Skip to content

This issue was moved to a discussion.

You can continue the conversation there. Go to discussion →

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Cannot run Auto3DSeg Two-Minute "Hello-World" Example because "dataset is not highly anisotropic" error #1902

Closed
AlexandreLaborde opened this issue Dec 17, 2024 · 1 comment

Comments

@AlexandreLaborde
Copy link

Describe the bug
I am trying to run the most basic hello_world tutorial and no matter what I do, I cannot get it to run the model.
All I did is follow the instructions to install all the dependencies and then copy and the notebook for the A Two-Minute "Hello-World" Example from https://github.com/Project-MONAI/tutorials/blob/main/auto3dseg/notebooks/auto3dseg_hello_world.ipynb and it failed on the runner.run() operation.

2024-12-17 17:26:06,597 - INFO - Running data analysis...
2024-12-17 17:26:06,613 - INFO - Found 1 GPUs for data analyzing!
100%|██████████| 12/12 [00:06<00:00,  1.78it/s]
2024-12-17 17:26:13,386 - INFO - Writing data stats to c:\Users\dsl\Documents\repos\monai_tests\helloworld_work_dir\datastats.yaml.
2024-12-17 17:26:13,388 - INFO - Writing by-case data stats to c:\Users\dsl\Documents\repos\monai_tests\helloworld_work_dir\datastats_by_case.yaml, this may take a while.
2024-12-17 17:26:13,414 - INFO - BundleGen from https://github.com/Project-MONAI/research-contributions/releases/download/algo_templates/e4cf5a1.tar.gz
algo_templates.tar.gz: 104kB [00:00, 215kB/s]                              
2024-12-17 17:26:13,911 - INFO - Downloaded: C:\Users\dsl\AppData\Local\Temp\tmpm7dibba1\algo_templates.tar.gz
2024-12-17 17:26:13,911 - INFO - Expected md5 is None, skip md5 check for file C:\Users\dsl\AppData\Local\Temp\tmpm7dibba1\algo_templates.tar.gz.
2024-12-17 17:26:13,911 - INFO - Writing into directory: c:\Users\dsl\Documents\repos\monai_tests\helloworld_work_dir.
2024-12-17 17:26:14,067 - INFO - Generated:c:\Users\dsl\Documents\repos\monai_tests\helloworld_work_dir\dints_0
2024-12-17 17:26:14,102 - INFO - Generated:c:\Users\dsl\Documents\repos\monai_tests\helloworld_work_dir\segresnet_0
2024-12-17 17:26:14,116 - INFO - segresnet2d_0 is skipped! SegresNet2D is skipped due to median spacing of [1.0, 1.0, 1.0],  which means the dataset is not highly anisotropic, e.g. spacing[2] < 3*(spacing[0] + spacing[1])/2) .
2024-12-17 17:26:14,150 - INFO - Generated:c:\Users\dsl\Documents\repos\monai_tests\helloworld_work_dir\swinunetr_0
2024-12-17 17:26:14,166 - INFO - The keys num_warmup_epochs cannot be found in the c:\Users\dsl\Documents\repos\monai_tests\helloworld_work_dir\dints_0\configs\hyper_parameters.yaml for training. Skipped overriding key num_warmup_epochs.
2024-12-17 17:26:14,166 - INFO - ['python', 'c:\\Users\\dsl\\Documents\\repos\\monai_tests\\helloworld_work_dir\\dints_0\\scripts\\train.py', 'run', "--config_file='c:/Users/dsl/Documents/repos/monai_tests/helloworld_work_dir/dints_0/configs/hyper_parameters.yaml,c:/Users/dsl/Documents/repos/monai_tests/helloworld_work_dir/dints_0/configs/hyper_parameters_search.yaml,c:/Users/dsl/Documents/repos/monai_tests/helloworld_work_dir/dints_0/configs/network.yaml,c:/Users/dsl/Documents/repos/monai_tests/helloworld_work_dir/dints_0/configs/network_search.yaml,c:/Users/dsl/Documents/repos/monai_tests/helloworld_work_dir/dints_0/configs/transforms_infer.yaml,c:/Users/dsl/Documents/repos/monai_tests/helloworld_work_dir/dints_0/configs/transforms_train.yaml,c:/Users/dsl/Documents/repos/monai_tests/helloworld_work_dir/dints_0/configs/transforms_validate.yaml'", '--training#num_epochs_per_validation=1', '--training#num_images_per_batch=2', '--training#num_epochs=2']

I tried the 'auto3dseg/notebooks/auto_runner.ipynb' notebook next and it also cannot run.

Looking at the output, in both tutorials all the datasets are skipped due to "dataset is not highly anisotropic". Following this, the runner cannot train as there are no datasets left.

Considering I did not change any of the tutorial code, what am I doing wrong?

Environment (please complete the following information):

  • OS : Windows 11 Pro 24H2
  • Python version : 3.11.9
  • MONAI version [e.g. git commit hash] / - CUDA/cuDNN version
    [ from print_config() ]
MONAI version: 1.4.0
Numpy version: 1.26.4
Pytorch version: 2.5.1+cu124
MONAI flags: HAS_EXT = False, USE_COMPILED = False, USE_META_DICT = False
MONAI rev id: 46a5272196a6c2590ca2589029eed8e4d56ff008
MONAI __file__: c:\Users\<username>\Documents\repos\monai_tests\.venv\Lib\site-packages\monai\__init__.py

Optional dependencies:
Pytorch Ignite version: 0.4.11
ITK version: 5.4.0
Nibabel version: 5.3.2
scikit-image version: 0.25.0
scipy version: 1.14.1
Pillow version: 11.0.0
Tensorboard version: 2.18.0
gdown version: 5.2.0
TorchVision version: 0.20.1+cu124
tqdm version: 4.67.1
lmdb version: 1.5.1
psutil version: 6.1.0
pandas version: 2.2.3
einops version: 0.8.0
transformers version: 4.47.1
mlflow version: 2.19.0
pynrrd version: 1.1.1
clearml version: 1.17.0rc0
  • GPU models and configuration : RTX 4070Ti

Thanks in advance!

@KumoLiu
Copy link
Contributor

KumoLiu commented Dec 19, 2024

Hi @AlexandreLaborde, thanks for your interest here.
As mentioned here: Project-MONAI/MONAI#6821 (comment)
The current version of Auto3DSeg algorithm use a Linux library and can't be used on Windows at this point.

@Project-MONAI Project-MONAI locked and limited conversation to collaborators Dec 19, 2024
@KumoLiu KumoLiu converted this issue into discussion #1904 Dec 19, 2024

This issue was moved to a discussion.

You can continue the conversation there. Go to discussion →

Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants