I am running an Intel Server board S26000 with a quadro RTX 4000. I am running Ubuntu 16.04 Anaconda 2019.03, CUDA 10.1, CUDNN 7.5.1.10. I am running under a Anaconda environment (Python 3.6) and have installed Tensorflow-gpu 12.0. Running nvidia-smi I get the following which shows only 1% utilization. (base) potero@Xeon:~$ nvidia-smi
Mon May 20 10:39:55 2019
±----------------------------------------------------------------------------+
| NVIDIA-SMI 418.56 Driver Version: 418.56 CUDA Version: 10.1 |
|-------------------------------±---------------------±---------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
|===============================+======================+======================|
| 0 Quadro RTX 4000 Off | 00000000:02:00.0 On | N/A |
| 30% 39C P8 11W / 125W | 390MiB / 7951MiB | 1% Default |
±------------------------------±---------------------±---------------------+
±----------------------------------------------------------------------------+
| Processes: GPU Memory |
| GPU PID Type Process name Usage |
|=============================================================================|
| 0 1948 G /usr/lib/xorg/Xorg 210MiB |
| 0 2497 G /usr/bin/gnome-shell 38MiB |
| 0 3104 G …6631945,3290711536085474279,131072 --en 39MiB |
| 0 3950 C /usr/lib/libreoffice/program/soffice.bin 97MiB |
±----------------------------------------------------------------------------+
(base) potero@Xeon:~$
I am running Tensorflow with Keras under Jupyter Notebook. Are there any setting that I need to activate to get the tensorflow utilization up?