2

I have installed tensorflow 1.5 with:

pip install tensorflow==1.5

because i had an error:

(ImportError: libcusolver.so.8.0: cannot open shared object file: No such file or directory

Failed to load the native TensorFlow runtime.)

After i installed Keras with:

pip install Keras

Finally I installed TensorFlow gpu with:

pip install tensorflow-gpu

When i run my code i searched with:

nvidia-smi

if there was my process but it didn't. What can i do?

MrCaptain Alex
  • 169
  • 1
  • 13
  • make sure you've installed the right version of nvidia driver, cuda, and cudnn as mentioned in this question. https://stackoverflow.com/questions/43558707/tensorflow-importerror-libcusolver-so-8-0-cannot-open-shared-object-file-no – Jialer Chew Sep 25 '18 at 09:29
  • CUDA is required to use the GPU in TensorFlow. – Dr. Snoopy Sep 25 '18 at 09:37

2 Answers2

0

Try to uninstall the previous versions of tenserflow, tenserflow-gpu and install them with specified version. Tenserflow-gpu also installs tenserflow as a dependancy, thus you may have a versioning difference. More info on this issue can be found on their git repo.

pip uninstall tensorflow
pip uninstall tensorflow-gpu
pip install tensorflow==1.5.0
pip install tensorflow-gpu==1.5.0
Karen Galstyan
  • 13
  • 1
  • 4
0

Have you installed cuda and cudnn? And add the env path to .bashrc?

Set the CUDA Path in ~/.bashrc:

# Cuda Nvidia path
$ export LD_LIBRARY_PATH="$LD_LIBRARY_PATH:/usr/local/cuda/lib64"
$ export CUDA_HOME=/usr/local/cuda

Pd: You have to install only tensorflow-gpu version, not cpu version.

After it you need to download a cuDNN version compatible with your Cuda version. For example, you can download cuda v7.1.3 via ssh/putty:

$ wget https://developer.download.nvidia.com/compute/machine-learning/cudnn/secure/v7.1.3/prod/8.0_20180414/cudnn-8.0-linux-x64-v7.1.tgz?2H2brULbeJyaEnHtkjrwUED5JR-vx6QBedn-A-nsa9Q73hUCfYCmWbFG_-eBnlAv3_CjRmI3qhRPJDiRz0wRC3oGviiYTX7M-H-RUiZQR_vuo21iCM5W-R0iaJOuwt0bmw-RFTg2XK_a8gdiV-uemVTH8Lf8-Q8rf3Msh52hznszsZKCP0hq2DvYNFuTSjyOSgPiH-3c_Th2uw

Once downloaded, just extract it and copy related files into CUDA install directory. This example is for cuda v8.0 and cudnn v7.1.3:

$ mv cudnn-8.0-linux-x64-v7.1.tgz?2H2brULbeJyaEnHtkjrwUED5JR-vx6QBedn-A-nsa9Q73hUCfYCmWbFG_-eBnlAv3_CjRmI3qhRPJDiRz0wRC3oGviiYTX7M-H-RUiZQR_vuo21iCM5W-R0iaJOuwt0bmw-RFTg2XK_a8gdiV-uemVTH8Lf8-Q8rf3Msh52hznszsZKCP0hq2DvYNFuTSjyOSgPiH-3c_Th2uw cudnn-8.0-linux-x64-v7.1.tgz
$ tar -zxvf cudnn-8.0-linux-x64-v7.1.tgz
$ sudo cp cuda/include/cudnn.h /usr/local/cuda/include/
$ sudo cp cuda/lib64/libcudnn* /usr/local/cuda/lib64/
$ sudo chmod a+r /usr/local/cuda/include/cudnn.h
$ sudo chmod a+r /usr/local/cuda/lib64/libcudnn*

Then, you can check that tensorflow is running over GPU:

# Python
import tensorflow as tf

# Creates a graph.
a = tf.constant([1.0, 2.0, 3.0, 4.0, 5.0, 6.0], shape=[2, 3], name='a')
b = tf.constant([1.0, 2.0, 3.0, 4.0, 5.0, 6.0], shape=[3, 2], name='b')
c = tf.matmul(a, b)
# Creates a session with log_device_placement set to True.
sess = tf.Session(config=tf.ConfigProto(log_device_placement=True))
# Runs the op.
print(sess.run(c))

If all it's ok you get something like this:

2018-09-25 13:27:23.614538: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1356] Found device 0 with properties: 
name: GeForce GTX 1070 Ti major: 6 minor: 1 memoryClockRate(GHz): 1.683
pciBusID: 0000:01:00.0
totalMemory: 7.93GiB freeMemory: 6.22GiB
2018-09-25 13:27:23.614552: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1435] Adding visible gpu devices: 0
2018-09-25 13:27:23.800175: I tensorflow/core/common_runtime/gpu/gpu_device.cc:923] Device interconnect StreamExecutor with strength 1 edge matrix:
2018-09-25 13:27:23.800210: I tensorflow/core/common_runtime/gpu/gpu_device.cc:929]      0 
2018-09-25 13:27:23.800215: I tensorflow/core/common_runtime/gpu/gpu_device.cc:942] 0:   N 
2018-09-25 13:27:23.800393: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1053] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 5997 MB memory) -> physical GPU (device: 0, name: GeForce GTX 1070 Ti, pci bus id: 0000:01:00.0, compute capability: 6.1)
Device mapping:
/job:localhost/replica:0/task:0/device:GPU:0 -> device: 0, name: GeForce GTX 1070 Ti, pci bus id: 0000:01:00.0, compute capability: 6.1
2018-09-25 13:27:23.857141: I tensorflow/core/common_runtime/direct_session.cc:284] Device mapping:
/job:localhost/replica:0/task:0/device:GPU:0 -> device: 0, name: GeForce GTX 1070 Ti, pci bus id: 0000:01:00.0, compute capability: 6.1

MatMul: (MatMul): /job:localhost/replica:0/task:0/device:GPU:0
2018-09-25 13:27:23.857701: I tensorflow/core/common_runtime/placer.cc:886] MatMul: (MatMul)/job:localhost/replica:0/task:0/device:GPU:0
b: (Const): /job:localhost/replica:0/task:0/device:GPU:0
2018-09-25 13:27:23.857729: I tensorflow/core/common_runtime/placer.cc:886] b: (Const)/job:localhost/replica:0/task:0/device:GPU:0
a: (Const): /job:localhost/replica:0/task:0/device:GPU:0
2018-09-25 13:27:23.857736: I tensorflow/core/common_runtime/placer.cc:886] a: (Const)/job:localhost/replica:0/task:0/device:GPU:0
[[22. 28.]
 [49. 64.]]
David Jimenez
  • 71
  • 6