Said earlier, I am in the model test and not appear this error when training, as for the model training report this error, please refer to my another blog: runtimeerror: about errors CUDA out of memory. The Tried to the allocate 1.17 GB
actually very simple solution, originally I procedures specify the gpu is 3, run the test code is called the title out of memory error, the diagram below: 
I specified the gpus as 2 and 3, and then ran the code without error.
Read More:
- Runtimeerror using Python training model: CUDA out of memory error resolution
- RuntimeError: CUDA out of memory. Tried to allocate 600.00 MiB (GPU 0； 23.69 GiB total capacity)
- MobaXterm error cuda:out of memory
- CUDA error:out of memory
- RuntimeError: cuda runtime error (100) : no CUDA-capable device is detected at /opt/conda/conda-bld/
- Python: CUDA error: an illegal memory access was accounted for
- Out of memory overflow solution for idea running error
- Idea pop-up window out of memory, modify the parameters and start the no response solution
- [Solved] RuntimeError: CUDA error: CUBLAS_STATUS_ALLOC_FAILED when calling `cublasCreate(handle)`
- Fatal error: Newspace:: rebalance allocation failed – process out of memory (memory overflow)
- Successfully solved runtimeerror: CUDA runtime error (30)
- CheXNet-master: CUDA out of memery [How to Solve]
- The difference, cause and solution of memory overflow and memory leak
- PyTorch CUDA error: an illegal memory access was encountered
- RuntimeError: CUDA error: CUBLAS_STATUS_ALLOC_FAILED when calling `cublasCreate(handle)`
- [PostgreSQL tutorial] · out of memory issue
- (29)RuntimeError: cuda runtime error (999)
- RuntimeError:cuda runtime error (11) : invalid argument at /pytorch/aten/src/THC/generic
- Oracle 11g installation prompt ora-27102: out of memory
- Error: 701 use DAC connection when out of memory