Hello! I tried to train custom YOLO model for my project.
Here is my environment
tensorflow-gpu 2.0
CUDA 10.0
cuDNN 7.4.2
GTX 1070
VS2017
I executed train file, but got this error.
> 2020-09-04 22:57:09.765952: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudart64_100.dll
> 2020-09-04 22:57:12.180849: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library nvcuda.dll
> 2020-09-04 22:57:12.207369: E tensorflow/stream_executor/cuda/cuda_driver.cc:318] failed call to cuInit: CUDA_ERROR_NO_DEVICE: no CUDA-capable device is detected
I use windows 10 and anaconda virtual environment.
I set CUDA_VISIBLE_DEVICES=‘0’.
When I load gpu list at python IDLE on virtual environment, gpu list is successfully loaded.
But still got same error.
Here is also nvidia-smi result. I don’t understand why compute processes are not supported.
Fri Sep 04 23:16:59 2020 +-----------------------------------------------------------------------------+ | NVIDIA-SMI 432.00 Driver Version: 452.06 CUDA Version: 11.0 | |-------------------------------+----------------------+----------------------+ | GPU Name TCC/WDDM | Bus-Id Disp.A | Volatile Uncorr. ECC | | Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. | |===============================+======================+======================| | 0 GeForce GTX 1070 WDDM | 00000000:09:00.0 On | N/A | | 0% 48C P0 36W / N/A | 779MiB / 8192MiB | 0% Default | +-------------------------------+----------------------+----------------------+ +-----------------------------------------------------------------------------+ | Processes: GPU Memory | | GPU PID Type Process name Usage | |=============================================================================| | 0 Not Supported | +-----------------------------------------------------------------------------+
Help me please…