-
Notifications
You must be signed in to change notification settings - Fork 152
Description
🐛 Bug description [Please make everyone to understand it]
在WSL 中使用pytorch 环境 出现获取硬件信息错误
🧑💻 Step to reproduce
swanlab: 👋 Hi JZ47321, welcome to swanlab!
swanlab: Syncing run ox-3 to the cloud
swanlab: 🌟 Run swanlab watch /home/fl/gh/nanoGPT/swanlog
to view SwanLab Experiment Dashboard locally
swanlab: 🏠 View project at https://swanlab.cn/@JZ47321/my-awesome-project
swanlab: 🚀 View run at https://swanlab.cn/@JZ47321/my-awesome-project/runs/43hry0zksv8604ry16imt
swanlab: Error happened while training
Error in sys.excepthook:
Traceback (most recent call last):
File "/home/fl/miniconda3/envs/nanogpt/lib/python3.9/site-packages/swanlab/data/callback_cloud.py", line 202, in _except_handler
get_run().finish(SwanLabRunState.CRASHED, error=self._traceback_error(tb, tp(val)))
TypeError: function takes exactly 5 arguments (1 given)
Original exception was:
Traceback (most recent call last):
File "/home/fl/gh/nanoGPT/testlog.py", line 5, in
swanlab.init(
File "/home/fl/miniconda3/envs/nanogpt/lib/python3.9/site-packages/swanlab/data/sdk.py", line 163, in init
run = register(
File "/home/fl/miniconda3/envs/nanogpt/lib/python3.9/site-packages/swanlab/data/run/init.py", line 15, in register
run = SwanLabRun(*args, **kwargs)
File "/home/fl/miniconda3/envs/nanogpt/lib/python3.9/site-packages/swanlab/data/run/main.py", line 121, in init
metadata, self.monitor_funcs = get_metadata(self.__settings.log_dir)
File "/home/fl/miniconda3/envs/nanogpt/lib/python3.9/site-packages/swanlab/data/run/metadata/init.py", line 21, in get_metadata
hardware_info, monitor_funcs = get_hardware_info()
File "/home/fl/miniconda3/envs/nanogpt/lib/python3.9/site-packages/swanlab/data/run/metadata/hardware/init.py", line 26, in get_hardware_info
nvidia = dec_hardware_func(get_nvidia_gpu_info, monitor_funcs)
File "/home/fl/miniconda3/envs/nanogpt/lib/python3.9/site-packages/swanlab/data/run/metadata/hardware/init.py", line 55, in dec_hardware_func
x, y = func()
File "/home/fl/miniconda3/envs/nanogpt/lib/python3.9/site-packages/swanlab/data/run/metadata/hardware/gpu/nvidia.py", line 41, in get_nvidia_gpu_info
gpu_name = pynvml.nvmlDeviceGetName(handle) # types: bytes | str
File "/home/fl/miniconda3/envs/nanogpt/lib/python3.9/site-packages/pynvml.py", line 2182, in wrapper
return res.decode()
UnicodeDecodeError: 'utf-8' codec can't decode byte 0xf8 in position 0: invalid start byte
swanlab: Experiment ox-3 has completed
Error in atexit._run_exitfuncs:
Traceback (most recent call last):
File "/home/fl/miniconda3/envs/nanogpt/lib/python3.9/site-packages/swanlab/data/callback_cloud.py", line 193, in _clean_handler
run.finish() if run.running else swanlog.debug("Duplicate finish, ignore it.")
File "/home/fl/miniconda3/envs/nanogpt/lib/python3.9/site-packages/swanlab/data/run/main.py", line 286, in finish
getattr(run, "_SwanLabRun__cleanup")(error)
File "/home/fl/miniconda3/envs/nanogpt/lib/python3.9/site-packages/swanlab/data/run/main.py", line 205, in __cleanup
if self.monitor_cron is not None:
AttributeError: 'SwanLabRun' object has no attribute 'monitor_cron'
👾 Expected result
Write down the results you expect
🚑 Any additional [like screenshots]
-
SwanLab Version: 0.4.0
-
Platform: wsl ubuntu-24.04 conda python 3.9.19