RuntimeError: No shared folder available

[qumengxue]

When I run python run_with_submitit.py --dataset_config configs/refcoco.json --batch_size 4 --load https://zenodo.org/record/4721981/files/pretrained_resnet101_checkpoint.pth?download=1 --ngpus 1 --nodes 2 --ema --text_encoder_lr 1e-5 --lr 5e-5, the following error occurred:
Traceback (most recent call last):
File "run_with_submitit.py", line 171, in
main()
File "run_with_submitit.py", line 130, in main
args.job_dir = get_shared_folder(args) / "%j"
File "run_with_submitit.py", line 41, in get_shared_folder
raise RuntimeError("No shared folder available")
RuntimeError: No shared folder available
How to deal with it?

[toku-n]

Hi @qumengxue,
How did you solved this problem?
It would be helpful if you share your solution.

[qumengxue]

Hi @qumengxue,
How did you solved this problem?
It would be helpful if you share your solution.

The "coco_path" in the .json file in ./config folder need to be modified to the correct path. It seems to be the only change I do for solving this problem.

[alcinos]

@toku-n you'll likely need to also adapt the default experimental path to something that works on your cluster.
See

mdetr/run_with_submitit.py

Lines 37 to 41 in 0e3c73f

	if Path("/checkpoint/").is_dir():
	p = Path(f"/checkpoint/{user}/experiments")
	p.mkdir(exist_ok=True)
	return p
	raise RuntimeError("No shared folder available")

This will fail if /checkpoint/{user}/experiments is not a writable path. Just change to where you want to save the experiments.

[toku-n]

Hi @qumengxue @alcinos
Thank you for your comment! The problem is solved.

[TopCoder2K]

@alcinos, thank you for your answer, but when I changed it into this:

user = os.getenv("USER")

if Path("/home/my_username/MDETR/mdetr/checkpoint/").is_dir():
    p = Path(f"/home/my_username/MDETR/mdetr/checkpoint/{user}/experiments")
    p.mkdir(exist_ok=True)

I got the following error:

Traceback (most recent call last):
  File "run_with_submitit.py", line 171, in <module>
    main()
  File "run_with_submitit.py", line 130, in main
    args.job_dir = get_shared_folder(args) / "%j"
  File "run_with_submitit.py", line 39, in get_shared_folder
    p.mkdir(exist_ok=True)
  File "/home/pchelintsev/anaconda3/envs/mdetr_env/lib/python3.8/pathlib.py", line 1288, in mkdir
    self._accessor.mkdir(self, mode)
FileNotFoundError: [Errno 2] No such file or directory: '/home/my_username/MDETR/mdetr/checkpoint/my_username/experiments'

Only creating this directory solved the problem.