Let me know version information by CUDA, CUDNN and tensorflow_gpu that recommended in 3D-Slicer 5.0.3 stable version for modeling segmentation of monai label
System information of my laptop is as following.
[display]
Operating System: Windows 10 Home, 64-bit
DirectX Version: 12.0
GPU Processor: NVIDIA GeForce RTX 3070 Laptop GPU
Driver Version: 471.51
Driver Type: DCH
Direct3D feature level: 12_1
CUDA Core: 5120
Resizable BAR Yes
Max-Q technology 3rd generation
Dynamic Boost 2.0 Yes
WhisperMode 2.0 Yes
Advanced Optimus No
Maximum graphics performance 100 W
Core clock: 1290 MHz
Memory data rate: 12.00 Gbps
Memory interface: 256-bit
Memory bandwidth: 384.06 GB/s
Total available graphics memory: 16079 MB
Dedicated video memory: 8192 MB GDDR6
System video memory: 0 MB
Shared system memory: 7887 MB
Video BIOS Version: 94.04.4F.00.19
IRQ: Not used
Bus: PCI Express x8 Gen3
Device Id: 10DE 249D 20141A58
Part number: 4735 0010
Please let me know version information by CUDA, CUDNN and tensorflow_gpu that recommended in 3D-Slicer 5.0.3 stable version for modeling segmentation of monai label.
For working 3D-Slicer 5.0.3 stable version based on windows 11 64bit OS , I installed in (my laptop) pc as following.
Python 3.9
monailabel 0.5.2
monai 1.0.1
torch 1.1.3
tensorflow-gpu 2.1.0
CUDA 11.3
CUDNN 8.6
And then using PyPi, I conducted command that is ‘python -c "import torch; print(torch.cuda.is_available())’.
The result shows ‘False’.
I think that it can’t conduct torch (with cuda) using gpu.
What should I do for showing ‘False’ into ‘True’ in result of ‘python -c "import torch; print(torch.cuda.is_available())’?
Your laptop with an RTX 3070 GPU should definitely be ready to enable CUDA and use MONIALabel in GPU mode. We have set up a MONAILabel segmentation project during the last 3D Slicer Project week:
On that page is a link How To Setup MONAILabel in Windows 11 that should help you with installation. Please follow this step by step, we’ll assist if something does not work.
when I installed ‘python –m torch-1.7.1-cp39-cp39-win_amd64.whl’, the result of ‘python -c "import torch; print(torch.cuda.is_available())’ shows ‘True’.