Open KrzysztofKaluza opened 5 years ago
Problem like in title, I'm using Ubuntu 18.04, python 3.6.8 tensorflow-gpu 1.14, cuda 10.1 with cudnn 7.6.1, could it be a problem with too new libraries?
i think it will not have any effect to train or test
Any update here? I am facing the same problem. For other networks on TF, GPU is being used. Not for pointnet.
Problem like in title, I'm using Ubuntu 18.04, python 3.6.8 tensorflow-gpu 1.14, cuda 10.1 with cudnn 7.6.1, could it be a problem with too new libraries?