Hello,
I am interested in your work and now trying to run the code for mini dataset. I find the terminal outputing "Error in magma_getdevice_arch: MAGMA not initialized (call magma_init() first) or bad device" 4 times every epoch and seems it occurs at line 84 in train.py. However the program is still running without being interrupted by the error. So should I ignore the problem or there are some solutions for it?
My running environment is Ubuntu 18.04 with python 3.6 and cuda 11.4.
Thanks.
Hello, I am interested in your work and now trying to run the code for mini dataset. I find the terminal outputing "Error in magma_getdevice_arch: MAGMA not initialized (call magma_init() first) or bad device" 4 times every epoch and seems it occurs at line 84 in train.py. However the program is still running without being interrupted by the error. So should I ignore the problem or there are some solutions for it? My running environment is Ubuntu 18.04 with python 3.6 and cuda 11.4. Thanks.