AWS Summit2017\uff5cDeep Learning with Apache MXNet on AWS<\/p>\n
\u3053\u306e\u65e5\u306e\u30bb\u30c3\u30b7\u30e7\u30f3\u306f DecOps\u3001 IoT\u3001 \u30d6\u30ed\u30c3\u30af\u30c1\u30a7\u30fc\u30f3\u3001 Deep Learning \u3068\u6700\u8fd1\u8eab\u8fd1\u306b\u306a\u3063\u3066\u304d\u3066\u3044\u308b\u6280\u8853\u3092 AWS\u3067\u5c55\u958b\u3059\u308b\u8a71\u3067\u3057\u305f
\n\u304b\u306a\u308a\u305f\u3081\u306b\u306a\u308b\u8a71\u304c\u591a\u304b\u3063\u305f\u3067\u3059
\n\u6700\u5f8c\u306b CTO\u30bb\u30c3\u30b7\u30e7\u30f3\u304c\u3042\u308a\u3001AWS\u306e CTO\uff08\u6700\u9ad8\u6280\u8853\u8cac\u4efb\u8005\uff09\u306e Werner Vogels \u306b\u5bfe\u3057\u3066 \u65e5\u672c\u306e\u30b9\u30bf\u30fc\u30c8\u30a2\u30c3\u30d7\u4f01\u696d\u306e CTO\u304c\u81ea\u793e\u30b5\u30fc\u30d3\u30b9\u3092\u30ec\u30d3\u30e5\u30fc\u3059\u308b\u30bb\u30c3\u30b7\u30e7\u30f3\u304c\u3042\u308a\u307e\u3057\u305f\u3002\u6280\u8853\u306e\u9078\u629e\u3068\u30b9\u30d4\u30fc\u30c9\u3068\u30b3\u30b9\u30c8\u3068\u30ea\u30bd\u30fc\u30b9\u3068\u3044\u308d\u3044\u308d\u9078\u629e\u30fb\u5224\u65ad\u3059\u308b CTO\u306f\u5927\u5909\u3067\u3059
\nIntel\u4e3b\u50ac\u306e\u30ca\u30a4\u30c8\u30bb\u30c3\u30b7\u30e7\u30f3\u3082\u3042\u308a\u307e\u3057\u305f\u304c \u304a\u9152\u304c\u3044\u3063\u3071\u3044\u3067\u9003\u3052\u51fa\u3057\u307e\u3057\u305f<\/p>\n
\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014<\/p>\n
Day3 Dev Day Track1 \u2013 D3T7-5<\/p>\n
Deep Learning with Apache MXNet on AWS<\/p>\n
Deep Learning \u306e\u30bb\u30c3\u30b7\u30e7\u30f3\u306f\u65e9\u3005\u306b\u4e88\u7d04\u3044\u3063\u3071\u3044\u3067\u4eba\u6c17\u3067\u3057\u305f\u304c \u5f53\u65e5\u306b\u306a\u3063\u3066\u307f\u305f\u3089\u5e2d\u306b\u4f59\u88d5\u304c\u3042\u3063\u305f\u306e\u3067\u516c\u8074\u3057\u307e\u3057\u305f<\/p>\n
MXNet \u306f AWS \u304c\u63a8\u3057\u3066\u3044\u308b Deep Learning \u306e\u57fa\u76e4\u3067\u3059
\nApache Fundation \u306a\u306e\u3067 \u30aa\u30fc\u30d7\u30f3\u30bd\u30fc\u30b9\u3067\u3042\u308a\u3001\u5229\u7528\u65b9\u6cd5\u3082 Google \u306e TensorFlow \u3084 MS \u306e Cognitive Toolkit \u3088\u308a\u4f7f\u3044\u3084\u3059\u304f\u3001\u30de\u30b7\u30f3\u30b3\u30fc\u30c9\u3092\u8a18\u8ff0\u3067\u304d\u308b\u30d7\u30ed\u30b0\u30e9\u30df\u30f3\u30b0\u8a00\u8a9e\u304c\u591a\u3044\u306e\u304c\u7279\u5fb4\u3067\u3059<\/p>\n
Deep Learning \u3067\u306f Input \u3055\u308c\u305f\u30c7\u30fc\u30bf\u3092\u69d8\u3005\u306a\u89d2\u5ea6\u3067\u89e3\u6790\u3057\u305f\u7d50\u679c\u306e Output \u3092\u691c\u8a3c\u3057 Output \u306e\u7cbe\u5ea6\u3092\u9ad8\u3081\u308b\u3053\u3068\u3067\u5b66\u7fd2\u3092\u6df1\u3081\u308b\u65b9\u6cd5\u3067\u3059
\n\u89e3\u6790\u306b\u968e\u5c64\u304c\u3042\u308b\u306e\u3067 Deep \u306a\u3093\u3060\u305d\u3046\u3067\u3059<\/p>\n
\u5b9f\u969b\u306e\u30c7\u30e2\u3067\u306f EC2 \u306b\u7528\u610f\u3055\u308c\u3066\u3044\u308b AMI \u3067 MXNet \u5165\u308a\u306e AmazonLinux \u3092 GPU\u30a4\u30f3\u30b9\u30bf\u30f3\u30b9\u3067\u8d77\u52d5\u3057\u3001\u30c6\u30b9\u30c8\u30c7\u30fc\u30bf\u3068\u3057\u3066 \u624b\u66f8\u304d\u306e\u6570\u5b57\u3092\u8aad\u307f\u8fbc\u307e\u305b\u3066\u5b66\u7fd2\u3057\u305f\u3042\u3068\u3001\u30de\u30a6\u30b9\u3067\u624b\u66f8\u304d\u3057\u305f\u6570\u5b57\u304c\u8a8d\u8b58\u3055\u308c\u308b\u3068\u3053\u308d\u307e\u3067\u3092 \u6570\u5206\u3067\u5b9f\u8df5\u3057\u3066\u3044\u307e\u3057\u305f<\/p>\n
\u4eca\u5f8c\u30b7\u30b9\u30c6\u30e0\u958b\u767a\u306f SE \u3067\u306f\u306a\u304f \u30de\u30b7\u30f3\u30b3\u30fc\u30c9\u306e\u958b\u767a\u306b\u30b7\u30d5\u30c8\u3059\u308b\u306e\u304b\u3082\u3057\u308c\u307e\u305b\u3093
\nAWS\u306e\u3072\u3068\u306f\u5c06\u6765\u7684\u306b\u305d\u3046\u306a\u308b\u3063\u3066\u8a00\u3063\u3066\u3057\u305f<\/p>\n
Deep Learning \u3063\u3066 \u7c21\u5358\u3067\u9762\u767d\u305d\u3046\u3067\u3059\u3057 \u5b66\u7fd2\u3055\u305b\u308b\u305f\u3081\u306e\u30de\u30b7\u30f3\u30b3\u30fc\u30c9\u306f\u305d\u308c\u307b\u3069\u591a\u304f\u306e\u30b3\u30fc\u30c9\u3092\u66f8\u304f\u5fc5\u8981\u304c\u306a\u3044\u3088\u3046\u3067\u3059
\nMXNet \u3067 Deep Learning \u306e\u6577\u5c45\u304c\u4f4e\u304f\u306a\u3063\u305f\u306e\u3067\u306f\u306a\u3044\u3067\u3057\u3087\u3046\u304b<\/p>\n
GPU\u30a4\u30f3\u30b9\u30bf\u30f3\u30b9\u306f \u304a\u9ad8\u3044\u306e\u3067 \u3054\u5229\u7528\u306f\u8a08\u753b\u7684\u306b<\/p>\n
\u30ad\u30fc\u30ef\u30fc\u30c9
\nDeep Learning (\u6df1\u5c64\u5b66\u7fd2\u30e2\u30c7\u30eb\u3001\u2252AI) , Apache MXNet<\/p>\n","protected":false},"featured_media":0,"template":"","news_category":[71],"class_list":["post-6294","news","type-news","status-publish","hentry","news_category-blog-aws"],"_links":{"self":[{"href":"https:\/\/www.fsz.co.jp\/index.php?rest_route=\/wp\/v2\/news\/6294","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.fsz.co.jp\/index.php?rest_route=\/wp\/v2\/news"}],"about":[{"href":"https:\/\/www.fsz.co.jp\/index.php?rest_route=\/wp\/v2\/types\/news"}],"wp:attachment":[{"href":"https:\/\/www.fsz.co.jp\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=6294"}],"wp:term":[{"taxonomy":"news_category","embeddable":true,"href":"https:\/\/www.fsz.co.jp\/index.php?rest_route=%2Fwp%2Fv2%2Fnews_category&post=6294"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}