Facing "Failed to build agent SeamlessStreaming" error
#12
by
wanderer797
- opened
I am facing "CUDA out of memory. Tried to allocate 20.00 MiB. GPU 0 has a total capacty of 3.81 GiB of which 12.62 MiB is free. Including non-PyTorch memory, this process has 3.38 GiB memory in use. Of the allocated memory 3.32 GiB is allocated by PyTorch, and 2.76 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF" error.
My GPU specs is: NVIDIA GeForce GTX 1650 Ti (4 GB memory)
What is to be done in this case.
Do i need to set "max_split_size_mb"? or reduce the batch size? Also where can I find the file where I can set split size in my system. As I am running on Ubuntu