [[128, 128, 128], [128, 128, 256], [256, 256, 512]] 2019-08-08 16:50:28,706 INFO cfg.RCNN.CLS_FC: [256, 256] 2019-08-08 16:50:28,706 INFO cfg.RCNN.REG_FC: [256, 256] 2019-08-08 16:50:28,707 INFO cfg.RCNN.LOSS_CLS: BinaryCrossEntropy 2019-08-08 16:50:28,707 INFO cfg.RCNN.FOCAL_ALPHA: [0.25, 0.75] 2019-08-08 16:50:28,707 INFO cfg.RCNN.FOCAL_GAMMA: 2.0 2019-08-08 16:50:28,707 INFO cfg.RCNN.CLS_WEIGHT: [1. 1. 1.] 2019-08-08 16:50:28,707 INFO cfg.RCNN.CLS_FG_THRESH: 0.6 2019-08-08 16:50:28,707 INFO cfg.RCNN.CLS_BG_THRESH: 0.45 2019-08-08 16:50:28,707 INFO cfg.RCNN.CLS_BG_THRESH_LO: 0.05 2019-08-08 16:50:28,707 INFO cfg.RCNN.REG_FG_THRESH: 0.55 2019-08-08 16:50:28,707 INFO cfg.RCNN.FG_RATIO: 0.5 2019-08-08 16:50:28,707 INFO cfg.RCNN.ROI_PER_IMAGE: 64 2019-08-08 16:50:28,707 INFO cfg.RCNN.HARD_BG_RATIO: 0.8 2019-08-08 16:50:28,707 INFO cfg.RCNN.SCORE_THRESH: 0.3 2019-08-08 16:50:28,707 INFO cfg.RCNN.NMS_THRESH: 0.1 2019-08-08 16:50:28,707 INFO
cfg.TRAIN = edict() 2019-08-08 16:50:28,707 INFO cfg.TRAIN.SPLIT: train 2019-08-08 16:50:28,707 INFO cfg.TRAIN.VAL_SPLIT: smallval 2019-08-08 16:50:28,707 INFO cfg.TRAIN.LR: 0.002 2019-08-08 16:50:28,707 INFO cfg.TRAIN.LR_CLIP: 1e-05 2019-08-08 16:50:28,707 INFO cfg.TRAIN.LR_DECAY: 0.5 2019-08-08 16:50:28,707 INFO cfg.TRAIN.DECAY_STEP_LIST: [100, 150, 180, 200] 2019-08-08 16:50:28,707 INFO cfg.TRAIN.LR_WARMUP: True 2019-08-08 16:50:28,707 INFO cfg.TRAIN.WARMUP_MIN: 0.0002 2019-08-08 16:50:28,708 INFO cfg.TRAIN.WARMUP_EPOCH: 1 2019-08-08 16:50:28,708 INFO cfg.TRAIN.BN_MOMENTUM: 0.1 2019-08-08 16:50:28,708 INFO cfg.TRAIN.BN_DECAY: 0.5 2019-08-08 16:50:28,708 INFO cfg.TRAIN.BNM_CLIP: 0.01 2019-08-08 16:50:28,708 INFO cfg.TRAIN.BN_DECAY_STEP_LIST: [1000] 2019-08-08 16:50:28,708 INFO cfg.TRAIN.OPTIMIZER: adam_onecycle 2019-08-08 16:50:28,708 INFO cfg.TRAIN.WEIGHT_DECAY: 0.001 2019-08-08 16:50:28,708 INFO cfg.TRAIN.MOMENTUM: 0.9 2019-08-08 16:50:28,708 INFO cfg.TRAIN.MOMS: [0.95, 0.85] 2019-08-08 16:50:28,708 INFO cfg.TRAIN.DIV_FACTOR: 10.0 2019-08-08 16:50:28,708 INFO cfg.TRAIN.PCT_START: 0.4 2019-08-08 16:50:28,708 INFO cfg.TRAIN.GRAD_NORM_CLIP: 1.0 2019-08-08 16:50:28,708 INFO cfg.TRAIN.RPN_PRE_NMS_TOP_N: 9000 2019-08-08 16:50:28,708 INFO cfg.TRAIN.RPN_POST_NMS_TOP_N: 512 2019-08-08 16:50:28,708 INFO cfg.TRAIN.RPN_NMS_THRESH: 0.85 2019-08-08 16:50:28,708 INFO cfg.TRAIN.RPN_DISTANCE_BASED_PROPOSE: True 2019-08-08 16:50:28,708 INFO
cfg.TEST = edict() 2019-08-08 16:50:28,708 INFO cfg.TEST.SPLIT: val 2019-08-08 16:50:28,708 INFO cfg.TEST.RPN_PRE_NMS_TOP_N: 9000 2019-08-08 16:50:28,708 INFO cfg.TEST.RPN_POST_NMS_TOP_N: 100 2019-08-08 16:50:28,708 INFO cfg.TEST.RPN_NMS_THRESH: 0.8 2019-08-08 16:50:28,708 INFO cfg.TEST.RPN_DISTANCE_BASED_PROPOSE: True 2019-08-08 16:50:28,709 INFO Load testing samples from ../data/KITTI/object/training 2019-08-08 16:50:28,710 INFO Done: total test samples 3769 2019-08-08 16:50:30,785 INFO ==> Loading from checkpoint '../PointRCNN.pth' 2019-08-08 16:50:31,038 INFO ==> Done 2019-08-08 16:50:31,074 INFO ---- EPOCH no_number JOINT EVALUATION ---- 2019-08-08 16:50:31,075 INFO ==> Output file: ../output/rcnn/default/eval/epoch_no_number/val eval: 0%| | 0/3769 [00:00<?, ?it/s]Traceback (most recent call last): File "eval_rcnn.py", line 902, in