RuntimeError: expected backend CPU and dtype Double but got backend CPU and dtype Float

fdjingyuan / Deep-Fashion-Analysis-ECCV2018

Codes of ECCV 2018 workshop paper "Deep Fashion Analysis with Feature Map Upsampling and Landmark-driven Attention"

84 stars 27 forks source link

Closed ahmedselim2017 closed 5 years ago

ahmedselim2017 commented 5 years ago

When I run train code (python -m src.train --conf src.conf.whole) I get this error.

ahmedselim2017 commented 5 years ago

I got it. It was because of my PyTorch and CUDA version. When I downgrade PyTorch version to 0.4.1 and CUDA version to 9 the problem was solved.