PyTorch使用tensorboard可以顯示網(wǎng)絡(luò)運行情況，但偶爾使用SummaryWriter時遇到Segmentation fault錯誤。
利用python3的faulthandler，可定位到出錯的代碼行，具體操作有兩種方式如下：
(1) 在代碼中寫入faulthandler

import faulthandler
# 在import之后直接添加以下啟用代碼即可
faulthandler.enable()
# 后邊正常寫你的代碼

(2)直接通過命令行來啟用，運行時添加-X faulthandler參數(shù)即可：

python -X faulthandler your_script.py

錯誤代碼：

import xxx
from torch.utils.tensorboard import SummaryWriter
import yyy
log_dir = './run_logs'
self.writer = SummaryWriter(log_dir)

報錯如下：

 File "/opt/conda/lib/python3.7/site-packages/tensorboard/compat/__init__.py", line 45 in tf
  File "/opt/conda/lib/python3.7/site-packages/tensorboard/lazy.py", line 50 in load_once
  File "/opt/conda/lib/python3.7/site-packages/tensorboard/lazy.py", line 97 in wrapper
  File "/opt/conda/lib/python3.7/site-packages/tensorboard/lazy.py", line 65 in __getattr__
  File "/opt/conda/lib/python3.7/site-packages/tensorboard/summary/writer/event_file_writer.py", line 72 in __init__
  File "/opt/conda/lib/python3.7/site-packages/torch/utils/tensorboard/writer.py",line 66 in __init__
  File "/opt/conda/lib/python3.7/site-packages/torch/utils/tensorboard/writer.py",line 256 in _get_file_writer
  File "/opt/conda/lib/python3.7/site-packages/torch/utils/tensorboard/writer.py", line 225 in __init__

經(jīng)過查詢，發(fā)現(xiàn)極有可能是導(dǎo)包(from torch.utils.tensorboard import SummaryWriter)順序?qū)е碌膯栴}。參考鏈接：[TensorBoard] The different order of import SummaryWriter may cause Segmentation fault

將from torch.utils.tensorboard import SummaryWriter順序調(diào)前，可以順利導(dǎo)包和實例化了。

import xxx
from torch.utils.tensorboard import SummaryWriter
import yyy

改為

from torch.utils.tensorboard import SummaryWriter
import xxx
import yyy

到此這篇關(guān)于PyTorch使用tensorboard的SummaryWriter報錯的文章就介紹到這了,更多相關(guān)PyTorch SummaryWriter報錯內(nèi)容請搜索腳本之家以前的文章或繼續(xù)瀏覽下面的相關(guān)文章希望大家以后多多支持腳本之家！

最新評論