Multi-GPU always allocates on cuda:0

InnovArul (Arul) August 31, 2018, 12:03am 2

This issue is fixed in version 0.5.0

Moving model to specific also allocates memory on GPU 0

Oh ok. I understand now. version 0.4.1: Maybe, pytorch uses cuda:0 always by default? pytorch master 0.5.0a0+ab6afc2: I have verified that this issue is fixed. Sorry for many edits of the answer

GPU utilization in Pytorch 0.4

show post in topic

Home
Categories
FAQ/Guidelines
Terms of Service
Privacy Policy

Powered by Discourse, best viewed with JavaScript enabled