Physical GPU are not recognize

There are 10 installed gpu.

However there is one device that process recognizes.

I can’t find cause. Tensorflow version is 2.5.0.

Error Message Not Occured.

Thank you

@resoto77,

Welcome to the Tensorflow Forum!

Can you share the output of the following code?

import tensorflow as tfprint("Num GPUs Available: ", len(tf.config.list_physical_devices('GPU')))

Num GPUs Available: 1

available gpu is only one. not recognize 9 gpu

According to tested build configurations

CUDA:11.2 and cuDNN: 8.1 are compatible with Tensorflow 2.5.

Can you please try as suggested above and let us know?

#ifndef CUDNN_VERSION_H_
#define CUDNN_VERSION_H_

#define CUDNN_MAJOR 8
#define CUDNN_MINOR 1
#define CUDNN_PATCHLEVEL 0

#define CUDNN_VERSION (CUDNN_MAJOR * 1000 + CUDNN_MINOR * 100 + CUDNN_PATCHLEVEL)

nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2020 NVIDIA Corporation
Built on Mon_Nov_30_19:08:53_PST_2020
Cuda compilation tools, release 11.2, V11.2.67
Build cuda_11.2.r11.2/compiler.29373293_0

I’m sorry for replying late. CUDA and cuDNN version are right.

hmm… I don’t know what the problem is.

Sometimes the ‘Graphics card not detected’ error will occur upon the installation of new drivers if something goes wrong. This could be a faulty driver or the new drivers’ incompatibility with another component inside the PC; the options are too numerous to list.

Can you please share operating system details?

Another Process use “CUDA_VISIBLE_DEVICES”.
That’s why only one device was recognized.
Thank you. help us.