When loading a classifier that ships with SleepECG, a bunch of warnings are printed. I wonder if these warnings indicate real problems (in which case we should fix them) or if they can be ignored (in which case we should silence them).
I tried changing the logger level after we import tensorflow, but that didn't seem to do anything. Setting the environment variable here (before the import) does get rid of all warnings except one message:
import os
os.environ["TF_CPP_MIN_LOG_LEVEL"] = "2"
The remaining warning is an error:
2023-05-02 12:06:33.352056: E tensorflow/compiler/xla/stream_executor/cuda/cuda_driver.cc:266] failed call to cuInit: CUDA_ERROR_NO_DEVICE: no CUDA-capable device is detected
So setting the level to "3" gets rid of that one too.
When loading a classifier that ships with SleepECG, a bunch of warnings are printed. I wonder if these warnings indicate real problems (in which case we should fix them) or if they can be ignored (in which case we should silence them).
WDYT @hofaflo?
I tried changing the logger level after we import
tensorflow
, but that didn't seem to do anything. Setting the environment variable here (before the import) does get rid of all warnings except one message:The remaining warning is an error:
So setting the level to
"3"
gets rid of that one too.