{"id":45791,"date":"2023-10-22T03:04:37","date_gmt":"2023-02-09T16:58:06","guid":{"rendered":"https:\/\/www.silicloud.com\/zh\/blog\/45791-2\/"},"modified":"2024-05-04T05:26:34","modified_gmt":"2024-05-03T21:26:34","slug":"45791-2","status":"publish","type":"post","link":"https:\/\/www.silicloud.com\/zh\/blog\/45791-2\/","title":{"rendered":""},"content":{"rendered":"<p>Jupyter Notebook\u3067\u3044\u3058\u3063\u3066\u5b66\u3076TensorFlow &#8211; MNIST For ML Beginners<br \/>\n\u3092python3 \u3067\u3084\u3063\u3066\u307f\u305f\u306e\u3067\u3081\u3082<\/p>\n<h1>\u74b0\u5883<\/h1>\n<p>linux mate 19<br \/>\nPython 3.6.7<br \/>\nconda 4.5.12<\/p>\n<h2>\u5b9f\u884c<\/h2>\n<ol>\u5fc5\u8981\u306a\u30e9\u30a4\u30d6\u30e9\u30ea\u3092\u8aad\u307f\u8fbc\u3080<\/ol>\n<p>\u5fc5\u8981\u306a\u30e9\u30a4\u30d6\u30e9\u30ea\u3092import<\/p>\n<pre class=\"post-pre\"><code>%matplotlib inline\r\n\r\nimport matplotlib.pyplot as plt\r\nimport tensorflow as tf\r\nimport numpy as np\r\n<\/code><\/pre>\n<ol>MNIST\u753b\u50cf\u3092\u8aad\u307f\u8fbc\u3080<\/ol>\n<p>MNIST\u306f\u3001\u6a5f\u68b0\u5b66\u7fd2\u306a\u3069\u306b\u3088\u304f\u4f7f\u308f\u308c\u308b\u624b\u66f8\u304d\u6587\u5b57\u306e\u753b\u50cf\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u3002\u6700\u521d\u306f\u30c0\u30a6\u30f3\u30ed\u30fc\u30c9\u306b\u5c11\u3057\u6642\u9593\u304c\u304b\u304b\u308b\u3002<\/p>\n<pre class=\"post-pre\"><code>old_v = tf.logging.get_verbosity()\r\ntf.logging.set_verbosity(tf.logging.ERROR)\r\n\r\nfrom tensorflow.contrib.learn.python.learn.datasets import mnist as mnist_loader\r\n\r\nmnist = mnist_loader.read_data_sets(\"MNIST_data\/\", one_hot=True)\r\n<\/code><\/pre>\n<p>Extracting MNIST_data\/train-images-idx3-ubyte.gz<br \/>\nExtracting MNIST_data\/train-labels-idx1-ubyte.gz<br \/>\nExtracting MNIST_data\/t10k-images-idx3-ubyte.gz<br \/>\nExtracting MNIST_data\/t10k-labels-idx1-ubyte.gz<\/p>\n<p>\u624b\u66f8\u304d\u753b\u50cf\u3092\u3072\u3068\u3064\u898b\u3066\u3002<\/p>\n<pre class=\"post-pre\"><code>plt.imshow(mnist.train.images[8].reshape([28, 28]))\r\nplt.gray()\r\n<\/code><\/pre>\n<div><img decoding=\"async\" class=\"post-images\" title=\"\" src=\"https:\/\/cdn.silicloud.com\/blog-img\/blog\/img\/657d615937434c4406cfd59d\/13-0.png\" alt=\"image.png\" \/><\/div>\n<p>\u3061\u306a\u307f\u306b\u3001\u305f\u307e\u305f\u307e\u898b\u3064\u3051\u305f\u3084\u3070\u3044\u3084\u3064<\/p>\n<pre class=\"post-pre\"><code>plt.imshow(mnist.train.images[65].reshape([28, 28]))\r\nplt.gray()\r\n<\/code><\/pre>\n<div><img decoding=\"async\" class=\"post-images\" title=\"\" src=\"https:\/\/cdn.silicloud.com\/blog-img\/blog\/img\/657d615937434c4406cfd59d\/16-0.png\" alt=\"image.png\" \/><\/div>\n<p>\u3053\u308c\u306f\u3001\u3001\u3001\uff18\uff1f<\/p>\n<p>\u30e9\u30d9\u30eb\u3092\u3072\u3068\u3064\u898b\u3066\u307f\u308b\u3002<\/p>\n<pre class=\"post-pre\"><code>plt.imshow(mnist.train.labels[8].reshape([1, -1]))\r\nplt.gray()\r\n<\/code><\/pre>\n<div><img decoding=\"async\" class=\"post-images\" title=\"\" src=\"https:\/\/cdn.silicloud.com\/blog-img\/blog\/img\/657d615937434c4406cfd59d\/20-0.png\" alt=\"image.png\" \/><\/div>\n<pre class=\"post-pre\"><code>plt.imshow(mnist.train.labels[65].reshape([1, -1]))\r\nplt.gray()\r\n<\/code><\/pre>\n<div><img decoding=\"async\" class=\"post-images\" title=\"\" src=\"https:\/\/cdn.silicloud.com\/blog-img\/blog\/img\/657d615937434c4406cfd59d\/22-0.png\" alt=\"image.png\" \/><\/div>\n<p>\u3069\u3046\u3084\u3089\u3055\u3063\u304d\u306e\u3088\u304f\u308f\u304b\u3089\u3093\u306e\u306f\u300c\uff12\u300d\u3089\u3057\u3044\u30fb\u30fb\u30fb<\/p>\n<p>\uff13. \u30e2\u30c7\u30eb\u3092\u5b9a\u7fa9\u3057\u3088\u3046<\/p>\n<p>Softmax\u56de\u5e30\u306e\u5b9f\u88c5<\/p>\n<p>\u5165\u529bx<\/p>\n<pre class=\"post-pre\"><code>x = tf.placeholder(\"float\",[None,784])\r\n<\/code><\/pre>\n<p>weight\u3068biases<\/p>\n<pre class=\"post-pre\"><code>W = tf.Variable(tf.zeros([784,10]))\r\nb = tf.Variable(tf.zeros([10]))\r\n<\/code><\/pre>\n<p>\u7c21\u5358\u306b\u8a00\u3046\u3068\u3001<\/p>\n<pre class=\"post-pre\"><code>y = W_{0}x_{0}+W_{1}x_{1}+...+W_{9}x_{9}+b\r\n<\/code><\/pre>\n<p>\u30e2\u30c7\u30eb\u306e\u5b9a\u7fa9<\/p>\n<pre class=\"post-pre\"><code>y = tf.nn.softmax(tf.matmul(x,W)+b)\r\n<\/code><\/pre>\n<p>\uff14. \u30e2\u30c7\u30eb\u306e\u8a13\u7df4\u65b9\u6cd5\u3092\u5b9a\u7fa9<\/p>\n<p>\u8a13\u7df4\u65b9\u6cd5\u306e\u5b9f\u88c5<\/p>\n<p>placeholder\u306e\u5b9a\u7fa9\u3002 \u8a08\u7b97\u3067\u6c42\u3081\u308b\uff59\u3092\u4ee5\u4e0b\u306e\u3088\u3046\u306a\u5f0f\u3067\u8868\u3055\u308c\u308b\u3068\u3059\u308b\u3002<\/p>\n<pre class=\"post-pre\"><code>y_ = tf.placeholder(\"float\",[None,10])\r\n<\/code><\/pre>\n<p>\u3068\u3001\u4ea4\u5dee\u30a8\u30f3\u30c8\u30ed\u30d4\u30fc\uff08\u8aa4\u5dee\uff09\u306f<\/p>\n<pre class=\"post-pre\"><code>cross_entropy = -tf.reduce_sum(y_*tf.log(y))\r\n<\/code><\/pre>\n<p>\u3067\u8868\u3055\u308c\u308b<\/p>\n<p>\u8a13\u7df4\u65b9\u6cd5\u306e\u5b9a\u7fa9<\/p>\n<pre class=\"post-pre\"><code>learning_rate = 0.01\r\ntrain_step = tf.train.GradientDescentOptimizer(learning_rate).minimize(cross_entropy)\r\n<\/code><\/pre>\n<p>learning_rate\u3000\u306e\u5909\u52d5\u3067 cross_entropy \u304c\u6700\u5c0f\u306b\u306a\u308b\u3088\u3046\u306b\u6025\u901f\u964d\u4e0b\u6cd5\u3067\u8a13\u7df4\u3059\u308b\u3002<\/p>\n<p>\uff15. \u30e2\u30c7\u30eb\u3092\u5b9f\u969b\u306b\u8a13\u7df4<\/p>\n<pre class=\"post-pre\"><code>init = tf.initialize_all_variables()\r\n<\/code><\/pre>\n<pre class=\"post-pre\"><code>sess = tf.InteractiveSession()\r\nsess.run(init)\r\n<\/code><\/pre>\n<p>\u8a13\u7df4\u306e\u5b9f\u884c<\/p>\n<pre class=\"post-pre\"><code># \u8a13\u7df4\u30d1\u30e9\u30e1\u30fc\u30bf\r\nn_train = 1000\r\nn_batch = 100\r\n\r\n# \u30b0\u30e9\u30d5\u63cf\u753b\u7528\r\nfig, ax = plt.subplots(1, 1, figsize=(15, 5))\r\nxvalues = np.arange(n_train)\r\nyvalues = np.zeros(n_train)\r\nlines, = ax.plot(xvalues, yvalues, label=\"cross_entropy\")\r\n\r\nfor i in range(n_train):\r\n\r\n    # \u30d0\u30c3\u30c1\u5b66\u7fd2\r\n    batch_xs, batch_ys = mnist.train.next_batch(n_batch)\r\n    sess.run(train_step, feed_dict={x: batch_xs, y_: batch_ys})\r\n\r\n    # \u30b0\u30e9\u30d5\u63cf\u753b\u7528\r\n    yvalues[i] = cross_entropy.eval(feed_dict={x: mnist.test.images[0:100], y_: mnist.test.labels[0:100]})\r\n    lines.set_data(xvalues, yvalues)\r\n    ax.set_ylim((yvalues.min(), yvalues.max()))\r\n    plt.legend()\r\n\r\n# \u30b0\u30e9\u30d5\u306e\u63cf\u5199\r\nplt.pause(.00001)\r\n<\/code><\/pre>\n<div><img decoding=\"async\" class=\"post-images\" title=\"\" src=\"https:\/\/cdn.silicloud.com\/blog-img\/blog\/img\/657d615937434c4406cfd59d\/49-0.png\" alt=\"image.png\" \/><\/div>\n<p>\uff16. \u4f5c\u6210\u3057\u305f\u30e2\u30c7\u30eb\u3092\u8a55\u4fa1<\/p>\n<pre class=\"post-pre\"><code>correct_prediction = tf.equal(tf.argmax(y,1),tf.argmax(y_,1))\r\n<\/code><\/pre>\n<p>bool\u5024\u306e\u30c6\u30f3\u30bd\u30eb\u3002\u3002\u3002\u3064\u307e\u308a\u3001\u6b63\u89e3\u3092\u300c\uff11\u300d\u3001\u8aa4\u7b54\u3092\u300c\uff10\u300d\u3068\u3057\u3066<\/p>\n<pre class=\"post-pre\"><code>{{1,1,1,1,0,0,1,1,1,1,10,1,....,1},\r\n{1,0,1,1,1,0,1,1,1,1,10,1,....,1},\r\n{1,1,1,1,1,0,1,1,1,1,10,1,....,1},\r\n{1,1,1,1,0,0,1,1,1,1,10,1,....,1},\r\n{1,1,1,1,0,0,1,1,1,1,10,1,....,1},\r\n{1,1,1,1,0,0,1,1,1,1,10,1,....,1},\r\n{1,1,1,1,0,0,1,1,1,1,10,1,....,1},\r\n{1,1,1,1,0,0,1,1,1,1,10,1,....,1}.....\r\n{1,1,1,1,0,0,1,1,1,1,10,1,....,1}}\r\n<\/code><\/pre>\n<p>\u307f\u305f\u3044\u306a\u72b6\u614b\u3002\u3053\u308c\u3092float\u306b\u30ad\u30e3\u30b9\u30c8\u3057\u3066\u5e73\u5747\u3092\u3068\u308b<\/p>\n<pre class=\"post-pre\"><code>accuracy = tf.reduce_mean(tf.cast(correct_prediction, \"float\"))\r\n\r\nprint(sess.run(accuracy, feed_dict={x: mnist.test.images, y_: mnist.test.labels}))\r\n<\/code><\/pre>\n<p>0.9211<\/p>\n<pre class=\"post-pre\"><code>w = W.eval().T\r\nfig = plt.figure(figsize=(10, 4))\r\n\r\nfor i in range(10):\r\n    ax = fig.add_subplot(2, 5, i + 1)\r\n    ax.imshow(w[i].reshape([28, 28]), cmap=\"seismic\")\r\n<\/code><\/pre>\n<div><img decoding=\"async\" class=\"post-images\" title=\"\" src=\"https:\/\/cdn.silicloud.com\/blog-img\/blog\/img\/657d615937434c4406cfd59d\/58-0.png\" alt=\"image.png\" \/><\/div>\n<ol>\u4f5c\u6210\u3057\u305f\u30e2\u30c7\u30eb\u3092\u4f7f\u3063\u3066\u5206\u985e<\/ol>\n<pre class=\"post-pre\"><code># \u624b\u66f8\u304d\u753b\u50cf\u306e\u30a4\u30f3\u30c7\u30c3\u30af\u30b9\r\nindex_test_image = 5\r\n\r\n# \u5206\u985e\r\nresult = y.eval(feed_dict={x: [mnist.test.images[index_test_image]]})\r\nprint(result)\r\n\r\n# \u63cf\u753b\u306e\u6e96\u5099\r\nfig = plt.figure(figsize=(8, 6))\r\n\r\n# \u30c6\u30b9\u30c8\u753b\u50cf\u3092\u63cf\u753b\r\nax0 = fig.add_subplot(2, 1, 1)\r\nax0.imshow(mnist.test.images[index_test_image].reshape([28, 28]))\r\n\r\n# \u5206\u985e\u7d50\u679c\u3092\u63cf\u753b\r\nax1 = fig.add_subplot(2, 1, 2)\r\nax1.imshow(result)\r\n<\/code><\/pre>\n<p>[[8.7355545e-08 9.8954684e-01 2.1719618e-03 2.6251078e-03 1.2656899e-05<br \/>\n3.2387452e-05 2.0649401e-05 3.0584279e-03 2.3757515e-03 1.5612924e-04]]<\/p>\n<div><img decoding=\"async\" class=\"post-images\" title=\"\" src=\"https:\/\/cdn.silicloud.com\/blog-img\/blog\/img\/657d615937434c4406cfd59d\/62-0.png\" alt=\"image.png\" \/><\/div>\n<p>98.954%\u306e\u78ba\u7387\u3067\u300c\uff11\u300d<\/p>\n<p>\u3053\u308c\u3068\u304b\u3061\u3087\u3063\u3068\u304a\u3082\u3057\u308d\u3044<\/p>\n<pre class=\"post-pre\"><code># \u624b\u66f8\u304d\u753b\u50cf\u306e\u30a4\u30f3\u30c7\u30c3\u30af\u30b9\r\nindex_test_image = 77\r\n\r\n# \u5206\u985e\r\nresult = y.eval(feed_dict={x: [mnist.test.images[index_test_image]]})\r\nprint(result)\r\n\r\n# \u63cf\u753b\u306e\u6e96\u5099\r\nfig = plt.figure(figsize=(8, 6))\r\n\r\n# \u30c6\u30b9\u30c8\u753b\u50cf\u3092\u63cf\u753b\r\nax0 = fig.add_subplot(2, 1, 1)\r\nax0.imshow(mnist.test.images[index_test_image].reshape([28, 28]))\r\n\r\n# \u5206\u985e\u7d50\u679c\u3092\u63cf\u753b\r\nax1 = fig.add_subplot(2, 1, 2)\r\nax1.imshow(result)\r\n<\/code><\/pre>\n<p>[[1.1343922e-03 2.7609523e-03 6.7110407e-01 4.1867211e-03 4.3900713e-04<br \/>\n9.8160030e-03 2.5389763e-03 2.6510671e-01 6.9887578e-03 3.5924386e-02]]<\/p>\n<div><img decoding=\"async\" class=\"post-images\" title=\"\" src=\"https:\/\/cdn.silicloud.com\/blog-img\/blog\/img\/657d615937434c4406cfd59d\/68-0.png\" alt=\"image.png\" \/><\/div>\n<p>\u300c\uff12\u300d\u306e\u78ba\u7387\u304c67,11%\u3000\u300c\uff17\u300d\u306e\u78ba\u7387\u304c26.51\uff05<\/p>\n<p>\u3069\u3046\u3084\u3089\u3001\u300c\uff12\u300d\u306e\u3057\u305f\u306e\u6a2a\u68d2\u3092\u9664\u3051\u3070\u300c\uff17\u300d\u306b\u898b\u3048\u308b\u3089\u3057\u3044<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Jupyter Notebook\u3067\u3044\u3058\u3063\u3066\u5b66\u3076TensorFlow &#8211; MNIST For ML  [&hellip;]<\/p>\n","protected":false},"author":5,"featured_media":0,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[],"class_list":["post-45791","post","type-post","status-publish","format-standard","hentry","category-uncategorized"],"yoast_head":"<!-- This site is optimized with the Yoast SEO Premium plugin v21.5 (Yoast SEO v21.5) - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>- 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\/45791-2\/\" \/>\n<meta property=\"og:locale\" content=\"zh_CN\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:description\" content=\"Jupyter Notebook\u3067\u3044\u3058\u3063\u3066\u5b66\u3076TensorFlow &#8211; MNIST For ML [&hellip;]\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.silicloud.com\/zh\/blog\/45791-2\/\" \/>\n<meta property=\"og:site_name\" content=\"Blog - Silicon Cloud\" \/>\n<meta property=\"article:published_time\" content=\"2023-02-09T16:58:06+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2024-05-03T21:26:34+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/cdn.silicloud.com\/blog-img\/blog\/img\/657d615937434c4406cfd59d\/13-0.png\" \/>\n<meta name=\"author\" content=\"\u6e05, \u5b87\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"\u4f5c\u8005\" \/>\n\t<meta name=\"twitter:data1\" content=\"\u6e05, \u5b87\" \/>\n\t<meta name=\"twitter:label2\" content=\"\u9884\u8ba1\u9605\u8bfb\u65f6\u95f4\" \/>\n\t<meta name=\"twitter:data2\" content=\"2 \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\/45791-2\/\",\"url\":\"https:\/\/www.silicloud.com\/zh\/blog\/45791-2\/\",\"name\":\"- Blog - Silicon Cloud\",\"isPartOf\":{\"@id\":\"https:\/\/www.silicloud.com\/zh\/blog\/#website\"},\"datePublished\":\"2023-02-09T16:58:06+00:00\",\"dateModified\":\"2024-05-03T21:26:34+00:00\",\"author\":{\"@id\":\"https:\/\/www.silicloud.com\/zh\/blog\/#\/schema\/person\/1a6ecd3d914d22a5ac32791ffc1fbd8e\"},\"inLanguage\":\"zh-Hans\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/www.silicloud.com\/zh\/blog\/45791-2\/\"]}]},{\"@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\/1a6ecd3d914d22a5ac32791ffc1fbd8e\",\"name\":\"\u6e05, \u5b87\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"zh-Hans\",\"@id\":\"https:\/\/www.silicloud.com\/zh\/blog\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/secure.gravatar.com\/avatar\/4b2016c18459a605fc469c7566608f5686491baa112d0871ee613f61b7210565?s=96&d=mm&r=g\",\"contentUrl\":\"https:\/\/secure.gravatar.com\/avatar\/4b2016c18459a605fc469c7566608f5686491baa112d0871ee613f61b7210565?s=96&d=mm&r=g\",\"caption\":\"\u6e05, \u5b87\"},\"url\":\"https:\/\/www.silicloud.com\/zh\/blog\/author\/qingyu\/\"},{\"@type\":\"ImageObject\",\"inLanguage\":\"zh-Hans\",\"@id\":\"https:\/\/www.silicloud.com\/zh\/blog\/45791-2\/#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\/45791-2\/","og_locale":"zh_CN","og_type":"article","og_description":"Jupyter Notebook\u3067\u3044\u3058\u3063\u3066\u5b66\u3076TensorFlow &#8211; MNIST For ML [&hellip;]","og_url":"https:\/\/www.silicloud.com\/zh\/blog\/45791-2\/","og_site_name":"Blog - Silicon Cloud","article_published_time":"2023-02-09T16:58:06+00:00","article_modified_time":"2024-05-03T21:26:34+00:00","og_image":[{"url":"https:\/\/cdn.silicloud.com\/blog-img\/blog\/img\/657d615937434c4406cfd59d\/13-0.png"}],"author":"\u6e05, \u5b87","twitter_card":"summary_large_image","twitter_misc":{"\u4f5c\u8005":"\u6e05, \u5b87","\u9884\u8ba1\u9605\u8bfb\u65f6\u95f4":"2 \u5206"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/www.silicloud.com\/zh\/blog\/45791-2\/","url":"https:\/\/www.silicloud.com\/zh\/blog\/45791-2\/","name":"- Blog - Silicon Cloud","isPartOf":{"@id":"https:\/\/www.silicloud.com\/zh\/blog\/#website"},"datePublished":"2023-02-09T16:58:06+00:00","dateModified":"2024-05-03T21:26:34+00:00","author":{"@id":"https:\/\/www.silicloud.com\/zh\/blog\/#\/schema\/person\/1a6ecd3d914d22a5ac32791ffc1fbd8e"},"inLanguage":"zh-Hans","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.silicloud.com\/zh\/blog\/45791-2\/"]}]},{"@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\/1a6ecd3d914d22a5ac32791ffc1fbd8e","name":"\u6e05, \u5b87","image":{"@type":"ImageObject","inLanguage":"zh-Hans","@id":"https:\/\/www.silicloud.com\/zh\/blog\/#\/schema\/person\/image\/","url":"https:\/\/secure.gravatar.com\/avatar\/4b2016c18459a605fc469c7566608f5686491baa112d0871ee613f61b7210565?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/4b2016c18459a605fc469c7566608f5686491baa112d0871ee613f61b7210565?s=96&d=mm&r=g","caption":"\u6e05, \u5b87"},"url":"https:\/\/www.silicloud.com\/zh\/blog\/author\/qingyu\/"},{"@type":"ImageObject","inLanguage":"zh-Hans","@id":"https:\/\/www.silicloud.com\/zh\/blog\/45791-2\/#local-main-organization-logo","url":"","contentUrl":"","caption":"Blog - Silicon Cloud"}]}},"_links":{"self":[{"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/posts\/45791","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\/5"}],"replies":[{"embeddable":true,"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/comments?post=45791"}],"version-history":[{"count":2,"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/posts\/45791\/revisions"}],"predecessor-version":[{"id":96386,"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/posts\/45791\/revisions\/96386"}],"wp:attachment":[{"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/media?parent=45791"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/categories?post=45791"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/tags?post=45791"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}