SemTorch icon indicating copy to clipboard operation
SemTorch copied to clipboard

Published 20 hours ago • verified publisher icon WaterKnight1998

unet + resnet error

Open tiagojcarvalho opened this issue 3 years ago • 2 comments

Hi,

when using get_segmentation_learner function with architecture_name="unet" and backbone_name="resnet50" I'm getting "RuntimeError: cuDNN error: CUDNN_STATUS_NOT_INITIALIZED" error. I'm using torch 1.7.1, torchvision 0.8.2 and cuda 10.2 because of fastai compatibility.

Do you have some idea why this happens ?

tiagojcarvalho avatar Apr 28 '21 18:04 tiagojcarvalho

Hi,

It looks like CUDA is not porperly installed. What is your GPU model?

WaterKnight1998 avatar Apr 28 '21 18:04 WaterKnight1998

GeForce GTX 1080 Ti

tiagojcarvalho avatar Apr 28 '21 23:04 tiagojcarvalho