48G 显存跑不起来，提示显存不足 #79
Description
torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 32.00 MiB. GPU 0 has a total capacity of 47.40 GiB of which 35.69 MiB is free. Process 3307 has 0 bytes memory in use. Including non-PyTorch memory, this process has 0 bytes memory in use. Of the allocated memory 21.52 GiB is allocated by PyTorch, and 9.81 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)
[rank1]:[W109 18:46:15.961672379 ProcessGroupNCCL.cpp:1496] Warning: WARNING: destroy_process_group() was not called before program exit, which can leak resources. For more info, please see https://pytorch.org/docs/stable/distributed.html#shutdown (function operator())
[rank0]:[W109 18:46:15.362591747 ProcessGroupNCCL.cpp:1496] Warning: WARNING: destroy_process_group() was not called before program exit, which can leak resources. For more info, please see https://pytorch.org/docs/stable/distributed.html#shutdown (function operator())
W0109 18:46:16.524000 3177 site-packages/torch/distributed/elastic/multiprocessing/api.py:897] Sending process 3306 closing signal SIGTERM
E0109 18:46:16.993000 3177 site-packages/torch/distributed/elastic/multiprocessing/api.py:869] failed (exitcode: 1) local_rank: 1 (pid: 3307) of binary: /usr/local/anaconda3/envs/longcat-video/bin/python3.10
跑官方 demo 失败。 求解