GPU availability check

Question

I am trying to test that my Jupyter notebook is using the GPU or not but when I check with this code, It shows me '0' GPU's available. But, My system is i7 10th Generation and GEFORCE RTX 2060. I have downloaded cuda and NVIDIA CNN added to the system variables and in anaconda downlaoded tensorflow GPU but I don't know why its not recognizing my GPU.

'''
    print("Num GPUs Available: ", len(tf.config.experimental.list_physical_devices('GPU')))
'''
>>>>Num GPUs Available:  0

Tal Folkman · Answer 1 · 2022-03-03T13:52:39.727

0

try use gputil:

import GPUtil
GPUtil.showUtilization()

try also look here - How to get current available GPUs in tensorflow?

edited Mar 03 '22 at 13:52

answered Jan 19 '22 at 17:30

Tal Folkman

2,368
1
7
21

It did not work. Is there any way to achieve this? – hemanth vema Mar 02 '22 at 20:22
what is the error? @hemanthvema – Tal Folkman Mar 03 '22 at 13:50
I want to make my Jupyter notebook always use my GPU instead of using CPU, but I tried so many times every time it is saying, GPU is 0 . I don't know whether it is possible to use my Jupyter notebook every time with my GPU or not. – hemanth vema Mar 05 '22 at 17:54
Unless I'm mistaken, 0 is the first available GPU. If you had none, `GPUtil.getAvailable()` would return `[]`instead of `[0]`. – Danferno Apr 26 '23 at 08:49

GPU availability check

1 Answers1