RuntimeError: CUDA error: out of memory，解决方案：map_location=device

weixin_45741855

已于 2023-12-31 12:18:20 修改

阅读量953

点赞数 26

文章标签：深度学习人工智能

于 2023-12-31 12:17:13 首次发布

本文链接：https://blog.csdn.net/weixin_45741855/article/details/135314845

版权

解决方案
Traceback (most recent call last):
File “/home/ljm/layoutlmv3x/train_question_answer_pair_v22_f1_0.9696.py”, line 621, in
model.load_state_dict(torch.load(model_weights_path), map_location=device)
File “/home/ljm/miniconda3/envs/layoutlmv3x/lib/python3.9/site-packages/torch/serialization.py”, line 712, in load
return _load(opened_zipfile, map_location, pickle_module, **pickle_load_args)
File “/home/ljm/miniconda3/envs/layoutlmv3x/lib/python3.9/site-packages/torch/serialization.py”, line 1046, in _load
result = unpickler.load()
File “/home/ljm/miniconda3/envs/layoutlmv3x/lib/python3.9/site-packages/torch/serialization.py”, line 1016, in persistent_load
load_tensor(dtype, nbytes, key, _maybe_decode_ascii(location))
File “/home/ljm/miniconda3/envs/layoutlmv3x/lib/python3.9/site-packages/torch/serialization.py”, line 1001, in load_tensor
wrap_storage=restore_location(storage, location),
File “/home/ljm/miniconda3/envs/layoutlmv3x/lib/python3.9/site-packages/torch/serialization.py”, line 176, in default_restore_location
result = fn(storage, location)
File “/home/ljm/miniconda3/envs/layoutlmv3x/lib/python3.9/site-packages/torch/serialization.py”, line 158, in _cuda_deserialize
return obj.cuda(device)
File “/home/ljm/miniconda3/envs/layoutlmv3x/lib/python3.9/site-packages/torch/_utils.py”, line 79, in cuda
return new_type(self.size()).copy(self, non_blocking)
RuntimeError: CUDA error: out of memory
CUDA kernel errors might be asynchronously reported at some other API call,so the stacktrace below might be incorrect.
For debugging consider passing CUDA_LAUNCH_BLOCKING=1.