Runtimeerror: CUDA out of memory (solved) [easy to understand]

Hello everyone , I meet you again , I'm your friend, Quan Jun .

Use today pytorch When training neural networks , The following error occurred ：

RuntimeError: CUDA out of memory. Tried to allocate 144.00 MiB (GPU 0; 2.00 GiB total capacity; 1.29 GiB already allocated; 79.00 MiB free; 1.30 GiB reserved in total by PyTorch)

Mingming GPU 0 Yes 2G Capacity , Why only 79M You can use ？ also 1.30G Has been PyTorch Occupied . This means that PyTorch The amount of GPU Space is not released , Cause the next run , appear CUDA out of memory.

The solution is as follows ：

（1） Create a new terminal

（2） Input nvidia-smi, Will be displayed GPU Usage situation , And occupy GPU Applications for

（3） Input taskkill -PID Process number -F End the process of occupation , such as taskkill -PID 7392 -F

（4） Input again nvidia-smi see GPU usage , Will find GPU The occupied space is greatly reduced , In this way, we can use GPU Run the program

【 Reference documents 】

CUDA out of memory.( resolved ）

Publisher ： Full stack programmer stack length , Reprint please indicate the source ：https://javaforall.cn/127655.html Link to the original text ：https://javaforall.cn