{"id":345,"date":"2023-08-23T12:46:21","date_gmt":"2023-08-23T03:46:21","guid":{"rendered":"http:\/\/192.168.0.139:8080\/2023\/08\/23\/nvidia%e3%81%ae%e3%82%b0%e3%83%a9%e3%83%9c%e3%82%92%e6%90%ad%e8%bc%89%e3%81%97%e3%81%9fpc%e3%81%abcuda%e3%82%92%e3%82%a4%e3%83%b3%e3%82%b9%e3%83%88%e3%83%bc%e3%83%ab%e3%81%97%e3%81%9f%e3%81%bb\/"},"modified":"2023-08-23T12:46:21","modified_gmt":"2023-08-23T03:46:21","slug":"nvidia%e3%81%ae%e3%82%b0%e3%83%a9%e3%83%9c%e3%82%92%e6%90%ad%e8%bc%89%e3%81%97%e3%81%9fpc%e3%81%abcuda%e3%82%92%e3%82%a4%e3%83%b3%e3%82%b9%e3%83%88%e3%83%bc%e3%83%ab%e3%81%97%e3%81%9f%e3%81%bb","status":"publish","type":"post","link":"http:\/\/192.168.0.139:8080\/2023\/08\/23\/345\/","title":{"rendered":"NVIDIA\u306e\u30b0\u30e9\u30dc\u3092\u642d\u8f09\u3057\u305fPC\u306bCUDA\u3092\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3057\u305f\u307b\u3046\u304c\u3044\u3044\u306e\u304b\uff1f"},"content":{"rendered":"\n<p>NVIDIA\u306eGPU\u3092\u642d\u8f09\u3057\u305f\u30d1\u30bd\u30b3\u30f3\u3067\u3001\u6df1\u5c64\u5b66\u7fd2\u3084GPU\u3092\u6d3b\u7528\u3057\u305f\u8a08\u7b97\u51e6\u7406\u3092\u884c\u3044\u305f\u3044\u5834\u5408\u3001CUDA\uff08Compute Unified Device Architecture\uff09\u3092\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3059\u308b\u3053\u3068\u304c\u63a8\u5968\u3055\u308c\u307e\u3059\u3002CUDA\u306fNVIDIA\u304c\u63d0\u4f9b\u3059\u308b\u4e26\u5217\u8a08\u7b97\u30d7\u30e9\u30c3\u30c8\u30d5\u30a9\u30fc\u30e0\u3068API\u30bb\u30c3\u30c8\u3067\u3001GPU\u4e0a\u3067\u306e\u4e26\u5217\u51e6\u7406\u3092\u5bb9\u6613\u306b\u884c\u3046\u3053\u3068\u304c\u3067\u304d\u307e\u3059\u3002<\/p>\n\n\n\n<p>\u5148\u65e5NVIDIA\u306e\u30b0\u30e9\u30dc\u642d\u8f09\u306ePC\u3092\u8cfc\u5165\u3057\u305f\u306e\u3067\u3059\u304c\u3001CUDA\u306f\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3055\u308c\u3066\u3044\u307e\u305b\u3093\u3067\u3057\u305f\u3002\u78ba\u8a8d\u65b9\u6cd5\u306b\u306f\u4e0b\u8a18\u304c\u3042\u308a\u307e\u3059\u3002<\/p>\n\n\n\n<p>\u30fb\u30b3\u30de\u30f3\u30c9\u30d7\u30ed\u30f3\u30d7\u30c8\u3092\u958b\u3044\u3066\u3001nvcc -V\u3068\u5165\u529b\u3057\u3066\u30a8\u30f3\u30bf\u30fc\u30ad\u30fc\u3092\u62bc\u3057\u307e\u3059\u3002\u3053\u306e\u30b3\u30de\u30f3\u30c9\u306f\u3001CUDA\u30b3\u30f3\u30d1\u30a4\u30e9\u306e\u30d0\u30fc\u30b8\u30e7\u30f3\u3092\u8868\u793a\u3057\u307e\u3059\u3002\u3082\u3057CUDA\u304c\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3055\u308c\u3066\u3044\u308c\u3070\u3001\u4ee5\u4e0b\u306e\u3088\u3046\u306a\u51fa\u529b\u304c\u8868\u793a\u3055\u308c\u307e\u3059\u3002<\/p>\n\n\n\n<p>nvcc: NVIDIA (R) Cuda compiler driver<br>Copyright (c) 2005-2020 NVIDIA Corporation<br>Built on Mon_Nov_30_19:15:10_Pacific_Standard_Time_2020<br>Cuda compilation tools, release 11.2, V11.2.67<br>Build cuda_11.2.r11.2\/compiler.29373293_0<\/p>\n\n\n\n<figure class=\"wp-block-image size-large\"><img decoding=\"async\" src=\"https:\/\/minokamo.xyz\/img\/blog\/2023\/08\/20230823.png\" alt=\"\"\/><\/figure>\n\n\n\n<p>\u3053\u306e\u74b0\u5883\u3067\u306f\u4e0b\u8a18\u306e\u30e1\u30c3\u30bb\u30fc\u30b8\u304c\u8868\u793a\u3055\u308c\u307e\u3057\u305f\u3002<\/p>\n\n\n\n<p>nvcc: The term &#8216;nvcc&#8217; is not recognized as a name of a cmdlet, function, script file, or executable program.<br>Check the spelling of the name, or if a path was included, verify that the path is correct and try again.<\/p>\n\n\n\n<p>\u30fb\u30a8\u30af\u30b9\u30d7\u30ed\u30fc\u30e9\u30fc\u3092\u958b\u3044\u3066\u3001C:Program FilesNVIDIA GPU Computing ToolkitCUDA\u3068\u3044\u3046\u30d5\u30a9\u30eb\u30c0\u3092\u63a2\u3057\u307e\u3059\u3002\u3053\u306e\u30d5\u30a9\u30eb\u30c0\u306f\u3001CUDA\u306e\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u5148\u306e\u30c7\u30d5\u30a9\u30eb\u30c8\u306e\u5834\u6240\u3067\u3059\u3002\u3082\u3057CUDA\u304c\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3055\u308c\u3066\u3044\u308c\u3070\u3001\u3053\u306e\u30d5\u30a9\u30eb\u30c0\u306e\u4e2d\u306b\u30d0\u30fc\u30b8\u30e7\u30f3\u3054\u3068\u306e\u30b5\u30d6\u30d5\u30a9\u30eb\u30c0\u304c\u5b58\u5728\u3057\u307e\u3059\u3002<\/p>\n\n\n\n<p>\u30fbNVIDIA\u30b3\u30f3\u30c8\u30ed\u30fc\u30eb\u30d1\u30cd\u30eb\u3092\u958b\u3044\u3066\u3001\u5de6\u5074\u306e\u30e1\u30cb\u30e5\u30fc\u304b\u3089\u300c\u30b7\u30b9\u30c6\u30e0\u60c5\u5831\u300d\u3092\u9078\u629e\u3057\u307e\u3059\u3002\u53f3\u5074\u306e\u30a6\u30a3\u30f3\u30c9\u30a6\u306b\u8868\u793a\u3055\u308c\u308b\u300c\u30b3\u30f3\u30dd\u30fc\u30cd\u30f3\u30c8\u300d\u306e\u9805\u76ee\u306b\u300cCUDA\u300d\u304c\u542b\u307e\u308c\u3066\u3044\u308c\u3070\u3001CUDA\u304c\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3055\u308c\u3066\u3044\u308b\u3053\u3068\u3092\u793a\u3057\u307e\u3059\u3002<\/p>\n\n\n\n<p>\u4e00\u65b9\u3067\u3001\u30b2\u30fc\u30e0\u3084\u4e00\u822c\u7684\u306a\u30bf\u30b9\u30af\u306b\u306fCUDA\u306f\u5fc5\u8981\u3067\u306f\u3042\u308a\u307e\u305b\u3093\u3002\u30b2\u30fc\u30e0\u3067\u306f\u3001\u30b0\u30e9\u30d5\u30a3\u30c3\u30af\u30dc\u30fc\u30c9\u306b\u642d\u8f09\u3055\u308c\u305fGPU\u304c\u753b\u50cf\u3084\u6620\u50cf\u306e\u63cf\u5199\u3092\u884c\u3044\u307e\u3059\u304c\u3001\u305d\u306e\u969b\u306bCUDA\u3092\u4f7f\u3046\u5fc5\u8981\u306f\u3042\u308a\u307e\u305b\u3093\u3002\u4e00\u822c\u7684\u306a\u30bf\u30b9\u30af\u3067\u306f\u3001CPU\u304c\u9023\u7d9a\u7684\u306b\u30bf\u30b9\u30af\u3092\u884c\u3044\u3001\u5e45\u5e83\u3044\u51e6\u7406\u3092\u3059\u308b\u3053\u3068\u3067\u6a5f\u80fd\u3057\u307e\u3059\u304c\u3001\u305d\u306e\u969b\u306bCUDA\u3092\u4f7f\u3046\u5fc5\u8981\u306f\u3042\u308a\u307e\u305b\u3093\u3002\u30bd\u30fc\u30b9\u306f\u4e0b\u8a18\u306e\u30b9\u30af\u30b7\u30e7\u3067\u3059\u3002<\/p>\n\n\n\n<figure class=\"wp-block-image size-large\"><img decoding=\"async\" src=\"https:\/\/minokamo.xyz\/img\/blog\/2023\/08\/202308231.png\" alt=\"\"\/><\/figure>\n\n\n\n<p>\u3068\u3044\u3046\u308f\u3051\u3067\u3001\u6700\u7d42\u7684\u306b\u306f\u4f7f\u7528\u76ee\u7684\u3084\u30cb\u30fc\u30ba\u306b\u5fdc\u3058\u3066CUDA\u306e\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3092\u691c\u8a0e\u3057\u3066\u304f\u3060\u3055\u3044\u3002\u4eca\u8a71\u984c\u306estable diffusion\u306e\u3088\u3046\u306a\u753b\u50cf\u751f\u6210\u306b\u306f\u5fc5\u8981\u3067\u3059\u3002\u307e\u305f\u3001neutrino diffusion\u306f\u3001AI\u306b\u6b4c\u308f\u305b\u308b\u3053\u3068\u304c\u3067\u304d\u308b\u30cb\u30e5\u30fc\u30e9\u30eb\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u30d9\u30fc\u30b9\u306e\u6b4c\u58f0\u5408\u6210\u30bd\u30d5\u30c8\u3067\u3059\u3002\u79c1\u304c\u591a\u7528\u3057\u3066\u3044\u308bVOICEVOX\u306f\u3001\u7121\u6599\u3067\u4f7f\u3048\u308b\u4e2d\u54c1\u8cea\u306a\u30c6\u30ad\u30b9\u30c8\u8aad\u307f\u4e0a\u3052\u30bd\u30d5\u30c8\u30a6\u30a7\u30a2\u3067\u3059\u3002<br>PowerDirector\u3067\u306fCUDA\u306f\u5fc5\u8981\u3067\u306f\u3042\u308a\u307e\u305b\u3093\u304c\u3001\u5229\u7528\u3059\u308b\u3053\u3068\u3067\u52d5\u753b\u306e\u30a8\u30f3\u30b3\u30fc\u30c9\u3084\u30a8\u30d5\u30a7\u30af\u30c8\u306e\u30ec\u30f3\u30c0\u30ea\u30f3\u30b0\u3092\u9ad8\u901f\u5316\u3067\u304d\u307e\u3059\u3002<\/p>\n\n\n\n<p><br>CUDA\u3092\u30b5\u30dd\u30fc\u30c8\u3059\u308b\u30aa\u30fc\u30d7\u30f3\u30bd\u30fc\u30b9\u3084\u7121\u511f\u306e\u30bd\u30d5\u30c8\u30a6\u30a7\u30a2\u306f\u6570\u591a\u304f\u5b58\u5728\u3057\u307e\u3059\u3002\u4ee5\u4e0b\u306f\u305d\u306e\u4e00\u90e8\u306e\u30ea\u30b9\u30c8\u3067\u3059\uff1a<\/p>\n\n\n\n<ol>\n<li><strong>\u6df1\u5c64\u5b66\u7fd2\u30d5\u30ec\u30fc\u30e0\u30ef\u30fc\u30af<\/strong>:\n<ul>\n<li><strong>TensorFlow<\/strong>: Google\u304c\u958b\u767a\u3057\u305f\u6df1\u5c64\u5b66\u7fd2\u30d5\u30ec\u30fc\u30e0\u30ef\u30fc\u30af\u3002\u5e83\u304f\u5229\u7528\u3055\u308c\u3066\u304a\u308a\u3001\u591a\u304f\u306e\u30a2\u30d7\u30ea\u30b1\u30fc\u30b7\u30e7\u30f3\u3084\u7814\u7a76\u3067\u4f7f\u7528\u3055\u308c\u3066\u3044\u307e\u3059\u3002<\/li>\n\n\n\n<li><strong>PyTorch<\/strong>: Facebook\u304c\u958b\u767a\u3057\u305f\u6df1\u5c64\u5b66\u7fd2\u30d5\u30ec\u30fc\u30e0\u30ef\u30fc\u30af\u3002\u7814\u7a76\u30b3\u30df\u30e5\u30cb\u30c6\u30a3\u3067\u975e\u5e38\u306b\u4eba\u6c17\u304c\u3042\u308a\u307e\u3059\u3002<\/li>\n\n\n\n<li><strong>Keras<\/strong>: TensorFlow\u306e\u4e0a\u3067\u52d5\u4f5c\u3059\u308b\u9ad8\u30ec\u30d9\u30eb\u306e\u30cb\u30e5\u30fc\u30e9\u30eb\u30cd\u30c3\u30c8\u30ef\u30fc\u30afAPI\u3002\u521d\u5fc3\u8005\u306b\u3082\u4f7f\u3044\u3084\u3059\u3044\u3067\u3059\u3002<\/li>\n\n\n\n<li><strong>Caffe<\/strong>: Berkeley Vision and Learning Center (BVLC)\u306b\u3088\u3063\u3066\u958b\u767a\u3055\u308c\u305f\u30d5\u30ec\u30fc\u30e0\u30ef\u30fc\u30af\u3002<\/li>\n\n\n\n<li><strong>Neural Network Console<\/strong>: Neural Network Console\u306f\u3001Sony\u304c\u63d0\u4f9b\u3059\u308b\u6df1\u5c64\u5b66\u7fd2\u306e\u958b\u767a\u30c4\u30fc\u30eb\u3067\u3059\u3002\u30b0\u30e9\u30d5\u30a3\u30ab\u30eb\u30e6\u30fc\u30b6\u30fc\u30a4\u30f3\u30bf\u30fc\u30d5\u30a7\u30fc\u30b9\u3092\u6301\u3064\u3053\u306e\u30c4\u30fc\u30eb\u306f\u3001\u30c7\u30a3\u30fc\u30d7\u30e9\u30fc\u30cb\u30f3\u30b0\u30e2\u30c7\u30eb\u306e\u8a2d\u8a08\u3001\u5b66\u7fd2\u3001\u8a55\u4fa1\u3092\u76f4\u611f\u7684\u306b\u884c\u3046\u3053\u3068\u304c\u3067\u304d\u307e\u3059\u3002<br>\uff11\u3001\u30c9\u30e9\u30c3\u30b0\uff06\u30c9\u30ed\u30c3\u30d7\u306e\u8a2d\u8a08: \u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u306e\u30a2\u30fc\u30ad\u30c6\u30af\u30c1\u30e3\u3092\u76f4\u611f\u7684\u306b\u30c7\u30b6\u30a4\u30f3\u3067\u304d\u307e\u3059\u3002<br>\uff12\u3001\u81ea\u52d5\u5316\u3055\u308c\u305f\u5b66\u7fd2: \u5b66\u7fd2\u306e\u8a2d\u5b9a\u3084\u6700\u9069\u5316\u306e\u30d7\u30ed\u30bb\u30b9\u3092\u7c21\u5358\u306b\u884c\u3048\u307e\u3059\u3002<br>\uff13\u3001\u30d3\u30b8\u30e5\u30a2\u30eb\u8a55\u4fa1: \u5b66\u7fd2\u306e\u9032\u884c\u72b6\u6cc1\u3084\u7d50\u679c\u3092\u30ea\u30a2\u30eb\u30bf\u30a4\u30e0\u3067\u8996\u899a\u7684\u306b\u78ba\u8a8d\u3067\u304d\u307e\u3059\u3002<br>\uff14\u3001CUDA\u30b5\u30dd\u30fc\u30c8: NVIDIA\u306eGPU\u3092\u6d3b\u7528\u3057\u3066\u3001\u9ad8\u901f\u306a\u5b66\u7fd2\u3084\u8a55\u4fa1\u304c\u53ef\u80fd\u3067\u3059\u3002<\/li>\n<\/ul>\n<\/li>\n\n\n\n<li><strong>GPU\u8a08\u7b97\u30e9\u30a4\u30d6\u30e9\u30ea<\/strong>:\n<ul>\n<li><strong>cuDNN<\/strong>: NVIDIA\u304c\u63d0\u4f9b\u3059\u308b\u6df1\u5c64\u5b66\u7fd2\u7528\u306eGPU\u52a0\u901f\u30e9\u30a4\u30d6\u30e9\u30ea\u3002<\/li>\n\n\n\n<li><strong>cuBLAS<\/strong>: CUDA\u5bfe\u5fdc\u306eBLAS\u30e9\u30a4\u30d6\u30e9\u30ea\u3067\u3001\u7dda\u5f62\u4ee3\u6570\u306e\u9ad8\u901f\u8a08\u7b97\u3092\u30b5\u30dd\u30fc\u30c8\u3057\u307e\u3059\u3002<\/li>\n\n\n\n<li><strong>cuFFT<\/strong>: CUDA\u5bfe\u5fdc\u306e\u9ad8\u901f\u30d5\u30fc\u30ea\u30a8\u5909\u63db\u30e9\u30a4\u30d6\u30e9\u30ea\u3002<\/li>\n<\/ul>\n<\/li>\n\n\n\n<li><strong>GPU\u30d7\u30ed\u30b0\u30e9\u30df\u30f3\u30b0<\/strong>:\n<ul>\n<li><strong>Numba<\/strong>: Python\u306e\u30b8\u30e3\u30b9\u30c8\u30a4\u30f3\u30bf\u30a4\u30e0\u30b3\u30f3\u30d1\u30a4\u30e9\u3002CUDA\u30b3\u30fc\u30c9\u3092Python\u304b\u3089\u7c21\u5358\u306b\u66f8\u304f\u3053\u3068\u304c\u3067\u304d\u307e\u3059\u3002<\/li>\n\n\n\n<li><strong>Thrust<\/strong>: C++\u30c6\u30f3\u30d7\u30ec\u30fc\u30c8\u30e9\u30a4\u30d6\u30e9\u30ea\u3067\u3001CUDA\u3084OpenMP\u306a\u3069\u306e\u30d0\u30c3\u30af\u30a8\u30f3\u30c9\u3092\u4f7f\u7528\u3057\u3066\u4e26\u5217\u30a2\u30eb\u30b4\u30ea\u30ba\u30e0\u3092\u5b9f\u88c5\u3057\u307e\u3059\u3002<\/li>\n<\/ul>\n<\/li>\n\n\n\n<li><strong>\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3 &amp; \u30ec\u30f3\u30c0\u30ea\u30f3\u30b0<\/strong>:\n<ul>\n<li><strong>Blender<\/strong>: 3D\u30b3\u30f3\u30c6\u30f3\u30c4\u4f5c\u6210\u30b9\u30a4\u30fc\u30c8\u3002Cycles\u3068\u3044\u3046\u30ec\u30f3\u30c0\u30e9\u304cCUDA\u3092\u30b5\u30dd\u30fc\u30c8\u3057\u3066\u304a\u308a\u3001GPU\u30ec\u30f3\u30c0\u30ea\u30f3\u30b0\u304c\u53ef\u80fd\u3067\u3059\u3002<\/li>\n<\/ul>\n<\/li>\n\n\n\n<li><strong>\u305d\u306e\u4ed6<\/strong>:\n<ul>\n<li><strong>Darknet<\/strong>: YOLO (You Only Look Once) \u3068\u3044\u3046\u7269\u4f53\u691c\u51fa\u30a2\u30eb\u30b4\u30ea\u30ba\u30e0\u306e\u305f\u3081\u306e\u30d5\u30ec\u30fc\u30e0\u30ef\u30fc\u30af\u3002CUDA\u3092\u30b5\u30dd\u30fc\u30c8\u3057\u3066\u3044\u307e\u3059\u3002<\/li>\n<\/ul>\n<\/li>\n<\/ol>\n\n\n\n<figure class=\"wp-block-embed is-type-video is-provider-youtube wp-block-embed-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"wp-block-embed__wrapper\">\n<iframe loading=\"lazy\" title=\"NVIDIA\u306e\u30b0\u30e9\u30dc\u3092\u642d\u8f09\u3057\u305fPC\u306bCUDA\u3092\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3057\u305f\u307b\u3046\u304c\u3044\u3044\u306e\u304b\uff1f\" width=\"500\" height=\"281\" src=\"https:\/\/www.youtube.com\/embed\/Ja7FV-HpE-A?feature=oembed\" frameborder=\"0\" allow=\"accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share\" referrerpolicy=\"strict-origin-when-cross-origin\" allowfullscreen><\/iframe>\n<\/div><\/figure>\n","protected":false},"excerpt":{"rendered":"<p>NVIDIA\u306eGPU\u3092\u642d\u8f09\u3057\u305f\u30d1\u30bd\u30b3\u30f3\u3067\u3001\u6df1\u5c64\u5b66\u7fd2\u3084GPU\u3092\u6d3b\u7528\u3057\u305f\u8a08\u7b97\u51e6\u7406\u3092\u884c\u3044\u305f\u3044\u5834\u5408\u3001CUDA\uff08Compute Unified Device Architecture\uff09\u3092\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3059\u308b\u3053\u3068\u304c\u63a8\u5968\u3055\u308c\u307e\u3059\u3002CUDA [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[],"_links":{"self":[{"href":"http:\/\/192.168.0.139:8080\/wp-json\/wp\/v2\/posts\/345"}],"collection":[{"href":"http:\/\/192.168.0.139:8080\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"http:\/\/192.168.0.139:8080\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"http:\/\/192.168.0.139:8080\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"http:\/\/192.168.0.139:8080\/wp-json\/wp\/v2\/comments?post=345"}],"version-history":[{"count":0,"href":"http:\/\/192.168.0.139:8080\/wp-json\/wp\/v2\/posts\/345\/revisions"}],"wp:attachment":[{"href":"http:\/\/192.168.0.139:8080\/wp-json\/wp\/v2\/media?parent=345"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/192.168.0.139:8080\/wp-json\/wp\/v2\/categories?post=345"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/192.168.0.139:8080\/wp-json\/wp\/v2\/tags?post=345"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}