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

HRNet / DEKR

This is an official implementation of our CVPR 2021 paper "Bottom-Up Human Pose Estimation Via Disentangled Keypoint Regression" (https://arxiv.org/abs/2104.02300)

MIT License

437 stars 76 forks source link

RutimeError:CUDA out of memory. when i training on COCO train2017 dataset #32

Open houshuaishuai opened 2 years ago

houshuaishuai commented 2 years ago

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?

AgntBrwr commented 2 years ago

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

wusaisa commented 1 year ago

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