在ubuntu18.04中导入错误tensorflow-gpu

Question

I'm getting some error while importing tensorflow.

My computer's Specifications:

OS:ubuntu 18.04

Nvidia RTX 2080 Ti*2

Nvidia driver-415

CUDA：10.0

cuDNN：7.3.0 tensorflow：1.11.0

import tensorflow

Error:

Traceback (most recent call last): File "/usr/local/lib/python3.6/dist-packages/tensorflow/python/pywrap_tensorflow.py", line 58, in from tensorflow.python.pywrap_tensorflow_internal import * File "/usr/local/lib/python3.6/dist-packages/tensorflow/python/pywrap_tensorflow_internal.py", line 28, in _pywrap_tensorflow_internal = swig_import_helper() File "/usr/local/lib/python3.6/dist-packages/tensorflow/python/pywrap_tensorflow_internal.py", line 24, in swig_import_helper _mod = imp.load_module('_pywrap_tensorflow_internal', fp, pathname, description) File "/usr/lib/python3.6/imp.py", line 243, in load_module return load_dynamic(name, filename, file) File "/usr/lib/python3.6/imp.py", line 343, in load_dynamic return _load(spec) ImportError: libcublas.so.9.0: cannot open shared object file: No such file or directory

During handling of the above exception, another exception occurred:

Traceback (most recent call last): File "", line 1, in File "/usr/local/lib/python3.6/dist-packages/tensorflow/ init .py", line 22, in from tensorflow.python import pywrap_tensorflow # pylint: disable=unused-import File "/usr/local/lib/python3.6/dist-packages/tensorflow/python/ init .py", line 49, in from tensorflow.python import pywrap_tensorflow File "/usr/local/lib/python3.6/dist-packages/tensorflow/python/pywrap_tensorflow.py", line 74, in raise ImportError(msg) ImportError: Traceback (most recent call last): File "/usr/local/lib/python3.6/dist-packages/tensorflow/python/pywrap_tensorflow.py", line 58, in from tensorflow.python.pywrap_tensorflow_internal import * File "/usr/local/lib/python3.6/dist-packages/tensorflow/python/pywrap_tensorflow_internal.py", line 28, in _pywrap_tensorflow_internal = swig_import_helper() File "/usr/local/lib/python3.6/dist-packages/tensorflow/python/pywrap_tensorflow_internal.py", line 24, in swig_import_helper _mod = imp.load_module('_pywrap_tensorfl ow_internal', fp, pathname, description) File "/usr/lib/python3.6/imp.py", line 243, in load_module return load_dynamic(name, filename, file) File "/usr/lib/python3.6/imp.py", line 343, in load_dynamic return _load(spec) ImportError: libcublas.so.9.0: cannot open shared object file: No such file or directory

Failed to load the native TensorFlow runtime.

See https://www.tensorflow.org/install/install_sources#common_installation_problems

for some common reasons and solutions. Include the entire stack trace above this error message when asking for help.

I already tried to use ubuntu 16.04 ,but GPU didn't support. Installation of CUDA9.0 & CUDA9.2 was not supported too.

how can I use tensorflow-gpu?

I already app path in ~/.bashrc

export PATH=/usr/local/cuda-10.0/bin${PATH:+:${PATH}}
export LD_LIBRARY_PATH=/usr/local/cuda-10.0/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}
export CUDA_HOME=/usr/local/cuda-10.0
export LD_LIBRARY_PATH=/usr/local/cuda-10.0/lib64
export LD_LIBRARY_PATH=/usr/local/cuda-10.0/lib64$LD_LIBRARY_PATH

Answer 1

The tensorflow-gpu package is built against Cuda 9.0, but you have Cuda 10.0 installed.

You need either to downgrade your version of Cuda to 9.0 (but if I recall, that's not possible with a 2080Ti), or build tensorflow from the sources. There is extensive documentation on how to do so on the tensorflow webiste

You can also try to install the package tf-nightly-gpu . You should note that this version is more experimental though, as it has not been tested as extensively.