CudaHostAlloc takes a lot of time during training #124456
Labels
module: cuda
Related to torch.cuda, and CUDA support in general
module: CUDACachingAllocator
module: dataloader
Related to torch.utils.data.DataLoader and Sampler
triaged
This issue has been looked at a team member, and triaged and prioritized into an appropriate module
馃悰 Describe the bug
set: pin_memory =True in Dataloader
Versions
pytorch 1.13
numpy 1.21
python3.8
cc @ssnl @VitalyFedyunin @ejguan @dzhulgakov @ptrblck
The text was updated successfully, but these errors were encountered: