You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Even when using 150 pre annotated images to label 10 additional images, CUDA memory finishes on a 3090 GPU.
torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate 31.36 GiB (GPU 0; 24.00 GiB total capacity; 49.21 GiB already allocated; 0 bytes free; 49.35 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
It stops working after about 50 pre annotated images due to this error.
The text was updated successfully, but these errors were encountered:
brr53
changed the title
CUDA memory finishes quickly when using larger pre-annotated data
Memory Leak? CUDA memory finishes quickly when using more than 50 pre-annotated data
Sep 30, 2023
Even when using 150 pre annotated images to label 10 additional images, CUDA memory finishes on a 3090 GPU.
torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate 31.36 GiB (GPU 0; 24.00 GiB total capacity; 49.21 GiB already allocated; 0 bytes free; 49.35 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
It stops working after about 50 pre annotated images due to this error.
The text was updated successfully, but these errors were encountered: