cskkxjk / mononerd Goto Github PK

/bin/bash: line 1: ./scripts/dist_test_ckpt.sh: Permission denied

Could anyone help me with this error when running the code on Google Colab? Thank you!

我下载了kitti_object，里面没有imagesets这个文件夹

Traceback (most recent call last):
File "tools/train.py", line 214, in
main()
File "tools/train.py", line 130, in main
model = build_network(model_cfg=cfg.MODEL, num_class=len(
File "/mnt/petrelfs/dengken_nerf_perception/perscene_OD/MonoNeRD/mononerd/models/init.py", line 12, in build_network
model = build_mono_detector(
File "/mnt/petrelfs/dengken_nerf_perception/perscene_OD/MonoNeRD/mononerd/models/detectors_mono/init.py", line 9, in build_detector
model = all[model_cfg.NAME](
File "/mnt/petrelfs/dengken_nerf_perception/perscene_OD/MonoNeRD/mononerd/models/detectors_mono/mononerd.py", line 10, in init
self.module_list = self.build_networks()
File "/mnt/petrelfs/dengken_nerf_perception/perscene_OD/MonoNeRD/mononerd/models/detectors_mono/stereo_detector3d_template.py", line 55, in build_networks
module, model_info_dict = getattr(self, 'build_%s' % module_name)(
File "/mnt/petrelfs/dengken_nerf_perception/perscene_OD/MonoNeRD/mononerd/models/detectors_mono/stereo_detector3d_template.py", line 85, in build_backbone_3d
backbone_3d_module = backbones_3d_mono.all[self.model_cfg.BACKBONE_3D.NAME](
File "/mnt/petrelfs/dengken_nerf_perception/perscene_OD/MonoNeRD/mononerd/models/backbones_3d_mono/mononerd_backbone.py", line 89, in init
self.feature_backbone = build_backbone(model_cfg.feature_backbone)
File "/mnt/lustre/houyuenan/anaconda3/envs/mononerd/lib/python3.8/site-packages/mmdet/models/builder.py", line 37, in build_backbone
return build(cfg, BACKBONES)
File "/mnt/lustre/houyuenan/anaconda3/envs/mononerd/lib/python3.8/site-packages/mmdet/models/builder.py", line 32, in build
return build_from_cfg(cfg, registry, default_args)
File "/mnt/lustre/houyuenan/anaconda3/envs/mononerd/lib/python3.8/site-packages/mmcv/utils/registry.py", line 171, in build_from_cfg
return obj_cls(**args)
TypeError: init() got an unexpected keyword argument 'with_max_pool'

运行命令：./scripts/dist_train.sh 1 'exp_name' ./configs/stereo/kitti_models/mononerd.3d-and-bev.yaml

In the monocular 3D object detection task, the file 'mononerd/models/detectors_mono/stereo_detector3d_template.py' contains the function post_processing() which is used to generate predictions of KITTI dataset.

My question is: How can I get all scores of an object (one of the objects belongs to the same image) instead of the highest one?

For example, for a car object, its scores may be [0.1333, 0.04, 0.05] (corresponding to 3 classes). I want to get the whole output [0.1333, 0.04, 0.05] rather than the highest value, i.e., 0.1333

Thanks for your patient reply :)

First of all, thanks for your work to the community!

I am new in this area, and I wonder why mononerd needs the right color images of kitti dataset in the evaluation stage?

Since the size of the Waymo dataset is too large to download, I wonder if you could provide the pretrained weight and config file of MonoNeRD on the Waymo dataset for me?

My email address is [email protected]

Looking forward to your timely reply!

When I try to run ./scripts/dist_test_ckpt.sh ${NUM_GPUS=1} ./configs/stereo/kitti_models/mononerd.yaml ./ckpt/pretrained_mononerd.pth

It still runs the cmd python -m torch.distributed.launch --nproc_per_node=8 tools/test.py --launcher pytorch --save_to_file --cfg_file pretrained-config.yaml --ckpt checkpoint.pth

Besides, the code fails when I run python -m torch.distributed.launch --nproc_per_node=1 tools/test.py --launcher pytorch --save_to_file --cfg_file pretrained-config.yaml --ckpt checkpoint.pth or python tools/test.py --save_to_file --cfg_file pretrained-config.yaml --ckpt checkpoint.pth

Looking forward for your reply!

I see the stereo configs for KITTI and mono configs for Waymo OD. Do you have mono configs for KITTI? Thanks.

Traceback (most recent call last):
  File "/home/user/anaconda3/envs/mono3d/lib/python3.8/runpy.py", line 194, in _run_module_as_main
    return _run_code(code, main_globals, None,
  File "/home/user/anaconda3/envs/mono3d/lib/python3.8/runpy.py", line 87, in _run_code
    exec(code, run_globals)
  File "/home/user/anaconda3/envs/mono3d/lib/python3.8/site-packages/torch/distributed/launch.py", line 340, in <module>
    main()
  File "/home/user/anaconda3/envs/mono3d/lib/python3.8/site-packages/torch/distributed/launch.py", line 326, in main
    sigkill_handler(signal.SIGTERM, None)  # not coming back
  File "/home/user/anaconda3/envs/mono3d/lib/python3.8/site-packages/torch/distributed/launch.py", line 301, in sigkill_handler
    raise subprocess.CalledProcessError(returncode=last_return_code, cmd=cmd)
subprocess.CalledProcessError: Command '['/home/user/anaconda3/envs/mono3d/bin/python', '-u', 'tools/train.py', '--local_rank=3', '--launcher', 'pytorch', '--fix_random_seed', '--sync_bn', '--save_to_file', '--cfg_file', './configs/stereo/kitti_models/mononerd.3d-and-bev.yaml', '--exp_name', 'exp_name']' died with <Signals.SIGSEGV: 11>.

/home/user/anaconda3/envs/mono3d/lib/python3.8/multiprocessing/resource_tracker.py:216: UserWarning: resource_tracker: There appear to be 29 leaked semaphore objects to clean up at shutdown
  warnings.warn('resource_tracker: There appear to be %d '
/home/user/anaconda3/envs/mono3d/lib/python3.8/multiprocessing/resource_tracker.py:216: UserWarning: resource_tracker: There appear to be 29 leaked semaphore objects to clean up at shutdown
  warnings.warn('resource_tracker: There appear to be %d '

请问这个问题该如何解决?

May I ask how to solve this problem?

或者您可以提供docker吗?

Or can you provide a Docker?

May I ask if the result of your KITTI val set is the result of the last epoch selected?
I found that your results on the KITTI val set are much lower than those on the KITTI test set, and when I reproduced them, I found that the highest KITTI AP | 3D from 52 epoch can reach 22.50/17.61/16.94, and from then on, the AP gradually decreases to
18.22/14.17/13.28.

2. If the result of your KITTI test set is also selecting the last epoch and submitting it to the official benchmark?

Thanks for your interesting work.

But "Segmentation fault" occur at "x = self.conv_input(input_sp_tensor)", i.e. "

Waiting for your helps!!!