How to compile lds module on CUDA 12.0

stemylonas / DeepSurf

A surface-based deep learning approach for the prediction of ligand binding sites on proteins

GNU Affero General Public License v3.0

37 stars 9 forks source link

How to compile lds module on CUDA 12.0 #15

Closed mainguyenanhvu closed 1 year ago

mainguyenanhvu commented 1 year ago

I am using the system:

GPU: V100 Tesla
OS:  SMP Debian 5.10.149-2 (2022-10-21) x86_64 GNU/Linux
Cuda compilation tools, release 12.0, V12.0.76
Build cuda_12.0.r12.0/compiler.31968024_0

Please show me the way to install it on my system. Thank you very much.

mainguyenanhvu commented 1 year ago

I installed multiple cudatoolkit by using https://pytorch.org/get-started/previous-versions/