CUDA out of memory

[berkat28]

When running :

python -u tools/train_net.py --cfg configs/a2v/foot.yaml \
                             --model finetune-foot \
                             TRAIN.CHECKPOINT_PATH checkpoints/a2v/pretrained_res50.pth \
                             GPU_ID 0

Got error :

CUDA out of memory. Tried to allocate 1.43 GiB (GPU 0; 5.80 GiB total capacity; 2.25 GiB already allocated; 1.16 GiB free; 3.61 GiB reserved in total by PyTorch)

Is there any solutions? or i need to upgrade my GPU?

Thank you.

[hwfan]

Thanks for playing with Activity2Vec! This information means that the memory of your GPU is not enough for training the network. You should kill the redundant processes that occupy the GPU memory, or use the GPUs with larger memory instead.

[berkat28]

Thanks!!!