0

I am trying to execute a machine learning project on my laptop having Ubuntu (18.04) and this requires GPU. My laptop has NVIDIA Geforce MX-150 (2GB) GPU and I have the following installed.

Kernel                : 5.3.0-53-generic
GCC                   : 7.5.0
Nvidia Driver Version : 440.33
CUDA Version          : 10.0
cuDNN Version         : 7.4.1.5-1+cuda10.0 
Python                : 3.6.3
tensorflow            : 1.14.0
tensorflow-gpu        : 1.14.0

Below is the output of nvidia-smi command

nvidia-smi

+-----------------------------------------------------------------------------+
Sat May 23 03:36:54 2020       
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 440.33.01    Driver Version: 440.33.01    CUDA Version: 10.2     |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|===============================+======================+======================|
|   0  GeForce MX150       On   | 00000000:01:00.0 Off |                  N/A |
| N/A   57C    P0    N/A /  N/A |    320MiB /  2002MiB |      0%      Default |
+-------------------------------+----------------------+----------------------+

|Processes:                                                        GPU Memory |
|  GPU       PID   Type   Process name                             Usage      |
|=============================================================================|
|    0      1054      G   /usr/lib/xorg/Xorg                            20MiB |
|    0      1204      G   /usr/bin/gnome-shell                          46MiB |
|    0      1509      G   /usr/lib/xorg/Xorg                            89MiB |
|    0      1680      G   /usr/bin/gnome-shell                          76MiB |
|    0      2078      G   ...AAAAAAAAAAAACAAAAAAAAAA= --shared-files    84MiB |
+-----------------------------------------------------------------------------+

On checking for GPU availability I am getting the following message from python:

import tensorflow as tf

tf.test.is_gpu_available()
2020-05-23 03:22:00.113303: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU
supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA

2020-05-23 03:22:00.139002: I tensorflow/core/platform/profile_utils/cpu_utils.cc:94] 
CPU Frequency: 1800000000 Hz

2020-05-23 03:22:00.139733: I tensorflow/compiler/xla/service/service.cc:168] XLA service 
0x55bf899c6a30 executing computations on platform Host. Devices:

2020-05-23 03:22:00.139791: I tensorflow/compiler/xla/service/service.cc:175]   
StreamExecutor device (0): Host, Default Version

False

My Graphic card supports CUDA , this can be verified from the link below:

https://www.geforce.com/hardware/notebook-gpus/geforce-mx150/specifications/

I did all these installations after carefully examining the compatibility all the related softwares and now I still not able to successfully work with my GPU. Can anyone please tell me what is the issue here with my GPU or processor and why isn't the GPU being recognised ?

talonmies
  • 70,661
  • 34
  • 192
  • 269
Sarvagya Dubey
  • 435
  • 1
  • 7
  • 22
  • Did you check this link[About Subject](https://stackoverflow.com/questions/47068709/your-cpu-supports-instructions-that-this-tensorflow-binary-was-not-compiled-to-u) before – baysal celik May 22 '20 at 22:32
  • OK, so your graphic card is absolutely fine, and your driver is also absolutely fine. Maybe your CUDA's version is too low, `tensorflow >= 2.1.0` requires `CUDA 10.1`. It could be the reason. So we also need to know your TensorFlow's version. – Sraw May 22 '20 at 22:53
  • @Sraw I referred to this answer where all the compatibility matrix has been shared. Maybe it has something to do with bazel and build options, I am trying to explore that https://stackoverflow.com/questions/50622525/which-tensorflow-and-cuda-version-combinations-are-compatible/ – Sarvagya Dubey May 22 '20 at 23:00
  • Sorry, I see your tensorflow's version. My bad. – Sraw May 22 '20 at 23:05
  • https://stackoverflow.com/questions/38073432/how-to-suppress-verbose-tensorflow-logging How about print verbose debug information so that we can address the problem. – Sraw May 22 '20 at 23:10

1 Answers1

-1

Closed-code (proprietary) drives cannot be examined by Linux or Ubuntu developers by nature. my advice, be careful not to have a closed source video card and try to make the downloads complete

Sir slayzer
  • 26
  • 1