How to really install tensorflow-gpu from source on a clean Ubuntu
level: beginners
the correct versions of the components must be compatible with each other, and they have to be installed in the correct order. i chose versions of the components that were released around the same period of time. also, tensorflow-gpu should be built from source.
os: Ubuntu - all versions (anything above 16.0.0 - main releases only)
gpu type: nvidia
python version: 3
components: cuda toolkit 9.0, cudnn 7, nccl 2.3.7, bazel 0.8.0, gcc 6, g++ 6, tensorflow r1.5 sources
install a new Ubuntu instance and nvidia drivers
sudo apt-get update && sudo apt-get upgrade
sudo ubuntu-drivers autoinstall
sudo reboot
sudo apt-get update && sudo apt-get upgrade
install cuda toolkit 9.0 - Linux, x86_64, Ubuntu, 17.04, deb (local)
sudo apt-key add /var/cuda-repo-9-0-local/7fa2af80.pub
sudo dpkg -i cuda-repo-ubuntu1704-9-0-local_9.0.176-1_amd64.deb
install cudnn 7.0 - cuDNN v7.4.2 (Dec 14, 2018), for CUDA 9.0:
runtime:
sudo apt-get update && sudo apt-get upgrade
sudo dpkg -i libcudnn7_7.4.2.24-1+cuda9.0_amd64.deb
developer:
sudo apt-get update && sudo apt-get upgrade
sudo dpkg -i libcudnn7-dev_7.4.2.24-1+cuda9.0_amd64.deb
sudo apt-get install aptitude
sudo aptitude install cuda
accept all aptitude solutions
sudo apt-get install libcupti-dev
export PATH=/usr/local/cuda-9.0/bin${PATH:+:${PATH}}
export LD_LIBRARY_PATH=/usr/local/cuda/lib64:${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}
install nccl - NCCL v2.3.7, for CUDA 9.0, Nov 8 & Dec 14, 2018:
sudo apt-get update && sudo apt-get upgrade
sudo dpkg -i nccl-repo-ubuntu1604-2.3.7-ga-cuda9.0_1-1_amd64.deb
sudo apt-get update && sudo apt-get upgrade
sudo apt install libnccl2 libnccl-dev
install bazel 0.8.0: https://github.com/bazelbuild/bazel/releases/download/0.8.0/bazel-0.8.0-installer-linux-x86_64.sh (158 MB):**
sudo apt-get update && sudo apt-get upgrade
sudo apt-get install pkg-config zip g++ zlib1g-dev unzip python
sudo chmod +x bazel-0.8.0-installer-linux-x86_64.sh
sudo ./bazel-0.8.0-installer-linux-x86_64.sh
export PATH="$PATH:/usr/local/bin"
install gcc 6:
sudo apt-get update && sudo apt-get upgrade
sudo apt-get install gcc-6 g++-6
build tensorflow-gpu from source:
sudo apt-get update && sudo apt-get upgrade
sudo apt-get install python3-dev python3-pip git
git clone https://github.com/tensorflow/tensorflow.git
cd tensorflow
git checkout r1.5
bazel shutdown
bazel clean
git clean -xdf
./configure
- Please specify the location of python. [Default is /usr/bin/python]: /usr/bin/python3
- Please input the desired Python library path to use. Default is [/usr/lib/python3/dist-packages]
- Do you wish to build TensorFlow with jemalloc as malloc support? [Y/n]: n
- Do you wish to build TensorFlow with Google Cloud Platform support? [Y/n]: n
- Do you wish to build TensorFlow with Hadoop File System support? [Y/n]: n
- Do you wish to build TensorFlow with Amazon S3 File System support? [Y/n]: n
- Do you wish to build TensorFlow with XLA JIT support? [y/N]: N
- Do you wish to build TensorFlow with GDR support? [y/N]: N
- Do you wish to build TensorFlow with VERBS support? [y/N]: N
- Do you wish to build TensorFlow with OpenCL SYCL support? [y/N]: N
- Do you wish to build TensorFlow with CUDA support? [y/N]: Y
- Please specify the CUDA SDK version you want to use, e.g. 7.0. [Leave empty to default to CUDA 9.0]:
- Please specify the location where CUDA 9.0 toolkit is installed. Refer to README.md for more details. [Default is /usr/local/cuda]:
- Please specify the cuDNN version you want to use. [Leave empty to default to cuDNN 7.0]:
- Please specify the location where cuDNN 7 library is installed. Refer to README.md for more details. [Default is /usr/local/cuda]:
- Please specify a list of comma-separated Cuda compute capabilities you want to build with. You can find the compute capability of your device at: https://developer.nvidia.com/cuda-gpus. Please note that each additional compute capability significantly increases your build time and binary size. [Default is: 5.0]
- Do you want to use clang as CUDA compiler? [y/N]: N
- Please specify which gcc should be used by nvcc as the host compiler. [Default is /usr/bin/gcc]: /usr/bin/gcc-6
- Do you wish to build TensorFlow with MPI support? [y/N]: N
- Please specify optimization flags to use during compilation when bazel option "--config=opt" is specified [Default is -march=native]:
sudo pip3 install numpy
sudo bazel build --config=opt --config=cuda //tensorflow/tools/pip_package:build_pip_package --verbose_failures
INFO: Build completed successfully, 5984 total actions
build the package:
"The bazel build command creates an executable named build_pip_package—this is the program that builds the pip package. For example, the following builds a .whl package in the /tmp/tensorflow_pkg directory:"
sudo ./bazel-bin/tensorflow/tools/pip_package/build_pip_package /tmp/tensorflow_pkg
Fri Jan 11 06:51:40 IST 2019 : === Output wheel file is in: /tmp/tensorflow_pkg
ls /tmp/tensorflow_pkg
tensorflow-1.5.1-cp36-cp36m-linux_x86_64.whl
install the package:
sudo python3 -m pip uninstall pip && sudo apt install python3-pip --reinstall
sudo pip3 install /tmp/tensorflow_pkg/tensorflow-1.5.1-cp36-cp36m-linux_x86_64.whl
verify installation:
cd ..
python3
import tensorflow
quit()
if everything is ok you should see nothing. type quit()
install useful libraries
sudo pip3 install pandas keras
make a snapshot or image of the machine
fast route - installing compiled packages rather than building from source:
sudo pip3 install --upgrade tensorflow-gpu
sudo pip3 uninstall tensorflow-gpu
sudo pip3 install tensorflow-gpu==0.0
0.12.1, 1.0.0, 1.0.1, 1.1.0rc1, 1.1.0rc2, 1.1.0, 1.2.0rc0, 1.2.0rc1, 1.2.0rc2, 1.2.0, 1.2.1, 1.3.0rc0, 1.3.0rc1, 1.3.0rc2, 1.3.0, 1.4.0rc0, 1.4.0rc1, 1.4.0, 1.4.1, 1.5.0rc0, 1.5.0rc1, 1.5.0, 1.5.1, 1.6.0rc0, 1.6.0rc1, 1.6.0, 1.7.0rc0, 1.7.0rc1, 1.7.0, 1.7.1, 1.8.0rc0, 1.8.0rc1, 1.8.0, 1.9.0rc0, 1.9.0rc1, 1.9.0rc2, 1.9.0, 1.10.0rc0, 1.10.0rc1, 1.10.0, 1.10.1, 1.11.0rc0, 1.11.0rc1, 1.11.0rc2, 1.11.0, 1.12.0rc0, 1.12.0rc1, 1.12.0rc2, 1.12.0
if this saved you:
1KJ1SNVoMT9zXRJxrRC8y8awuxRfmUd3Vb