How to run in GPU? - Githubissues

mkocabas / VIBE

Official implementation of CVPR2020 paper "VIBE: Video Inference for Human Body Pose and Shape Estimation"

https://arxiv.org/abs/1912.05656

Other

2.88k stars 548 forks source link

How to run in GPU? #236

Closed lakpa-tamang9 closed 2 years ago

lakpa-tamang9 commented 2 years ago

Running demo.py automatically runs on CPU and takes huge amount of time. At the end, it throws cuda runtime error. RuntimeError: cuda runtime error (209) : no kernel image is available for execution on the device at /pytorch/aten/src/THCUNN/generic/LeakyReLU.cu:29 In the readme, there in no any instruction provided about usage of CPU for inference.

mrpositron commented 2 years ago

Hey! How did you solved it?

baolin-hri commented 2 years ago

update pytorch and cudatoolkit works for me. conda install pytorch==1.8.0 torchvision==0.9.0 torchaudio==0.8.0 cudatoolkit=11.1 -c pytorch -c conda-forge