I think a recent update with either Colab or Cuda is throwing off the YOLOv4 model. I have previously built this notebook and trained a complete model with ...
CUDA Error: out of memory darknet: ./src/cuda.c:36: check_error: Assertio `0' failed. Need to modify the parameters of subdivision in the model cfg file ...
So when you try to execute the training, and you don't have enough free CUDA memory available, then the framework you're using throws this out of memory error.
03.08.2021 · I think a recent update with either Colab or Cuda is throwing off the YOLOv4 model. I have previously built this notebook and trained a complete model with it with no problems but now when I run the exact same code, with no changes, I get this problem: 672 x 672 try to allocate additional workspace_size = 65.03 MB CUDA allocate done!
17.02.2021 · I got this Error: RuntimeError: CUDA out of memory. GPU 0; 1.95 GiB total capacity; 1.23 GiB already allocated 1.27 GiB reserved in total by PyTorch. But it is not out of memory, it …
Inside your yolov4_custom.cfg or the any cfg that you were using, you need to change the subdivisions to match your GPU performance. For example, for 32 GB GPU-VRAM, set …