![nvidia cuda toolkit 9.0 nvidia cuda toolkit 9.0](https://tech.amikelive.com/wp-content/uploads/2018/12/cuda_upgrade_standard-300x174.png)
![nvidia cuda toolkit 9.0 nvidia cuda toolkit 9.0](https://windows-cdn.softpedia.com/screenshots/NVIDIA-CUDA-Toolkit_8.png)
You can check whether your card is CUDA-compatible here and here (for older cards). You need a CUDA-compatible graphic card to use CNTK GPU capabilities. This is normal and not a problem.This section outlines the packages you need to setup in order for CNTK to leverage NVIDIA GPUs. For example if you have a r384 GPU driver installed, the CUDA driver API interface within that GPU driver supports CUDA 9 (but it also supports CUDA 8, and prior CUDA versions as well). The CUDA driver version is reflective of the GPU driver version that is installed. deviceQuery, it gives back cuda driver version 9.0 linked against CUDA 9.0.Īnother issue is that, I just installed cuda-8.0, but why when I run. Libcublas.so.9.0: cannot open shared object file: No such file or directoryĬertainly, without a doubt, indicates that something is looking for i.e. Anyway I have no idea what you did or what is happening exactly, but this: However if you build TF from sources, it’s possible for it to be linked against some other CUDA version. Please see the link as below.Ĭertainly that is true if you installed TF using a binary install method, such as pip However, from the discussion of tensorflow’s github, currently, tensorflow does not support cuda-9.0 currently.
Nvidia cuda toolkit 9.0 how to#
and installing it all over again several times but I always end up getting the same error.Īny suggestions what could possibly go wrong or how to fix this? Tried removing everything that is associated with nvidia from my linux including drivers, cuda toolkit, cudnn etc. I have checked the environmental variables and all seems to be perfectly correct.
![nvidia cuda toolkit 9.0 nvidia cuda toolkit 9.0](https://windows-cdn.softpedia.com/screenshots/NVIDIA-CUDA-Toolkit_21.jpg)
The import error shows that libcublas.so.8.0 is missing, however in both cuda/lib64 and cuda-9.0/lib64 the file does exists (newer version libcublas.so.9.0). Include the entire stack traceĪbove this error message when asking for help.
![nvidia cuda toolkit 9.0 nvidia cuda toolkit 9.0](https://cdn-images-1.medium.com/max/1600/1*1Cf_hc9pmWVHJOH3YvOI2w.png)
ImportError: Traceback (most recent call last):įailed to load the native TensorFlow runtime.įor some common reasons and solutions. ImportError: libcublas.so.8.0: cannot open shared object file: No such file or directoryĭuring handling of the above exception, another exception occurred:įile "/home/albert/anaconda3/lib/python3.6/site-packages/tensorflow/_init_.py", line 24, in įile "/home/albert/anaconda3/lib/python3.6/site-packages/tensorflow/python/_init_.py", line 49, in įrom tensorflow.python import pywrap_tensorflowįile "/home/albert/anaconda3/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow.py", line 72, in Return load_dynamic(name, filename, file)įile "/home/albert/anaconda3/lib/python3.6/imp.py", line 343, in load_dynamic _mod = imp.load_module('_pywrap_tensorflow_internal', fp, pathname, description)įile "/home/albert/anaconda3/lib/python3.6/imp.py", line 243, in load_module _pywrap_tensorflow_internal = swig_import_helper()įile "/home/albert/anaconda3/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow_internal.py", line 24, in swig_import_helper The stack trace from the shell: Traceback (most recent call last):įile "/home/albert/anaconda3/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow.py", line 58, in įrom _tensorflow_internal import *įile "/home/albert/anaconda3/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow_internal.py", line 28, in
Nvidia cuda toolkit 9.0 code#
Installed tensorflow-gpu, tried to verify this using sample code from the website (both python3 Shell and IDE Spyder) but unfortunately no luck. bandwidthTest, both of them passed correctly. Of course, after step one and two I have verified the installation using.