thuml/CDAN

Cannot train pytorch version on multi-gpu

chester256 opened this issue · 1 comments

When trained on multi-gpu, the transfer loss (CDAN + Entropy loss) is very low and the test accuracy alway decreases. Dataset: visda2017
tested on Ubuntu 18.04, Pytorch version: 1.2.0 / 1.4.0, GPU: GTX 1080 / RTX 2080Ti

我也会出现这样的情况