Hello! Thanks for sharing the code of Seeing Glass.But I have some trouble when running the code.I really hope you can help me solve the problem, and I am looking forward to your reply! Thanks！
The problems are as following：
1)The functions on line98 and line100 do not exist in train_franka.py.And when I ran train_franka.py,I got error as follow:
Traceback (most recent call last):
File "tools/train_franka.py", line 171, in
main()
File "tools/train_franka.py", line 67, in main
cfg = get_default_config(args.default_cfg)
File "/home/qjn/project/TranspareNet/saic_depth_completion/config/init.py", line 12, in get_default_config
return CONFIGS[type].clone()
KeyError: 'arch0'
2)According to the error which I got by running pccRunner.py directly,I created a new folder named frankascanv2 .Then I put TODD dataset in ./datasets/frankascanv2/.When I ran pccRunner.py,it occured error as follow:
[Open3D WARNING] Read PCD failed: unable to open file: ./datasets/frankascanv2/train/1617657685.630454/Ground_Truth_recenter_0.pcd
3)When I ran inference_transparenet.py ,I got error as follow:
Let's use 6 GPUs!
/home/qjn/software/miniconda3/envs/TranspareNet/lib/python3.7/site-packages/torch/nn/parallel/data_parallel.py:30: UserWarning:
There is an imbalance between your GPUs. You may want to exclude GPU 4 which
has less than 75% of the memory or cores of GPU 0. You can do so by setting
the device_ids argument to DataParallel, or by setting the CUDA_VISIBLE_DEVICES
environment variable.
warnings.warn(imbalance_warn.format(device_ids[min_pos], device_ids[max_pos]))
2022-04-14 23:02:57 - saic-dc: Snapshoter's arg 'save_dir' was not passed. Using default value(./).
2022-04-14 23:02:57 - saic-dc: No snapshot found. Initializing model from scratch
ROOT grnet_point_cloud_completion/datasets/frankascanv2
{'val_frankascanv2': <torch.utils.data.dataloader.DataLoader object at 0x7f7b58c25510>}
checkpoints/frankav2-ckpt-epoch-100.pth
Traceback (most recent call last):
File "inference_transparenet.py", line 103, in
main()
File "inference_transparenet.py", line 81, in main
checkpoint = torch.load(cfg.CONST.WEIGHTS)
File "/home/qjn/software/miniconda3/envs/TranspareNet/lib/python3.7/site-packages/torch/serialization.py", line 699, in load
with _open_file_like(f, 'rb') as opened_file:
File "/home/qjn/software/miniconda3/envs/TranspareNet/lib/python3.7/site-packages/torch/serialization.py", line 231, in _open_file_like
return _open_file(name_or_buffer, mode)
File "/home/qjn/software/miniconda3/envs/TranspareNet/lib/python3.7/site-packages/torch/serialization.py", line 212, in init
super(_open_file, self).init(open(name, mode))
FileNotFoundError: [Errno 2] No such file or directory: 'checkpoints/frankav2-ckpt-epoch-100.pth'
4）I want to know what is the relationship between pccRunner, img2pcd, pcd2img, train_franka and inference_transparenet, and how can I correctly understand and use them.
Hello! Thanks for sharing the code of Seeing Glass.But I have some trouble when running the code.I really hope you can help me solve the problem, and I am looking forward to your reply! Thanks！
The problems are as following：
1)The functions on line98 and line100 do not exist in train_franka.py.And when I ran train_franka.py,I got error as follow: Traceback (most recent call last): File "tools/train_franka.py", line 171, in
main()
File "tools/train_franka.py", line 67, in main
cfg = get_default_config(args.default_cfg)
File "/home/qjn/project/TranspareNet/saic_depth_completion/config/init.py", line 12, in get_default_config
return CONFIGS[type].clone()
KeyError: 'arch0'
2)According to the error which I got by running pccRunner.py directly,I created a new folder named frankascanv2 .Then I put TODD dataset in ./datasets/frankascanv2/.When I ran pccRunner.py,it occured error as follow: [Open3D WARNING] Read PCD failed: unable to open file: ./datasets/frankascanv2/train/1617657685.630454/Ground_Truth_recenter_0.pcd
3)When I ran inference_transparenet.py ,I got error as follow: Let's use 6 GPUs! /home/qjn/software/miniconda3/envs/TranspareNet/lib/python3.7/site-packages/torch/nn/parallel/data_parallel.py:30: UserWarning: There is an imbalance between your GPUs. You may want to exclude GPU 4 which has less than 75% of the memory or cores of GPU 0. You can do so by setting the device_ids argument to DataParallel, or by setting the CUDA_VISIBLE_DEVICES environment variable. warnings.warn(imbalance_warn.format(device_ids[min_pos], device_ids[max_pos])) 2022-04-14 23:02:57 - saic-dc: Snapshoter's arg 'save_dir' was not passed. Using default value(./). 2022-04-14 23:02:57 - saic-dc: No snapshot found. Initializing model from scratch ROOT grnet_point_cloud_completion/datasets/frankascanv2 {'val_frankascanv2': <torch.utils.data.dataloader.DataLoader object at 0x7f7b58c25510>} checkpoints/frankav2-ckpt-epoch-100.pth Traceback (most recent call last): File "inference_transparenet.py", line 103, in
main()
File "inference_transparenet.py", line 81, in main
checkpoint = torch.load(cfg.CONST.WEIGHTS)
File "/home/qjn/software/miniconda3/envs/TranspareNet/lib/python3.7/site-packages/torch/serialization.py", line 699, in load
with _open_file_like(f, 'rb') as opened_file:
File "/home/qjn/software/miniconda3/envs/TranspareNet/lib/python3.7/site-packages/torch/serialization.py", line 231, in _open_file_like
return _open_file(name_or_buffer, mode)
File "/home/qjn/software/miniconda3/envs/TranspareNet/lib/python3.7/site-packages/torch/serialization.py", line 212, in init
super(_open_file, self).init(open(name, mode))
FileNotFoundError: [Errno 2] No such file or directory: 'checkpoints/frankav2-ckpt-epoch-100.pth'
4）I want to know what is the relationship between pccRunner, img2pcd, pcd2img, train_franka and inference_transparenet, and how can I correctly understand and use them.