read
Ubuntu 16.04 LTS version Tensorflow install.
Version info
- Ubuntu 16.04 LTS
- CUDA 8.0
- CUDNN 6.0
- Tensorflow 1.4 (recent version)
To send a file from local to remote server, scp is useful command.
scp -P PORT_NUMBER FILE_NAME USER@REMOTE_IP:DIR
Install process
- NVIDIA driver install
- CUDA install
- Environment setting
- CUDnn install
- NVIDIA docker install
- Anaconda install
- Tensorflow Install
NVIDIA driver install
sudo apt-get purge nvidia-*
sudo add-apt-repository ppa:graphics-drivers/ppa
sudo apt-get update
sudo apt-get install nvidia-375
sudo reboot
CUDA Download
wget -c https://developer.nvidia.com/compute/cuda/8.0/Prod2/local_installers/cuda-repo-ubuntu1604-8-0-local-ga2_8.0.61-1_amd64-deb
sudo dpkg -i cuda-repo-ubuntu1604-8-0-local-ga2_8.0.61-1_amd64-deb
sudo apt update
sudo apt install cuda
Environment setting and test
export PATH=/usr/local/cuda-8.0/bin${PATH:+:${PATH}}
export LD_LIBRARY_PATH=/usr/local/cuda-8.0/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}
cd $LD_LIBRARY_PATH
cd ../bin/
bash cuda-install-samples-8.0.sh ~
cd ~/NVIDIA_CUDA-8.0_Samples/5_Simulations/nbody
make
./nbody
CUDNN Download
To download cuDNN, you need to register your account.
And after that, you can download the compressed file and uncompress and move to cuda directory.
tar -xvf cudnn-8.0-linux-x64-v6.0.tar
cd ~/cuda/lib64
sudo mv * /usr/local/cuda/lib64/
cd ~/cuda/include
sudo mv * /usr/local/cuda/include/
Install library
sudo apt-get install libcupti-dev
NVIDIA Docker Install
curl -L https://nvidia.github.io/nvidia-docker/gpgkey | \
sudo apt-key add -
sudo tee /etc/apt/sources.list.d/nvidia-docker.list <<< \
"deb https://nvidia.github.io/libnvidia-container/ubuntu16.04/amd64 /
deb https://nvidia.github.io/nvidia-container-runtime/ubuntu16.04/amd64 /
deb https://nvidia.github.io/nvidia-docker/ubuntu16.04/amd64 /"
sudo apt-get update
sudo apt-get install nvidia-docker2
sudo pkill -SIGHUP dockerd
Anaconda
Install
wget -c https://repo.continuum.io/archive/Anaconda3-5.0.1-Linux-x86_64.sh
bash Anaconda3-5.0.1-Linux-x86_64.sh
Anaconda setting
conda create -n tf python=3.5
source activate tf
pip install tensorflow-gpu
Tensorflow test
from tensorflow.python.client import device_lib
device_lib.list_local_devices()
import tensorflow as tf
# Creates a graph.
with tf.device('/gpu:0'):
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)
Image