unable to install MONAI Label plugin from Slicer Extension Manager

Support
1

Operating system: Windows 10
Slicer version: 4.11.20210226 r29738 / 7a593c8
Expected behavior: to install plugin View → Extension Manager → Active Learning → MONAI Label
Actual behavior: but In extension manager Active learning module is not found.

2

As it is described in the MONAILabel installation instructions, you must use the latest Slicer Preview Release. MONAILabel is not available for Slicer-4.11.20210226.

1 Like
3

Thank you Andras for your reply!
Now, when I am trying to annotate then getting following error
‘RuntimeError: Attempting to deserialize object on a CUDA device but torch.cuda.is_available() is False. If you are running on a CPU-only machine, please use torch.load with map_location=torch.device(‘cpu’) to map your storages to the CPU.’

So, Could you please tell me how I can set this value in my cpu based system.
Thanks in advance!

1 Like
4

@diazandr3s could you help with this question?

1 Like
5

Thanks, @lassoan for ping me :slight_smile:
@poonam1 thanks for your interest in MONAI Label.
We’ve developed MONAI Label to work on a computer with a GPU. Even a small one. They are important because the training/inference will be very slow otherwise.

1 Like
6

Hi @diazandr3s thank you for response!
I get that it is for GPU based system, but I am willing to try on my CPU based system once and then I will transfer that to GPU one.
So please if you can help me setting up on my system, then it would be really grateful.
Thanks,

1 Like
7

Some inference tasks work perfectly well on a CPU. For example ultrasound segmentation - the images are quite small and just 2D. Does MONAILabel enforces the presence of a GPU or a model/sample app imposes this limitation?

1 Like
8

The type of device used for inference and training is a variable configuration that is set by default to GPU. Depending on which MONAI Label App (i.e. DeepEdit, DeepGrow or segmentation), you should change that … for instance, if you want to train any App, you should change this to CPU (MONAILabel/basic_train.py at main · Project-MONAI/MONAILabel · GitHub)
If you want to do inference in the “segmentation” App, you should change this line: MONAILabel/infer.py at ca9ffe553889f0ea9d99e0133c359851a5ea2003 · Project-MONAI/MONAILabel · GitHub

If using a pre-trained model in the segmentation App, don’t forget to also change this line (MONAILabel/infer.py at ca9ffe553889f0ea9d99e0133c359851a5ea2003 · Project-MONAI/MONAILabel · GitHub) so the model is loaded to work on the CPU.

1 Like