{"id":46395,"date":"2023-01-24T02:55:21","date_gmt":"2023-08-12T15:57:07","guid":{"rendered":"https:\/\/www.silicloud.com\/zh\/blog\/46395-2\/"},"modified":"2024-04-29T11:12:10","modified_gmt":"2024-04-29T03:12:10","slug":"46395-2","status":"publish","type":"post","link":"https:\/\/www.silicloud.com\/zh\/blog\/46395-2\/","title":{"rendered":""},"content":{"rendered":"

Jupyter-tensorboard\u306e\u7d39\u4ecb\u3068Keras\u304b\u3089\u4f7f\u3063\u3066\u307f\u308b<\/h1>\n

Jupyter Notebook\u306b\u7d71\u5408\u3055\u308c\u305fTensorboard\u306e\u7d39\u4ecb\u3067\u3059\u3002
\nhttps:\/\/github.com\/lspvic\/jupyter_tensorboard<\/p>\n

\u30c7\u30a3\u30fc\u30d7\u30e9\u30fc\u30cb\u30f3\u30b0\u306e\u30e2\u30c7\u30eb\u958b\u767a\u3092\u3059\u308b\u969b\u3001\u3072\u308d\u304f\u4f7f\u308f\u308c\u308b\u958b\u767a\u30a4\u30f3\u30bf\u30fc\u30d5\u30a7\u30a4\u30b9\u304cJupyter Notebook\u3067\u3059\u3002
\n\u52a0\u3048\u3066\u30c7\u30a3\u30fc\u30d7\u30e9\u30fc\u30cb\u30f3\u30b0\u306e\u30d5\u30ec\u30fc\u30e0\u30ef\u30fc\u30af\u3068\u3057\u3066\u983b\u7e41\u306b\u4f7f\u308f\u308c\u308b2\u5f37\u304cTensorflow\u3068Keras\u3060\u3068\u601d\u3044\u307e\u3059\u3002
\nKeras\u81ea\u4f53\u304cTheano\u3084Tensorflow\u3092\u30ab\u30d0\u30fc\u3059\u308b\u30cf\u30a4\u30ec\u30d9\u30eb\u30fb\u30e9\u30a4\u30d6\u30e9\u30ea\u306e\u305f\u3081\u3001\u5b9f\u8ceaTensorflow\u304c\u4f7f\u308f\u308c\u3066\u3044\u308b\u306e\u3067\u3059\u304c\u3001Tensorflow\u306b\u306fTensorboard\u3068\u3044\u3046\u4fbf\u5229\u306a\u30a6\u30a7\u30d6\u30a2\u30d7\u30ea\u30b1\u30fc\u30b7\u30e7\u30f3\u304c\u3042\u308a\u307e\u3059\u3002
\nJupyter-tensorboard\u306fJupyter\u4e0a\u3067Tensorboard\u3092\u8d77\u52d5\u3059\u308b\u3053\u3068\u304c\u3067\u304d\u308b\u30d7\u30e9\u30b0\u30a4\u30f3\u3067\u3059\u3002<\/p>\n

\"101.PNG\"<\/div>\n

\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u65b9\u6cd5\u306f\u7c21\u5358\u3067\u3001pip install\u3067\u5c0e\u5165\u3067\u304d\u307e\u3059\u3002<\/p>\n

pip install <\/span>tensorflow tensorboard keras jupyter-tensorboard\r\n<\/code><\/pre>\n
Keras\u304b\u3089\u3067\u3082\u4f7f\u3048\u308b<\/h2>\n
Keras\u306f\u5f93\u6765\u304b\u3089Tensorboard\u3067\u53ef\u8996\u5316\u3059\u308b\u3053\u3068\u304c\u3067\u304d\u307e\u3059\u304c\u3001\u3082\u3061\u308d\u3093Jupyter-tensorboard\u3067\u3082\u4f7f\u3048\u307e\u3059\u3002
\n\u7c21\u5358\u306bMNIST\u306e\u30e2\u30c7\u30eb\u3092Jupyter-tensorboard\u3067\u8868\u793a\u3057\u305f\u3044\u3068\u601d\u3044\u307e\u3059\u3002
\n\u30d7\u30ed\u30b0\u30e9\u30e0\u306f\u4ee5\u4e0b\u3067\u3059\u304c\u3001”!!\u3053\u3053\u8ffd\u52a0!!”\u3068\u3042\u308b\u90e8\u5206\u3092\u8ffd\u52a0\u3059\u308b\u3060\u3051\u3067Tensorboard\u3092\u4f7f\u3046\u3053\u3068\u304c\u3067\u304d\u308b\u3088\u3046\u306b\u306a\u308a\u307e\u3059\u3002<\/p>\n
import<\/span> keras<\/span>\r\nfrom<\/span> keras.datasets<\/span> import<\/span> mnist<\/span>\r\nfrom<\/span> keras.models<\/span> import<\/span> Sequential<\/span>\r\nfrom<\/span> keras.layers<\/span> import<\/span> Dense<\/span>,<\/span> Dropout<\/span>\r\nfrom<\/span> keras.optimizers<\/span> import<\/span> RMSprop<\/span>\r\n\r\nbatch_size<\/span> =<\/span> 128<\/span>\r\nnum_classes<\/span> =<\/span> 10<\/span>\r\nepochs<\/span> =<\/span> 20<\/span>\r\n\r\n# the data, shuffled and split between train and test sets\r\n<\/span>(<\/span>x_train<\/span>,<\/span> y_train<\/span>),<\/span> (<\/span>x_test<\/span>,<\/span> y_test<\/span>)<\/span> =<\/span> mnist<\/span>.<\/span>load_data<\/span>()<\/span>\r\n\r\nx_train<\/span> =<\/span> x_train<\/span>.<\/span>reshape<\/span>(<\/span>60000<\/span>,<\/span> 784<\/span>)<\/span>\r\nx_test<\/span> =<\/span> x_test<\/span>.<\/span>reshape<\/span>(<\/span>10000<\/span>,<\/span> 784<\/span>)<\/span>\r\nx_train<\/span> =<\/span> x_train<\/span>.<\/span>astype<\/span>(<\/span>'float32'<\/span>)<\/span>\r\nx_test<\/span> =<\/span> x_test<\/span>.<\/span>astype<\/span>(<\/span>'float32'<\/span>)<\/span>\r\nx_train<\/span> \/=<\/span> 255<\/span>\r\nx_test<\/span> \/=<\/span> 255<\/span>\r\nprint<\/span>(<\/span>x_train<\/span>.<\/span>shape<\/span>[<\/span>0<\/span>],<\/span> 'train samples'<\/span>)<\/span>\r\nprint<\/span>(<\/span>x_test<\/span>.<\/span>shape<\/span>[<\/span>0<\/span>],<\/span> 'test samples'<\/span>)<\/span>\r\n\r\n# convert class vectors to binary class matrices\r\n<\/span>y_train<\/span> =<\/span> keras<\/span>.<\/span>utils<\/span>.<\/span>to_categorical<\/span>(<\/span>y_train<\/span>,<\/span> num_classes<\/span>)<\/span>\r\ny_test<\/span> =<\/span> keras<\/span>.<\/span>utils<\/span>.<\/span>to_categorical<\/span>(<\/span>y_test<\/span>,<\/span> num_classes<\/span>)<\/span>\r\n\r\nmodel<\/span> =<\/span> Sequential<\/span>()<\/span>\r\nmodel<\/span>.<\/span>add<\/span>(<\/span>Dense<\/span>(<\/span>512<\/span>,<\/span> activation<\/span>=<\/span>'relu'<\/span>,<\/span> input_shape<\/span>=<\/span>(<\/span>784<\/span>,)))<\/span>\r\nmodel<\/span>.<\/span>add<\/span>(<\/span>Dropout<\/span>(<\/span>0.2<\/span>))<\/span>\r\nmodel<\/span>.<\/span>add<\/span>(<\/span>Dense<\/span>(<\/span>512<\/span>,<\/span> activation<\/span>=<\/span>'relu'<\/span>))<\/span>\r\nmodel<\/span>.<\/span>add<\/span>(<\/span>Dropout<\/span>(<\/span>0.2<\/span>))<\/span>\r\nmodel<\/span>.<\/span>add<\/span>(<\/span>Dense<\/span>(<\/span>num_classes<\/span>,<\/span> activation<\/span>=<\/span>'softmax'<\/span>))<\/span>\r\n\r\nmodel<\/span>.<\/span>compile<\/span>(<\/span>loss<\/span>=<\/span>'categorical_crossentropy'<\/span>,<\/span>\r\n              optimizer<\/span>=<\/span>RMSprop<\/span>(),<\/span>\r\n              metrics<\/span>=<\/span>[<\/span>'accuracy'<\/span>])<\/span>\r\n\r\n# !!\u3053\u3053\u8ffd\u52a0!!\r\n<\/span>tb_cb<\/span> =<\/span> keras<\/span>.<\/span>callbacks<\/span>.<\/span>TensorBoard<\/span>(<\/span>log_dir<\/span>=<\/span>\"tflog\/\"<\/span>,<\/span> histogram_freq<\/span>=<\/span>1<\/span>)<\/span>\r\ncbks<\/span> =<\/span> [<\/span>tb_cb<\/span>]<\/span>\r\n\r\nhistory<\/span> =<\/span> model<\/span>.<\/span>fit<\/span>(<\/span>x_train<\/span>,<\/span> y_train<\/span>,<\/span>\r\n                    batch_size<\/span>=<\/span>batch_size<\/span>,<\/span>\r\n                    epochs<\/span>=<\/span>epochs<\/span>,<\/span>\r\n                    verbose<\/span>=<\/span>1<\/span>,<\/span>\r\n                    validation_data<\/span>=<\/span>(<\/span>x_test<\/span>,<\/span> y_test<\/span>),<\/span>\r\n                    callbacks<\/span>=<\/span>cbks<\/span>)<\/span>\u3000<\/span># !!\u3053\u3053\u8ffd\u52a0!!\r\n<\/span>\r\nscore<\/span> =<\/span> model<\/span>.<\/span>evaluate<\/span>(<\/span>x_test<\/span>,<\/span> y_test<\/span>,<\/span> verbose<\/span>=<\/span>0<\/span>)<\/span>\r\nprint<\/span>(<\/span>'Test loss:'<\/span>,<\/span> score<\/span>[<\/span>0<\/span>])<\/span>\r\nprint<\/span>(<\/span>'Test accuracy:'<\/span>,<\/span> score<\/span>[<\/span>1<\/span>])<\/span>\r\n\r\n<\/code><\/pre>\n
MNIST\u306e\u5b66\u7fd2\u3092\u958b\u59cb\u3059\u308b\u3068\u3001”tflog\/”\u30c7\u30a3\u30ec\u30af\u30c8\u30ea\u306b\u30ed\u30b0\u304c\u84c4\u7a4d\u3055\u308c\u3066\u3044\u304d\u307e\u3059\u3002
\nJupyter\u306e\u753b\u9762\u3067\u3053\u306e\u30c7\u30a3\u30ec\u30af\u30c8\u30ea\u3092\u9078\u629e\u3057\u3001Tensorboard\u3092\u9078\u629e\u3059\u308b\u3053\u3068\u3067Jupyter-tensorboard\u3092\u8d77\u52d5\u3059\u308b\u3053\u3068\u304c\u3067\u304d\u307e\u3059\u3002<\/p>\n
\"102.PNG\"<\/div>\n
\u3042\u3068\u306fTensorboard\u305d\u306e\u307e\u307e\u306b\u4f7f\u3046\u3053\u3068\u304c\u3067\u304d\u307e\u3059\u3002<\/p>\n
\"103.PNG\"<\/div>\n","protected":false},"excerpt":{"rendered":"
Jupyter-tensorboard\u306e\u7d39\u4ecb\u3068Keras\u304b\u3089\u4f7f\u3063\u3066\u307f\u308b Jupyter Notebook\u306b\u7d71\u5408 […]<\/p>\n","protected":false},"author":10,"featured_media":0,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[],"class_list":["post-46395","post","type-post","status-publish","format-standard","hentry","category-uncategorized"],"yoast_head":"\n- Blog - Silicon Cloud<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/www.silicloud.com\/zh\/blog\/46395-2\/\" \/>\n<meta property=\"og:locale\" content=\"zh_CN\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:description\" content=\"Jupyter-tensorboard\u306e\u7d39\u4ecb\u3068Keras\u304b\u3089\u4f7f\u3063\u3066\u307f\u308b Jupyter Notebook\u306b\u7d71\u5408 […]\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.silicloud.com\/zh\/blog\/46395-2\/\" \/>\n<meta property=\"og:site_name\" content=\"Blog - Silicon Cloud\" \/>\n<meta property=\"article:published_time\" content=\"2023-08-12T15:57:07+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2024-04-29T03:12:10+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/cdn.silicloud.com\/blog-img\/blog\/img\/657d649c37434c4406d04e2f\/3-0.png\" \/>\n<meta name=\"author\" content=\"\u5b87, \u534e\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"\u4f5c\u8005\" \/>\n\t<meta name=\"twitter:data1\" content=\"\u5b87, \u534e\" \/>\n\t<meta name=\"twitter:label2\" content=\"\u9884\u8ba1\u9605\u8bfb\u65f6\u95f4\" \/>\n\t<meta name=\"twitter:data2\" content=\"1 \u5206\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\/\/www.silicloud.com\/zh\/blog\/46395-2\/\",\"url\":\"https:\/\/www.silicloud.com\/zh\/blog\/46395-2\/\",\"name\":\"- Blog - Silicon Cloud\",\"isPartOf\":{\"@id\":\"https:\/\/www.silicloud.com\/zh\/blog\/#website\"},\"datePublished\":\"2023-08-12T15:57:07+00:00\",\"dateModified\":\"2024-04-29T03:12:10+00:00\",\"author\":{\"@id\":\"https:\/\/www.silicloud.com\/zh\/blog\/#\/schema\/person\/513018e4e121d3add1b7c5de8be21458\"},\"inLanguage\":\"zh-Hans\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/www.silicloud.com\/zh\/blog\/46395-2\/\"]}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/www.silicloud.com\/zh\/blog\/#website\",\"url\":\"https:\/\/www.silicloud.com\/zh\/blog\/\",\"name\":\"Blog - Silicon Cloud\",\"description\":\"\",\"inLanguage\":\"zh-Hans\"},{\"@type\":\"Person\",\"@id\":\"https:\/\/www.silicloud.com\/zh\/blog\/#\/schema\/person\/513018e4e121d3add1b7c5de8be21458\",\"name\":\"\u5b87, \u534e\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"zh-Hans\",\"@id\":\"https:\/\/www.silicloud.com\/zh\/blog\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/secure.gravatar.com\/avatar\/63cd45cbc05a35fc4ff7637a163c83c4962ef58d27472726c3a3e0c9c5194f0f?s=96&d=mm&r=g\",\"contentUrl\":\"https:\/\/secure.gravatar.com\/avatar\/63cd45cbc05a35fc4ff7637a163c83c4962ef58d27472726c3a3e0c9c5194f0f?s=96&d=mm&r=g\",\"caption\":\"\u5b87, \u534e\"},\"url\":\"https:\/\/www.silicloud.com\/zh\/blog\/author\/yuhua\/\"},{\"@type\":\"ImageObject\",\"inLanguage\":\"zh-Hans\",\"@id\":\"https:\/\/www.silicloud.com\/zh\/blog\/46395-2\/#local-main-organization-logo\",\"url\":\"\",\"contentUrl\":\"\",\"caption\":\"Blog - Silicon Cloud\"}]}<\/script>\n<!-- \/ Yoast SEO Premium plugin. -->","yoast_head_json":{"title":"- Blog - Silicon Cloud","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/www.silicloud.com\/zh\/blog\/46395-2\/","og_locale":"zh_CN","og_type":"article","og_description":"Jupyter-tensorboard\u306e\u7d39\u4ecb\u3068Keras\u304b\u3089\u4f7f\u3063\u3066\u307f\u308b Jupyter Notebook\u306b\u7d71\u5408 […]","og_url":"https:\/\/www.silicloud.com\/zh\/blog\/46395-2\/","og_site_name":"Blog - Silicon Cloud","article_published_time":"2023-08-12T15:57:07+00:00","article_modified_time":"2024-04-29T03:12:10+00:00","og_image":[{"url":"https:\/\/cdn.silicloud.com\/blog-img\/blog\/img\/657d649c37434c4406d04e2f\/3-0.png"}],"author":"\u5b87, \u534e","twitter_card":"summary_large_image","twitter_misc":{"\u4f5c\u8005":"\u5b87, \u534e","\u9884\u8ba1\u9605\u8bfb\u65f6\u95f4":"1 \u5206"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/www.silicloud.com\/zh\/blog\/46395-2\/","url":"https:\/\/www.silicloud.com\/zh\/blog\/46395-2\/","name":"- Blog - Silicon Cloud","isPartOf":{"@id":"https:\/\/www.silicloud.com\/zh\/blog\/#website"},"datePublished":"2023-08-12T15:57:07+00:00","dateModified":"2024-04-29T03:12:10+00:00","author":{"@id":"https:\/\/www.silicloud.com\/zh\/blog\/#\/schema\/person\/513018e4e121d3add1b7c5de8be21458"},"inLanguage":"zh-Hans","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.silicloud.com\/zh\/blog\/46395-2\/"]}]},{"@type":"WebSite","@id":"https:\/\/www.silicloud.com\/zh\/blog\/#website","url":"https:\/\/www.silicloud.com\/zh\/blog\/","name":"Blog - Silicon Cloud","description":"","inLanguage":"zh-Hans"},{"@type":"Person","@id":"https:\/\/www.silicloud.com\/zh\/blog\/#\/schema\/person\/513018e4e121d3add1b7c5de8be21458","name":"\u5b87, \u534e","image":{"@type":"ImageObject","inLanguage":"zh-Hans","@id":"https:\/\/www.silicloud.com\/zh\/blog\/#\/schema\/person\/image\/","url":"https:\/\/secure.gravatar.com\/avatar\/63cd45cbc05a35fc4ff7637a163c83c4962ef58d27472726c3a3e0c9c5194f0f?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/63cd45cbc05a35fc4ff7637a163c83c4962ef58d27472726c3a3e0c9c5194f0f?s=96&d=mm&r=g","caption":"\u5b87, \u534e"},"url":"https:\/\/www.silicloud.com\/zh\/blog\/author\/yuhua\/"},{"@type":"ImageObject","inLanguage":"zh-Hans","@id":"https:\/\/www.silicloud.com\/zh\/blog\/46395-2\/#local-main-organization-logo","url":"","contentUrl":"","caption":"Blog - Silicon Cloud"}]}},"_links":{"self":[{"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/posts\/46395","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/users\/10"}],"replies":[{"embeddable":true,"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/comments?post=46395"}],"version-history":[{"count":2,"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/posts\/46395\/revisions"}],"predecessor-version":[{"id":84928,"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/posts\/46395\/revisions\/84928"}],"wp:attachment":[{"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/media?parent=46395"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/categories?post=46395"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/tags?post=46395"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}