{"id":1951,"date":"2018-05-02T17:54:17","date_gmt":"2018-05-02T17:54:17","guid":{"rendered":"http:\/\/fayazrasheed.com\/?p=1951"},"modified":"2018-09-09T02:18:24","modified_gmt":"2018-09-09T02:18:24","slug":"installing-tensorflowgpu-opencv-and-dlib-on-ubuntu-18-04-bionic-beaver","status":"publish","type":"post","link":"https:\/\/fayazrasheed.com\/?p=1951","title":{"rendered":"Installing Tensorflow(GPU), OpenCV and dlib on Ubuntu 18.04 Bionic Beaver"},"content":{"rendered":"

Image source: https:\/\/www.tensorflow.org\/<\/p>\n

For those who are ready for machine learning and computer vision with the updated versions of OpenCV, dlib, Tensorflow (GPU) on the Bionic Beaver.<\/p>\n

Install synaptic and atom from Ubuntu’s package manager<\/p>\n

[code language=”bash”]
\nsudo ubuntu-drivers autoinstall
\nsudo add-apt-repository ppa:graphics-drivers\/ppa
\nsudo apt update
\n[\/code]<\/p>\n

search and install nvidia-390 from synaptic<\/p>\n

Download CUDA 9.0 (It has to be 9.0 for Tensorflow 1.8):<\/p>\n

https:\/\/developer.nvidia.com\/cuda-90-download-archive?target_os=Linux&target_arch=x86_64&target_distro=Ubuntu&target_version=1704&target_type=runfilelocal<\/a><\/p>\n

And Download cuDNN v7.1.3 for CUDA 9.0:<\/p>\n

https:\/\/developer.nvidia.com\/rdp\/cudnn-download<\/a><\/p>\n

[code language=”bash”]
\nsudo chmod +x cuda_9.0.176_384.81_linux.run
\nsudo chmod +x cuda_9.0.176.1_linux.run
\nsudo chmod +x cuda_9.0.176.2_linux.run
\nsudo .\/cuda_9.0.176_384.81_linux.run –override
\nsudo .\/cuda_9.0.176.1_linux.run
\nsudo .\/cuda_9.0.176.2_linux.run
\n[\/code]<\/p>\n

DO NOT INSTALL THE DRIVER AND SAMPLES IN THIS PART! Ignore the fact that they give you a warning for the driver not being installed. This is because the installer cannot detect the installed driver in your system, which we installed earlier through synaptic.<\/p>\n

[code language=”bash”]
\nsudo apt-get install cuda-9
\nsudo apt-get upgrade
\ntar -zxvf cudnn-9.0-linux-x64-v7.1.tgz
\nsudo cp -P cuda\/lib64\/* \/usr\/local\/cuda-9.0\/lib64\/
\nsudo cp cuda\/include\/* \/usr\/local\/cuda-9.0\/include\/
\nsudo chmod a+r \/usr\/local\/cuda-9.0\/include\/cudnn.h
\nsudo apt-get install libcupti-dev
\nsudo atom ~\/.bashrc
\n[\/code]<\/p>\n

And add these lines:<\/p>\n

export PATH=\/usr\/local\/cuda-9.0\/bin${PATH:+:${PATH}}<\/code>
\nexport LD_LIBRARY_PATH=\/usr\/local\/cuda-9.0\/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}<\/code><\/p>\n

[code language=”bash”]
\nsource ~\/.bashrc
\nsudo apt-get update
\nsudo apt-get install build-essential cmake libopenblas-dev liblapack-dev libx11-dev libgtk-3-dev python python-dev python-pip python3 python3-dev python3-pip<\/p>\n

[\/code]<\/p>\n

Download and install Anaconda Python 3.6 (I use 3.6 universally and 2.7 for scientific computations)<\/p>\n

Create environment using an environment name (envname)<\/p>\n

[code language=”bash”]
\nconda create -n envname python=2.7
\nsource activate envname
\npip install numpy pillow lxml jupyter matplotlib dlib protobuf
\nsudo apt -y install python-opencv
\nconda install -c conda-forge opencv
\nsudo snap install protobuf –classic
\npip install –upgrade tensorflow-gpu
\n[\/code]<\/p>\n

To KILL process and clear memory of GPU:<\/p>\n

[code language=”bash”]
\nnvidia-smi
\n[\/code]<\/p>\n

and kill the process causing unwanted memory usage<\/p>\n

[code language=”bash”]
\nsudo kill -20483 PID.
\n[\/code]<\/p>\n","protected":false},"excerpt":{"rendered":"

Image source: https:\/\/www.tensorflow.org\/ For those who are ready for machine learning and computer vision with the updated versions of OpenCV, dlib, Tensorflow (GPU) on the Bionic Beaver. Install synaptic and atom from Ubuntu’s package manager [code language=”bash”] sudo ubuntu-drivers autoinstall sudo add-apt-repository ppa:graphics-drivers\/ppa sudo apt update [\/code] search…<\/p>\n","protected":false},"author":1,"featured_media":1974,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[2,3,4,5],"tags":[118,120,122,121,108,119,68],"class_list":["post-1951","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-fluids-aerodynamics","category-informative-stuff","category-life-as-it-happens","category-tech-blogs","tag-bionic-beaver","tag-dlib","tag-gpu","tag-opencv","tag-tensorflow","tag-tensorflow-gpu","tag-ubuntu"],"_links":{"self":[{"href":"https:\/\/fayazrasheed.com\/index.php?rest_route=\/wp\/v2\/posts\/1951"}],"collection":[{"href":"https:\/\/fayazrasheed.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/fayazrasheed.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/fayazrasheed.com\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/fayazrasheed.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=1951"}],"version-history":[{"count":32,"href":"https:\/\/fayazrasheed.com\/index.php?rest_route=\/wp\/v2\/posts\/1951\/revisions"}],"predecessor-version":[{"id":2074,"href":"https:\/\/fayazrasheed.com\/index.php?rest_route=\/wp\/v2\/posts\/1951\/revisions\/2074"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/fayazrasheed.com\/index.php?rest_route=\/wp\/v2\/media\/1974"}],"wp:attachment":[{"href":"https:\/\/fayazrasheed.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=1951"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/fayazrasheed.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=1951"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/fayazrasheed.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=1951"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}