error in update_parameters
chestnut-zhao opened this issue · comments
CUDA out of memory when runing update_parameters
File "/data/med-NLP/li-BERT-NER-Pytorch-master/metaST/modeling_meta_bert.py", line 339, line 339, in update_parameters
create_graph=not first_order, allow_unused=True)
File "/home/anaconda3/envs/wlhtc/lib/python3.7/site-packages/torch/autograd/init.py", line 228, in grad
inputs, allow_unused, accumulate_grad=False)
RuntimeError: CUDA out of memory. Tried to allocate 48.00 MiB (GPU 0; 10.76 GiB total capacity; 9.44 GiB already allocated; 5.12 MiB free; 9.61 GiB reserved in total by PyTorch)