Problem extracting multi-scale features
etrulls opened this issue · 3 comments
etrulls commented
It crashes here, when operating on a byte() tensor and a boolean tensor. Works if you convert detections to .byte(). It was working the last time I tried, it could be a problem with library versions? I'm on torch 1.2.0 now. FYI.
mihaidusmanu commented
Thanks a lot for bringing this to my attention. I'm still on PyTorch 1.1 personally, but I'll upgrade and release a fix shortly.
mihaidusmanu commented
Fixed in the latest commit - PyTorch 1.2 introduced boolean tensors so VRAM usage should also be slightly lower now. Sorry for the delay.
etrulls commented
Pretty fast IMO, thanks. :)