小白求教 训练模型 刚开始正常 两三秒之后就报错RuntimeError: CUDA out of memory.

小白求教 训练模型 刚开始正常 两三秒之后就报错

- RuntimeError: CUDA out of memory. Tried to allocate 20.00 MiB (GPU 0; 2.00 GiB total capacity; 1.02 GiB already allocated; 3.96 MiB free; 1.06 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF

然后便自行终止 求大神帮忙看看 怎样解决

Start with the reported CUDA out-of-memory message and the linked PyTorch CUDA memory-management guidance. Reproduce the training run only after collecting the model, batch-size, and GPU details that are missing here; done means a confirmed cause and a documented, reproducible mitigation.

Written by the indexing model from the issue text.