Pytorch——谜之bug汇总

运行别人的代码的时候报错：RuntimeError: CUDA out of memory. Tried to allocate 512.00 MiB (GPU 0; 11.90 GiB total capacity; 11.15 GiB already allocated; 46.94 MiB free; 241.57 MiB cached) 

但是nvidia-smi查看显卡，显存充足   

解决：batch改小一点，本来代码里是8，我觉得够小了。。改成4就跑起来了。。

-------------------------------------------------------------------------------------------------------------------------------------------------