Grand Challenge

Sanity Check Submission Failure ¶

By: AdamFarrag on May 21, 2025, 6:16 a.m.

Dear LUNA25 organizing team,

I am trying to submit an algorithm to the sanity check phase, however, it seems to be failing with the error The algorithm failed on one or more cases.. Running the inference file and the do_test_run script works fine locally, so I am unsure where the issue lies. Would it be possible to access the logs and errors only for the sanity check phase? I understand this is not possible for the open development phase, but it would be very useful for fixing bugs in order to pass the sanity check phase.

Thanks in advance!

-Adam

Last edited by: AdamFarrag on May 21, 2025, 6:18 a.m., edited 2 times in total.

Re: Sanity Check Submission Failure ¶

By: bogdanobreja on May 21, 2025, 7:50 a.m.

Dear Adam,

This is the error we are seeing on our side: "Time limit exceeded". So the runtime of your algorithm exceeds the 5 minutes time limit.

Hope this helps.

Kind Regards, Bogdan.

Re: Sanity Check Submission Failure ¶

By: pvlima on May 27, 2025, 9:23 p.m.

Dear Bogdan,

The same just happened to me and I do not know exactly what the problem is. Is there an easier way to access the logs?

Best Regards, Pedro Lima

Re: Sanity Check Submission Failure ¶

By: bogdanobreja on May 28, 2025, 7:52 a.m.

Dear Pedro,

This is the error that we see on our side: "Output file 'lung-nodule-malginancy-likelihoods.json' was not produced".

Hope this helps!

Kind Regards, Bogdan.

Re: Sanity Check Submission Failure ¶

By: vegetable.dog on June 8, 2025, 1:46 p.m.

Dear Bogdan,

The same just happened to me and I do not know exactly what the problem is. Could you tell me the failure logs?

Best Wishes,Vegetable.

Re: Sanity Check Submission Failure ¶

By: drepeeters on June 9, 2025, 10:05 a.m.

Dear Vegetable,

It seems you do not have access to a NVIDIA Driver (GPU). See the Logs below:

2025-06-08T12:44:56.059000+00:00 =+==+==+==+==+==+==+==+==+==+=
2025-06-08T12:44:56.059000+00:00 Collecting Torch CUDA information
2025-06-08T12:44:56.059000+00:00 Torch version: 11.7
2025-06-08T12:44:56.059000+00:00 Torch CUDA is available: False
2025-06-08T12:44:56.059000+00:00 =+==+==+==+==+==+==+==+==+==+=

2025-06-08T12:44:56.059000+00:00 Traceback (most recent call last):
2025-06-08T12:44:56.059000+00:00   File "/opt/app/inference.py", line 255, in <module>
2025-06-08T12:44:56.059000+00:00     raise SystemExit(run(mode= mode,
2025-06-08T12:44:56.059000+00:00   File "/opt/app/inference.py", line 190, in run
2025-06-08T12:44:56.059000+00:00     processor = NoduleProcessor(ct_image_file=input_chest_ct,
2025-06-08T12:44:56.059000+00:00   File "/opt/app/inference.py", line 94, in __init__
2025-06-08T12:44:56.059000+00:00     self.processor = MalignancyProcessor(mode=mode, suppress_logs=True, model_name=model_name)
2025-06-08T12:44:56.059000+00:00   File "/opt/app/processor.py", line 40, in __init__
2025-06-08T12:44:56.059000+00:00     self.model_2d = ResNet18(weights=None).cuda()
2025-06-08T12:44:56.059000+00:00   File "/home/user/.local/lib/python3.10/site-packages/torch/nn/modules/module.py", line 905, in cuda
2025-06-08T12:44:56.059000+00:00     return self._apply(lambda t: t.cuda(device))
2025-06-08T12:44:56.059000+00:00   File "/home/user/.local/lib/python3.10/site-packages/torch/nn/modules/module.py", line 797, in _apply
2025-06-08T12:44:56.059000+00:00     module._apply(fn)
2025-06-08T12:44:56.059000+00:00   File "/home/user/.local/lib/python3.10/site-packages/torch/nn/modules/module.py", line 797, in _apply
2025-06-08T12:44:56.059000+00:00     module._apply(fn)
2025-06-08T12:44:56.059000+00:00   File "/home/user/.local/lib/python3.10/site-packages/torch/nn/modules/module.py", line 820, in _apply
2025-06-08T12:44:56.059000+00:00     param_applied = fn(param)
2025-06-08T12:44:56.059000+00:00   File "/home/user/.local/lib/python3.10/site-packages/torch/nn/modules/module.py", line 905, in <lambda>
2025-06-08T12:44:56.059000+00:00     return self._apply(lambda t: t.cuda(device))
2025-06-08T12:44:56.059000+00:00   File "/home/user/.local/lib/python3.10/site-packages/torch/cuda/__init__.py", line 247, in _lazy_init
2025-06-08T12:44:56.059000+00:00     torch._C._cuda_init()
2025-06-08T12:44:56.059000+00:00 RuntimeError: Found no NVIDIA driver on your system. Please check that you have an NVIDIA GPU and installed a driver from http://www.nvidia.com/Download/index.aspx

My advice would be to check if you configured your algorithm with GPU access. Check our documentation here for more information.

Please let me know if you need further information.

Kind regards,
Dre