RutimeError:CUDA out of memory. when i training on COCO train2017 dataset

Question

RutimeError:CUDA out of memory. when i training on COCO train2017 dataset

houshuaishuai opened this issue 3 years ago · 2 comments

when i run python tools/train.py --cfg experiments/coco/w32/w32_4x_reg03_bs10_512_adam_lr1e-3_coco_x140.yaml
RuntimeError: CUDA out of memory. Tried to allocate 86.00 MiB (GPU 0; 10.91 GiB total capacity; 9.39 GiB already allocated; 125.50 MiB free; 9.48 GiB reserved in total by PyTorch) , i didn't find batch_size in the w32_4x_reg03_bs10_512_adam_lr1e-3_coco_x140.yaml? is there anything else that need to be modified?

Answer 1 · 2022-06-01T11:41:01.000Z

Images per GPU (for training) can be lowered in that file as a solution

Answer 2 · 2023-10-30T01:12:28.000Z

I had the same problem as you, did you solve it?