{"id":637,"date":"2019-03-24T12:01:56","date_gmt":"2019-03-24T03:01:56","guid":{"rendered":"https:\/\/www.alprovs.com\/wordpress\/?p=637"},"modified":"2021-08-19T08:41:52","modified_gmt":"2021-08-18T23:41:52","slug":"centos7%e3%81%abtensorflowver1-13-1%e3%82%92build%e3%82%a4%e3%83%b3%e3%82%b9%e3%83%88%e3%83%bc%e3%83%abcuda10-1","status":"publish","type":"post","link":"https:\/\/www.alprovs.com\/wordpress\/?p=637","title":{"rendered":"CentOS7\u306bTensorFlow(Ver:1.13.1)\u3092Build\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb(CUDA:10.1)"},"content":{"rendered":"<p>\u4ee5\u524d\u306b\u3082\u5c0e\u5165\u8a18\u9332\u3092\u6b8b\u3057\u3066\u3044\u308b\u304c\uff08<a href=\"https:\/\/www.alprovs.com\/wordpress\/?p=549\" target=\"_blank\" rel=\"noopener noreferrer\">CentOS7(Ver:7.5)\u306bTensorFlow(Ver:1.8.0)\u5c0e\u5165<\/a>\uff09\u3001<br \/>\n\u6642\u9593\u304c\u7d4c\u904e\u3057\u3066\u30d0\u30fc\u30b8\u30e7\u30f3\u304c\u4e0a\u304c\u308a\u3001\u4ee5\u524d\u306e\u624b\u9806\u3067\u306f\u5c0e\u5165\u3067\u304d\u306a\u304b\u3063\u305f\u306e\u3067\u518d\u5ea6\u307e\u3068\u3081\u308b\u3002<\/p>\n<p>\u203b\u524d\u63d0\u3068\u3057\u3066CUDA\uff08Version 10.1\uff09\u304c\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u6e08\u307f\u306e\u3053\u3068<\/p>\n<p>\u3010\u76ee\u6b21\u3011<\/p>\n<ol>\n<li>\u74b0\u5883<\/li>\n<li>Python\u306e\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb(pyenv + Anaconda)<\/li>\n<li>cuDNN\u306e\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb<\/li>\n<li>Bazel\u306e\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb<\/li>\n<li>TensorFlow\u306e\u30a4\u30f3\u30b9\u30c8\u30c8\u30fc\u30eb<\/li>\n<\/ol>\n<h3>\uff11\uff0e\u74b0\u5883<\/h3>\n<ul>\n<li>OS : CentOS 7.6<\/li>\n<li>CUDA : 10.1<\/li>\n<li>cuDNN : 7.5<\/li>\n<li>Bazel : 1.19.2<\/li>\n<li>TensorFlow : 1.13.1<\/li>\n<li>Python : 3.6.0<\/li>\n<\/ul>\n<h3>\uff12\uff0ePython\u306e\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb<\/h3>\n<p>\u4e0b\u8a18\u30b5\u30a4\u30c8\u306e\u300cpyenv\u306e\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u300d\u301c\u300cPython\u306e\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u300d\u307e\u3067\u3092\u5b9f\u65bd<\/p>\n<p><a href=\"https:\/\/qiita.com\/harrynezumi\/items\/8e373a0563b92f3fc033\" target=\"_blank\" rel=\"noopener noreferrer\">CentOS7\u306bTensorFlow\u3092\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3059\u308b\u65b9\u6cd5<\/a><\/p>\n<h3>\uff13\uff0ecuDNN\u306e\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb<\/h3>\n<p>\u3053\u308c\u3092\u3057\u306a\u3044\u3068\u30d3\u30eb\u30c9\u6642\u306b\u30a8\u30e9\u30fc\u3068\u306a\u308b\u306e\u3067\u5c0e\u5165<\/p>\n<p>\u4e0b\u8a18\u306e\u30da\u30fc\u30b8\u306e\u300cDownload cuDNN\u300d\u304b\u3089\u30c0\u30a6\u30f3\u30ed\u30fc\u30c9\u3059\u308b\u3002<br \/>\n\u4eca\u56de\u306f\u300cDownload cuDNN v7.5.0 (Feb 25, 2019), for CUDA 10.1\u300d\u3092\u9078\u629e\u3057\u305f\u3002<br \/>\n\u203b\u30c0\u30a6\u30f3\u30ed\u30fc\u30c9\u3059\u308b\u305f\u3081\u306b\u306f\u30e1\u30f3\u30d0\u30fc\u30b7\u30c3\u30d7\u3078\u306e\u53c2\u52a0\u304c\u5fc5\u8981\u3067\u3059\u3002<\/p>\n<p><a href=\"https:\/\/developer.nvidia.com\/cudnn\" target=\"_blank\" rel=\"noopener noreferrer\">https:\/\/developer.nvidia.com\/cudnn<\/a><\/p>\n<p>\u30c0\u30a6\u30f3\u30ed\u30fc\u30c9\u3057\u305f\u30d5\u30a1\u30a4\u30eb\uff08cudnn-10.1-linux-x64-v7.5.0.56.tgz\uff09\u3092\u5c55\u958b\u3057\u3066\u914d\u7f6e\u3059\u308b\u3002<\/p>\n<pre class=\"EnlighterJSRAW\" data-enlighter-language=\"generic\">$ tar zxvf cudnn-10.1-linux-x64-v7.5.0.56.tgz\r\ncuda\/include\/cudnn.h\r\ncuda\/NVIDIA_SLA_cuDNN_Support.txt\r\ncuda\/lib64\/libcudnn.so\r\ncuda\/lib64\/libcudnn.so.7\r\ncuda\/lib64\/libcudnn.so.7.5.0\r\ncuda\/lib64\/libcudnn_static.a\r\n$ sudo cp .\/cuda\/include\/* \/usr\/local\/cuda-10.1\/include\/\r\n$ sudo cp .\/cuda\/lib64\/* \/usr\/local\/cuda-10.1\/lib64\/<\/pre>\n<p>&#8212; 2019\/05\/24 \u8ffd\u8a18 &#8212;<br \/>\n\u4e0a\u8a18\u306e\u65b9\u6cd5\u3060\u3068ldconfig\u3092\u5b9f\u884c\u3059\u308b\u3068\u6b21\u306e\u8b66\u544a\u304c\u767a\u751f\u3059\u308b\u305f\u3081\u3001\u4fee\u6b63\u3059\u308b\u3002<br \/>\n<code class=\"EnlighterJSRAW\" data-enlighter-language=\"generic\">ldconfig: \/usr\/local\/cuda-10.1\/targets\/x86_64-linux\/lib\/libcudnn.so.7 \u306f\u30b7\u30f3\u30dc\u30ea\u30c3\u30af\u30ea\u30f3\u30af\u3067\u306f\u3042\u308a\u307e\u305b\u3093<\/code><br \/>\n\u4ee5\u4e0b\u3001\u4fee\u6b63\u65b9\u6cd5\u3002\u4fee\u6b63\u5f8c\u306bldconfig\u3092\u5b9f\u884c\u3057\u3066\u8b66\u544a\u304c\u3067\u306a\u3051\u308c\u3070OK<\/p>\n<pre class=\"EnlighterJSRAW\" data-enlighter-language=\"generic\">$ cd \/usr\/local\/cuda-10.1\/targets\/x86_64-linux\/lib\r\n$ sudo rm libcudnn.so.7 libcudnn.so\r\n$ sudo ln -s libcudnn.so.7.5.0 libcudnn.so.7\r\n$ sudo ln -s libcudnn.so.7 libcudnn.so<\/pre>\n<p>&#8212;&#8212;&#8212;&#8212;&#8212;&#8212;&#8212;-<\/p>\n<h3>\uff14\uff0eBazel\u306e\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb<\/h3>\n<p>\u4eca\u56de\u306f\u30ea\u30dd\u30b8\u30c8\u30ea\u304b\u3089\u306e\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3092\u884c\u308f\u306a\u3044\u3002\uff08\u30ea\u30dd\u30b8\u30c8\u30ea\u306b\u5b58\u5728\u3059\u308b\u30d0\u30fc\u30b8\u30e7\u30f3\u304c\u65b0\u3057\u3059\u304e\u308b\u305f\u3081\uff09<br \/>\n\u4e0b\u8a18\u306e\u30da\u30fc\u30b8\u304b\u3089\u300c<span class=\"pl-2 flex-auto min-width-0 text-bold\">bazel-0.19.2-installer-linux-x8<\/span><span class=\"pl-2 flex-auto min-width-0 text-bold\">6_64.sh<\/span>\u300d\u3092\u30c0\u30a6\u30f3\u30ed\u30fc\u30c9\u3059\u308b\u3002\uff08\u7d50\u69cb\u6642\u9593\u304c\u304b\u304b\u308a\u307e\u3057\u305f\uff09<\/p>\n<p><a href=\"https:\/\/github.com\/bazelbuild\/bazel\/releases\" target=\"_blank\" rel=\"noopener noreferrer\">https:\/\/github.com\/bazelbuild\/bazel\/releases<\/a><\/p>\n<p>Bazel\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb<\/p>\n<pre class=\"EnlighterJSRAW\" data-enlighter-language=\"generic\"># bash bazel-0.19.2-installer-linux-x86_64.sh \r\nBazel installer\r\n---------------\r\n\r\nBazel is bundled with software licensed under the GPLv2 with Classpath exception.\r\nYou can find the sources next to the installer on our release page:\r\n   https:\/\/github.com\/bazelbuild\/bazel\/releases\r\n\r\n# Release 0.19.2 (2018-11-19)\r\n\r\nBaseline: ac880418885061d1039ad6b3d8c28949782e02d6\r\n\r\nCherry picks:\r\n\r\n   + 9bc3b20053a8b99bf2c4a31323a7f96fabb9f1ec:\r\n     Fix the \"nojava\" platform and enable full presubmit checks for\r\n     the various JDK platforms now that we have enough GCE resources.\r\n   + 54c2572a8cabaf2b29e58abe9f04327314caa6a0:\r\n     Add openjdk_linux_archive java_toolchain for nojava platform.\r\n   + 20bfdc67dc1fc32ffebbda7088ba49ee17e3e182:\r\n     Automated rollback of commit\r\n     19a401c38e30ebc0879925a5caedcbe43de0028f.\r\n   + 914b4ce14624171a97ff8b41f9202058f10d15b2:\r\n     Windows: Fix Precondition check for addDynamicInputLinkOptions\r\n   + 83d406b7da32d1b1f6dd02eae2fe98582a4556fd:\r\n     Windows, test-setup.sh: Setting RUNFILES_MANIFEST_FILE only when\r\n     it exists.\r\n   + e025726006236520f7e91e196b9e7f139e0af5f4:\r\n     Update turbine\r\n   + 5f312dd1678878fb7563eae0cd184f2270346352:\r\n     Fix event id for action_completed BEP events\r\n   + f0c844c77a2406518c4e75c49188390d5e281d3d:\r\n     Release 0.19.0 (2018-10-29)\r\n   + c3fb1db9e4e817e8a911f5b347b30f2674a82f7c:\r\n     Do not use CROSSTOOL to select cc_toolchain\r\n   + 8e280838e8896a6b5eb5421fda435b96b6f8de60:\r\n     Windows Add tests for msys gcc toolchain and mingw gcc toolchain\r\n   + fd52341505e725487c6bc6dfbe6b5e081aa037da:\r\n     update bazel-toolchains pin to latest release Part of changes to\r\n     allow bazelci to use 0.19.0 configs. RBE toolchain configs at or\r\n     before 0.17.0 are not compatible with bazel 0.19.0 or above.\r\n   + eb2af0f699350ad187048bf814a95af23f562c77:\r\n     Release 0.19.1 (2018-11-12)\r\n   + 6bc452874ddff69cbf7f66186238032283f1195f:\r\n     Also update cc_toolchain.toolchain_identifier when\r\n     CC_TOOLCHAIN_NAME is set\r\n   + f7e5aef145c33968f658eb2260e25630dc41cc67:\r\n     Add cc_toolchain targets for the new entries in the default\r\n     cc_toolchain_suite.\r\n   + 683c302129b66a8999f986be5ae7e642707e978c:\r\n     Read the CROSSTOOL from the package of the current cc_toolchain,\r\n     not from --crosstool_top\r\n\r\n- Fixes regression #6662, by fixing tools\/cpp\/BUILD\r\n- Fixes regression #6665, by setting the toolchain identifier.\r\n- CROSSTOOL file is now read from the package of cc_toolchain, not from the\r\n  package of cc_toolchain_suite. This is not expected to break anybody since\r\n  cc_toolchain_suite and cc_toolchain are commonly in the same package.\r\n\r\n## Build informations\r\n   - [Commit](https:\/\/github.com\/bazelbuild\/bazel\/commit\/6d6633a)\r\nUncompressing.......\r\n\r\nBazel is now installed!\r\n\r\nMake sure you have \"\/usr\/local\/bin\" in your path. You can also activate bash\r\ncompletion by adding the following line to your ~\/.bashrc:\r\n  source \/usr\/local\/lib\/bazel\/bin\/bazel-complete.bash\r\n\r\nSee http:\/\/bazel.build\/docs\/getting-started.html to start a new project!\r\n$ echo \"source \/usr\/local\/lib\/bazel\/bin\/bazel-complete.bash\" &gt;&gt; ~\/.bashrc<\/pre>\n<h3>\uff15\uff0eTensorFlow\u306e\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb<\/h3>\n<p>\u30d3\u30eb\u30c9\u524d\u306e\u6e96\u5099<\/p>\n<pre class=\"EnlighterJSRAW\" data-enlighter-language=\"generic\">$ sudo ln -s \/usr\/local\/cuda\/include\/crt\/math_functions.hpp \/usr\/local\/cuda\/include\/math_functions.hpp\r\n\r\n$ sudo ln -s \/usr\/lib64\/libcublas.so.10.1.0.105 \/usr\/local\/cuda-10.1\/lib64\/libcublas.so.10.1.0.105\r\n$ sudo ln -s \/usr\/local\/cuda-10.1\/lib64\/libcublas.so.10.1.0.105 \/usr\/local\/cuda-10.1\/lib64\/libcublas.so.10.1\r\n$ sudo ln -s \/usr\/local\/cuda-10.1\/lib64\/libcublas.so.10.1 \/usr\/local\/cuda-10.1\/lib64\/libcublas.so\r\n$ sudo ln -s \/usr\/local\/cuda-10.1\/targets\/x86_64-linux\/lib\/libcusolver.so.10.1.105 \/usr\/local\/cuda-10.1\/lib64\/libcusolver.so.10.1\r\n$ sudo ln -s \/usr\/local\/cuda-10.1\/targets\/x86_64-linux\/lib\/libcurand.so.10.1.105 \/usr\/local\/cuda-10.1\/lib64\/libcurand.so.10.1\r\n$ sudo ln -s \/usr\/local\/cuda-10.1\/targets\/x86_64-linux\/lib\/libcufft.so.10.1.105 \/usr\/local\/cuda-10.1\/lib64\/libcufft.so.10.1<\/pre>\n<p>\u30d3\u30eb\u30c9\u958b\u59cb \u203b\u7d50\u69cb\u6642\u9593\u304b\u304b\u308a\u307e\u3059&#8230;<\/p>\n<pre class=\"EnlighterJSRAW\" data-enlighter-language=\"generic\">$ git clone https:\/\/github.com\/tensorflow\/tensorflow.git\r\n$ cd tensorflow\r\n$ git checkout refs\/tags\/v1.13.1\r\n$ .\/configure\r\n# \u5fc5\u8981\u306b\u5fdc\u3058\u3066Yes No\u3092\u9078\u629e\r\n# \u4eca\u56de\u306fCUDA\u3092Yes\u306b\u3057\u3066\u30d0\u30fc\u30b8\u30e7\u30f3\u309210.1\u306b\r\n# \u307e\u305f\u3001cuDNN \u306b7.5\u3092\u6307\u5b9a\u3057\u305f\r\nWARNING: Running Bazel server needs to be killed, because the startup options are different.\r\nWARNING: --batch mode is deprecated. Please instead explicitly shut down your Bazel server using the command \"bazel shutdown\".\r\nYou have bazel 0.19.2 installed.\r\nPlease specify the location of python. [Default is \/home\/yuya\/.pyenv\/versions\/anaconda3-4.3.0\/bin\/python]:             \r\n\r\n\r\nFound possible Python library paths:\r\n  \/home\/yuya\/.pyenv\/versions\/anaconda3-4.3.0\/lib\/python3.6\/site-packages\r\nPlease input the desired Python library path to use.  Default is [\/home\/yuya\/.pyenv\/versions\/anaconda3-4.3.0\/lib\/python3.6\/site-packages]\r\n\r\nDo you wish to build TensorFlow with XLA JIT support? [Y\/n]: \r\nXLA JIT support will be enabled for TensorFlow.\r\n\r\nDo you wish to build TensorFlow with OpenCL SYCL support? [y\/N]: \r\nNo OpenCL SYCL support will be enabled for TensorFlow.\r\n\r\nDo you wish to build TensorFlow with ROCm support? [y\/N]: \r\nNo ROCm support will be enabled for TensorFlow.\r\n\r\nDo you wish to build TensorFlow with CUDA support? [y\/N]: y\r\nCUDA support will be enabled for TensorFlow.\r\n\r\nPlease specify the CUDA SDK version you want to use. [Leave empty to default to CUDA 10.0]: 10.1\r\n\r\n\r\nPlease specify the location where CUDA 10.1 toolkit is installed. Refer to README.md for more details. [Default is \/usr\/local\/cuda]: \r\n\r\n\r\nPlease specify the cuDNN version you want to use. [Leave empty to default to cuDNN 7]: 7.5\r\n\r\n\r\nPlease specify the location where cuDNN 7 library is installed. Refer to README.md for more details. [Default is \/usr\/local\/cuda]: \r\n\r\n\r\nDo you wish to build TensorFlow with TensorRT support? [y\/N]: \r\nNo TensorRT support will be enabled for TensorFlow.\r\n\r\nPlease specify the locally installed NCCL version you want to use. [Default is to use https:\/\/github.com\/nvidia\/nccl]: \r\n\r\n\r\nPlease specify a list of comma-separated Cuda compute capabilities you want to build with.\r\nYou can find the compute capability of your device at: https:\/\/developer.nvidia.com\/cuda-gpus.\r\nPlease note that each additional compute capability significantly increases your build time and binary size. [Default is: 6.1]: \r\n\r\n\r\nDo you want to use clang as CUDA compiler? [y\/N]: \r\nnvcc will be used as CUDA compiler.\r\n\r\nPlease specify which gcc should be used by nvcc as the host compiler. [Default is \/usr\/bin\/gcc]: \r\n\r\n\r\nDo you wish to build TensorFlow with MPI support? [y\/N]: \r\nNo MPI support will be enabled for TensorFlow.\r\n\r\nPlease specify optimization flags to use during compilation when bazel option \"--config=opt\" is specified [Default is -march=native -Wno-sign-compare]: \r\n\r\n\r\nWould you like to interactively configure .\/WORKSPACE for Android builds? [y\/N]: \r\nNot configuring the WORKSPACE for Android builds.\r\n\r\nPreconfigured Bazel build configs. You can use any of the below by adding \"--config=&lt;&gt;\" to your build command. See .bazelrc for more details.\r\n\t--config=mkl         \t# Build with MKL support.\r\n\t--config=monolithic  \t# Config for mostly static monolithic build.\r\n\t--config=gdr         \t# Build with GDR support.\r\n\t--config=verbs       \t# Build with libverbs support.\r\n\t--config=ngraph      \t# Build with Intel nGraph support.\r\n\t--config=dynamic_kernels\t# (Experimental) Build kernels into separate shared objects.\r\nPreconfigured Bazel build configs to DISABLE default on features:\r\n\t--config=noaws       \t# Disable AWS S3 filesystem support.\r\n\t--config=nogcp       \t# Disable GCP support.\r\n\t--config=nohdfs      \t# Disable HDFS support.\r\n\t--config=noignite    \t# Disable Apacha Ignite support.\r\n\t--config=nokafka     \t# Disable Apache Kafka support.\r\n\t--config=nonccl      \t# Disable NVIDIA NCCL support.\r\nConfiguration finished\r\n$ bazel build --config=opt --config=cuda \/\/tensorflow\/tools\/pip_package:build_pip_package<\/pre>\n<p>\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u30d1\u30c3\u30b1\u30fc\u30b8\u3092\u751f\u6210<br \/>\n<code class=\"EnlighterJSRAW\" data-enlighter-language=\"generic\">$ bazel-bin\/tensorflow\/tools\/pip_package\/build_pip_package \/tmp\/tensorflow_pkg<\/code><br \/>\n\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb<\/p>\n<pre class=\"EnlighterJSRAW\" data-enlighter-language=\"generic\">$ source activate tensorflow\r\n$ pip install \/tmp\/tensorflow_pkg\/tensorflow-1.13.1-cp36-cp36m-linux_x86_64.whl<\/pre>\n<p>\u52d5\u4f5c\u30c1\u30a7\u30c3\u30af<\/p>\n<pre class=\"EnlighterJSRAW\" data-enlighter-language=\"generic\">$ cd ~\/\r\n$ python\r\nPython 3.6.0 |Continuum Analytics, Inc.| (default, Dec 23 2016, 12:22:00) \r\n[GCC 4.4.7 20120313 (Red Hat 4.4.7-1)] on linux\r\nType \"help\", \"copyright\", \"credits\" or \"license\" for more information.\r\n&gt;&gt;&gt; import tensorflow as tf\r\n&gt;&gt;&gt; hello = tf.constant('Hello, TensorFlow!')\r\n&gt;&gt;&gt; sess = tf.Session()\r\n2019-03-24 11:58:58.077953: I tensorflow\/stream_executor\/cuda\/cuda_gpu_executor.cc:998] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\r\n2019-03-24 11:58:58.081272: I tensorflow\/compiler\/xla\/service\/service.cc:150] XLA service 0x3623d70 executing computations on platform CUDA. Devices:\r\n2019-03-24 11:58:58.081300: I tensorflow\/compiler\/xla\/service\/service.cc:158]   StreamExecutor device (0): GeForce GTX 1050 Ti, Compute Capability 6.1\r\n2019-03-24 11:58:58.084051: I tensorflow\/core\/platform\/profile_utils\/cpu_utils.cc:94] CPU Frequency: 3200050000 Hz\r\n2019-03-24 11:58:58.084286: I tensorflow\/compiler\/xla\/service\/service.cc:150] XLA service 0x368bfa0 executing computations on platform Host. Devices:\r\n2019-03-24 11:58:58.084308: I tensorflow\/compiler\/xla\/service\/service.cc:158]   StreamExecutor device (0): &lt;undefined&gt;, &lt;undefined&gt;\r\n2019-03-24 11:58:58.084948: I tensorflow\/core\/common_runtime\/gpu\/gpu_device.cc:1433] Found device 0 with properties: \r\nname: GeForce GTX 1050 Ti major: 6 minor: 1 memoryClockRate(GHz): 1.392\r\npciBusID: 0000:06:00.0\r\ntotalMemory: 3.95GiB freeMemory: 3.89GiB\r\n2019-03-24 11:58:58.084972: I tensorflow\/core\/common_runtime\/gpu\/gpu_device.cc:1512] Adding visible gpu devices: 0\r\n2019-03-24 11:58:58.088318: I tensorflow\/core\/common_runtime\/gpu\/gpu_device.cc:984] Device interconnect StreamExecutor with strength 1 edge matrix:\r\n2019-03-24 11:58:58.088342: I tensorflow\/core\/common_runtime\/gpu\/gpu_device.cc:990]      0 \r\n2019-03-24 11:58:58.088350: I tensorflow\/core\/common_runtime\/gpu\/gpu_device.cc:1003] 0:   N \r\n2019-03-24 11:58:58.088869: I tensorflow\/core\/common_runtime\/gpu\/gpu_device.cc:1115] Created TensorFlow device (\/job:localhost\/replica:0\/task:0\/device:GPU:0 with 3687 MB memory) -&gt; physical GPU (device: 0, name: GeForce GTX 1050 Ti, pci bus id: 0000:06:00.0, compute capability: 6.1)\r\n&gt;&gt;&gt; print(sess.run(hello))\r\nb'Hello, TensorFlow!'<\/pre>\n<p>\u4ee5\u4e0a<\/p>\n<p>&nbsp;<\/p>\n<p>\u53c2\u8003<\/p>\n<p><a href=\"https:\/\/www.tensorflow.org\/install\/source\" target=\"_blank\" rel=\"noopener noreferrer\">TensorFlow Build from source<\/a><\/p>\n<p><a href=\"https:\/\/github.com\/tensorflow\/tensorflow\/issues\/26150\" target=\"_blank\" rel=\"noopener noreferrer\">need guide to build with CUDA 10.1<\/a><\/p>\n<p><a href=\"https:\/\/qiita.com\/hiro_nishi\/items\/bea29f2345b1f21d015f\" target=\"_blank\" rel=\"noopener noreferrer\">Tensorflow\u3092CentOS7\u306b\u30aa\u30d5\u30e9\u30a4\u30f3\u74b0\u5883\u3067\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3059\u308b<\/a><\/p>\n<p>&nbsp;<\/p>\n","protected":false},"excerpt":{"rendered":"<p>\u4ee5\u524d\u306b\u3082\u5c0e\u5165\u8a18\u9332\u3092\u6b8b\u3057\u3066\u3044\u308b\u304c\uff08CentOS7(Ver:7.5)\u306bTensorFlow(Ver:1.8.0)\u5c0e\u5165\uff09\u3001 \u6642\u9593\u304c\u7d4c\u904e\u3057\u3066\u30d0\u30fc\u30b8\u30e7\u30f3\u304c\u4e0a\u304c\u308a\u3001\u4ee5\u524d\u306e\u624b\u9806\u3067\u306f\u5c0e\u5165\u3067\u304d\u306a\u304b\u3063\u305f\u306e\u3067\u518d\u5ea6\u307e\u3068\u3081\u308b\u3002 \u203b\u524d\u63d0\u3068\u3057\u3066CUD [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[3,6,10],"tags":[],"class_list":["post-637","post","type-post","status-publish","format-standard","hentry","category-linux","category-6","category-10"],"_links":{"self":[{"href":"https:\/\/www.alprovs.com\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/637","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.alprovs.com\/wordpress\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.alprovs.com\/wordpress\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.alprovs.com\/wordpress\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.alprovs.com\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=637"}],"version-history":[{"count":7,"href":"https:\/\/www.alprovs.com\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/637\/revisions"}],"predecessor-version":[{"id":804,"href":"https:\/\/www.alprovs.com\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/637\/revisions\/804"}],"wp:attachment":[{"href":"https:\/\/www.alprovs.com\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=637"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.alprovs.com\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=637"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.alprovs.com\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=637"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}