{"id":426,"date":"2018-08-16T23:00:26","date_gmt":"2018-08-16T15:00:26","guid":{"rendered":"https:\/\/mmaqa.com\/blog\/?p=426"},"modified":"2021-12-13T23:22:23","modified_gmt":"2021-12-13T15:22:23","slug":"%e3%80%90%e8%b5%84%e6%ba%90%e3%80%91wolfram-neural-net-repository%ef%bc%88%e7%a5%9e%e7%bb%8f%e7%bd%91%e7%bb%9c%e6%a8%a1%e5%9e%8b%e5%ba%93%ef%bc%89","status":"publish","type":"post","link":"https:\/\/mmaqa.com\/blog\/426\/2018\/08\/16\/%e3%80%90%e8%b5%84%e6%ba%90%e3%80%91wolfram-neural-net-repository%ef%bc%88%e7%a5%9e%e7%bb%8f%e7%bd%91%e7%bb%9c%e6%a8%a1%e5%9e%8b%e5%ba%93%ef%bc%89\/","title":{"rendered":"\u3010\u8d44\u6e90\u3011Wolfram Neural Net Repository\uff08\u795e\u7ecf\u7f51\u7edc\u6a21\u578b\u5e93\uff09"},"content":{"rendered":"<div id=\"toc_container\" class=\"no_bullets\"><p class=\"toc_title\">\u6587\u7ae0\u76ee\u5f55<\/p><ul class=\"toc_list\"><li><a href=\"#i\">\u8d44\u6e90\u4e0b\u8f7d<\/a><\/li><li><a href=\"#i-2\">0. \u8bf4\u660e<\/a><\/li><li><a href=\"#1\">1. \u8d44\u6e90\u4e0b\u8f7d\u4e0e\u4fdd\u5b58<\/a><ul><li><a href=\"#11\">1.1 \u6765\u6e90<\/a><\/li><li><a href=\"#12_Wolfram\">1.2 \u767b\u5f55 Wolfram \u8d26\u53f7<\/a><\/li><li><a href=\"#13\">1.3 \u4e0b\u8f7d\u5e76\u4fdd\u5b58\u8d44\u6e90<\/a><\/li><\/ul><\/li><li><a href=\"#2\">2. \u4f7f\u7528<\/a><ul><li><a href=\"#21\">2.1 \u76f4\u63a5\u4f7f\u7528\u65b9\u6cd5<\/a><\/li><li><a href=\"#22\">2.2 \u4fee\u6539\u6a21\u578b\u4e0e\u4f7f\u7528\u65b9\u6cd5<\/a><\/li><\/ul><\/li><\/ul><\/div>\n<h1><span id=\"i\">\u8d44\u6e90\u4e0b\u8f7d<\/span><\/h1>\n<ul>\n<li><a href=\"https:\/\/pan.baidu.com\/s\/1atPNFYQerpGRrMrXXNdoAQ\">\u30102020-03-22\u3011\u3010NetModel\u3011\u3010\u767e\u5ea6\u7f51\u76d8\u3011\u3010 95\u4e2a\u6a21\u578b\u3011\u3010\u4ee5\u6a21\u578b\u540d\u79f0\u4f5c\u4e3a\u6587\u4ef6\u540d\u5b58\u50a8\u3011<\/a>\uff08\u63d0\u53d6\u5bc6\u7801\uff1a46zp\uff09<\/li>\n<li><a href=\"https:\/\/pan.baidu.com\/s\/1q_G5GsJ1d8vBkgSdpeSe-w\">\u30102018-11-19\u3011\u3010NetModel\u3011\u3010\u767e\u5ea6\u7f51\u76d8\u3011\u3010 79\u4e2a\u6a21\u578b\u3011\u3010\u4ee5\u6a21\u578b\u540d\u79f0\u4f5c\u4e3a\u6587\u4ef6\u540d\u5b58\u50a8\u3011<\/a>\uff08\u63d0\u53d6\u5bc6\u7801\uff1addrk\uff09<\/li>\n<li><a href=\"https:\/\/pan.baidu.com\/s\/1r-4EDFSJVulmd8wIAEh34w\">\u30102018-08-16\u3011<\/a><a href=\"https:\/\/pan.baidu.com\/s\/1r-4EDFSJVulmd8wIAEh34w\">\u3010NetModel\u3011\u3010\u767e\u5ea6\u7f51\u76d8\u3011\u3010 7\uff16\u4e2a\u6a21\u578b\u3011\u3010\u4ee5UUID\u4f5c\u4e3a\u6587\u4ef6\u540d\u5b58\u50a8\u3011<\/a><\/li>\n<\/ul>\n<h1><span id=\"i-2\">0. \u8bf4\u660e<\/span><\/h1>\n<p>\u6700\u8fd1\u5728\u5b66\u4e60 Mathematica \u673a\u5668\u5b66\u4e60\u65b9\u9762\u7684\u529f\u80fd\uff0c\u4e0e TensorFlow \u76f8\u6bd4\uff0cMathematica \u5728\u673a\u5668\u5b66\u4e60\u65b9\u9762\u4e3b\u8981\u6709\u4ee5\u4e0b\u4f18\u70b9\uff1a<\/p>\n<ul>\n<li>\u6784\u5efa\u6a21\u578b\u7b80\u6d01\u660e\u4e86<\/li>\n<li>\u8bad\u7ec3\u8fc7\u7a0b\u53ef\u89c6\u5316<\/li>\n<li>\u4ea4\u4e92\u5f0f\u7684\u5c55\u793a\u6a21\u578b\u7ed3\u6784\uff0c\u5bf9\u4e8e\u5927\u578b\u7f51\u7edc\u5f88\u65b9\u4fbf<\/li>\n<li>\u52a0\u8f7d\u6570\u636e\u65b9\u4fbf<\/li>\n<\/ul>\n<p><center><br \/>\n<a href=\"https:\/\/mmaqa.com\/blog\/wp-content\/uploads\/2018\/08\/TIM%E5%9B%BE%E7%89%8720180816230145.png\"><img src=\"https:\/\/mmaqa.com\/blog\/wp-content\/uploads\/2018\/08\/TIM%E5%9B%BE%E7%89%8720180816230145-236x300.png\" alt=\"\" \/><\/a><\/p>\n<p>\uff08\u8bad\u7ec3\u8fc7\u7a0b\uff0c\u70b9\u51fb\u56fe\u7247\u53ef\u770b\u5927\u56fe\uff0c\u4e0b\u540c\uff09<\/p>\n<p><\/center><\/p>\n<p>\u5728\u5b66\u4e60\u8fc7\u7a0b\u4e2d\uff0c\u6709\u4e9b\u90e8\u5206\u9700\u8981\u4e0b\u8f7d Wolfram \u63d0\u4f9b\u7684\u795e\u7ecf\u7f51\u7edc\u9884\u8bad\u7ec3\u6a21\u578b\u3002<br \/>\n\u4f46\u7531\u4e8e\u56fd\u5185\u4e0b\u8f7d\u65f6\u5b58\u5728\u7f51\u7edc\u4e0d\u7a33\u5b9a\u7684\u60c5\u51b5\uff0c\u6240\u4ee5\u6211\u5c31\u628a <code>NetModule<\/code> \u4e2d\u6240\u6709\u8d44\u6e90\u90fd\u4e0b\u8f7d\u4e0b\u6765\uff0c\u907f\u514d\u91cd\u88c5\u8f6f\u4ef6\u6216\u5728\u5176\u7535\u8111\u4f7f\u7528\u65f6\u8fd8\u5f97\u518d\u4e0b\u8f7d\u3002\u5e76\u5728\u8fd9\u513f\u5168\u90e8\u5206\u4eab\u7ed9\u5927\u5bb6\u3002<\/p>\n<blockquote><p>\u63d0\u793a\uff1a<code>NetModule<\/code> \u4e2d\u7684\u6a21\u578b\u4f1a\u4e0d\u65ad\u589e\u52a0\uff0c\u672c\u6587\u4f1a\u9646\u7eed\u66f4\u65b0\u65b0\u7684\u8d44\u6e90\u3002<\/p><\/blockquote>\n<h1><span id=\"1\">1. \u8d44\u6e90\u4e0b\u8f7d\u4e0e\u4fdd\u5b58<\/span><\/h1>\n<h2><span id=\"11\">1.1 \u6765\u6e90<\/span><\/h2>\n<p><strong>\u5b98\u65b9\u8d44\u6e90\u5217\u8868\u94fe\u63a5<\/strong><\/p>\n<ul>\n<li><a href=\"http:\/\/resources.wolframcloud.com\/NeuralNetRepository\">Wolfram Neural Net Repository<\/a><\/li>\n<li><a href=\"https:\/\/resources.wolframcloud.com\/NeuralNetRepository\/all\/\">\u6240\u6709\u8d44\u6e90\u5217\u8868<\/a><\/li>\n<\/ul>\n<p>\u4e0a\u9762\u4e24\u4e2a\u9875\u9762\u90fd\u53ef\u4ee5\u6309\u8f93\u5165\u7c7b\u578b\u548c\u4efb\u52a1\u7c7b\u578b\u6765\u8fc7\u6ee4\u6a21\u578b\uff0c\u6bcf\u4e2a\u6a21\u578b\u9875\u9762\u90fd\u6709\u8be5\u6a21\u578b\u7684\u4ecb\u7ecd\u548c\u4f7f\u7528\u6848\u4f8b\u3002\u53e6\u5916\uff0c\u9875\u9762\u4e2d\u4e5f\u4f1a\u8bf4\u660e\u8bad\u7ec3\u96c6\u7684\u6765\u6e90\uff0c\u8fd9\u4e3a\u81ea\u5df1\u8bad\u7ec3\u6a21\u578b\u63d0\u4f9b\u4e86\u4e00\u7cfb\u5217\u4e0d\u9519\u7684\u6570\u636e\u6e90\u3002<\/p>\n<h2><span id=\"12_Wolfram\">1.2 \u767b\u5f55 Wolfram \u8d26\u53f7<\/span><\/h2>\n<p>\u4e0b\u8f7d <code>NetModule<\/code> \u8d44\u6e90\uff0c\u9700\u8981\u5148\u767b\u5f55 Wolfram \u8d26\u53f7\u3002<\/p>\n<p>\u542f\u52a8 Notebook \u540e\uff0c\u4e00\u822c\u4f1a\u8fdb\u884c\u6b22\u8fce\u9875\u9762\uff0c\u6b64\u65f6\u76f4\u63a5\u767b\u5f55\u5373\u53ef\u3002\u5982\u679c\u8df3\u8fc7\u4e86\u6b22\u8fce\u9875\u9762\uff0c\u4e5f\u53ef\u4ee5\u5728 <code>\u5e2e\u52a9 &gt; \u767b\u5f55<\/code> \u4e2d\u767b\u5f55\u60a8\u7684 Wolfram \u8d26\u53f7\u3002<\/p>\n<p>\u5982\u679c\u4ee5\u547d\u4ee4\u884c\u65b9\u6cd5\u8fd0\u884c\uff08\u4e0d\u542f\u52a8 Notebook\uff09\uff0c\u53ef\u4ee5\u5728\u4ee3\u7801\u4e2d\u4f7f\u7528<code>CloudConnect<\/code>\u51fd\u6570\u767b\u5f55\u60a8\u7684 Wolfram \u8d26\u53f7\uff0c\u6bd4\u5982\uff1a<\/p>\n<pre><code class=\"language-mathematica \">CloudConnect[\"\u60a8\u7684 Wolfram \u7528\u6237\u540d\", \"\u60a8\u7684 Wolfram \u5bc6\u7801\"];\n<\/code><\/pre>\n<h2><span id=\"13\">1.3 \u4e0b\u8f7d\u5e76\u4fdd\u5b58\u8d44\u6e90<\/span><\/h2>\n<p>\u5728 Linux \u4e2d\uff0c\u5c06\u4e0b\u5217\u4ee3\u7801\u4fdd\u5b58\u5728 <code>~\/net-model\/download.ws<\/code>\uff0c\u4e4b\u540e\u8fd0\u884c<code>wolframscript -f ~\/net-model\/download.ws<\/code> \u5373\u53ef\u5f00\u59cb\u4e0b\u8f7d\u6a21\u578b\u3002<\/p>\n<p>\u5b8c\u6574\u7684\u4ee3\u7801\uff08\u5305\u542b\u8d26\u53f7\u767b\u5f55\uff09\u5982\u4e0b\uff1a<\/p>\n<pre><code class=\"language-mathematica \">Echo[\"Logining\"];\nrootDir = \"~\/net-model\/\";(*\u3010\u5efa\u8bae\u4fee\u6539\u3011\u5b58\u50a8\u8def\u5f84*)\nCloudConnect[\"\u60a8\u7684 Wolfram \u7528\u6237\u540d\", \"\u60a8\u7684 Wolfram \u5bc6\u7801\"];(*\u3010\u9700\u8981\u4fee\u6539\u3011\u767b\u5f55Wolfram\u8d26\u53f7*)\nEcho[\"Logined\"];\n\nallNetModel = NetModel[];(*\u83b7\u53d6\u6240\u6709\u6a21\u578b\u7684\u540d\u79f0*)\now = OpenWrite[rootDir &lt;&gt; \"net-module-index.txt\"];(*\u6253\u5f00\u5b58\u50a8\u6a21\u578b\u4fe1\u606f\u7684\u6587\u4ef6*)\nIf[! DirectoryQ[rootDir &lt;&gt; \"wlnet\/\"], \n CreateDirectory[rootDir &lt;&gt; \"wlnet\/\"]];(*\u521b\u5efa\u5b58\u50a8\u6a21\u578b\u7684\u76ee\u5f55*)\n\nDo[\n (*\u6a21\u578b\u4fe1\u606f*)\n netName = allNetModel[[k]];(*\u6a21\u578b\u540d\u79f0*)\n Echo[\"loading:\\t\" &lt;&gt; netName];\n uuid = ResourceObject[netName][\"UUID\"];(*\u83b7\u53d6\u6a21\u578bUUID*)\n hash = Hash[netName, \"SHA\", \"HexString\"];\n fileName = rootDir &lt;&gt; \"wlnet\/\" &lt;&gt; netName &lt;&gt; \".wlnet\";\n\n (*\u8df3\u8fc7\u5df2\u7ecf\u4e0b\u8f7d\u7684\u6a21\u578b*)\n If[FileExistsQ[fileName],\n  WriteString[ow, \n   \"load_ed:\\t\" &lt;&gt; uuid &lt;&gt; \"\\t\" &lt;&gt; hash &lt;&gt; \"\\t\" &lt;&gt; netName &lt;&gt; \n    \"\\n\"];(*\u5b58\u50a8\u5df2\u4e0b\u8f7d\u8fc7\u7684\u6a21\u578b\u4fe1\u606f*)\n  Echo[\"skip loaded NetModel:\\t\" &lt;&gt; uuid &lt;&gt; \"\\t\" &lt;&gt; hash &lt;&gt; \"\\t\" &lt;&gt; \n    netName];\n  Continue[]\n  ];\n\n (*\u4e0b\u8f7d\u6a21\u578b*)\n net = NetModel[netName];\n If[net === $Failed,\n  WriteString[ow, \n   \"load_fail\\t\" &lt;&gt; uuid &lt;&gt; \"\\t\" &lt;&gt; hash &lt;&gt; \"\\t\" &lt;&gt; netName &lt;&gt; \n    \"\\n\"];(*\u5b58\u50a8\u4e0b\u8f7d\u5931\u8d25\u7684\u6a21\u578b\u4fe1\u606f*)\n  Echo[\"skip load file NetModel:\\t\" &lt;&gt; uuid &lt;&gt; \"\\t\" &lt;&gt; hash &lt;&gt; \"\\t\" &lt;&gt;\n     netName];\n  Continue[]\n  ];\n\n (*\u5b58\u50a8\u6a21\u578b*)\n Export[fileName, net];\n WriteString[ow, \n  \"load_succ:\\t\" &lt;&gt; uuid &lt;&gt; \"\\t\" &lt;&gt; hash &lt;&gt; \"\\t\" &lt;&gt; netName &lt;&gt; \n   \"\\n\"];(*\u5b58\u50a8\u4e0b\u8f7d\u6210\u529f\u7684\u6a21\u578b\u4fe1\u606f*)\n Echo[\"load succ:\\t\" &lt;&gt; netName];\n , {k, Length[allNetModel]}];\nClose[ow];\n<\/code><\/pre>\n<h1><span id=\"2\">2. \u4f7f\u7528<\/span><\/h1>\n<h2><span id=\"21\">2.1 \u76f4\u63a5\u4f7f\u7528\u65b9\u6cd5<\/span><\/h2>\n<p>\u7531\u4e8e\u4f7f\u7528 Mathematica \u81ea\u5df1\u7684\u6587\u4ef6\u683c\u5f0f\u5b58\u50a8\u7684\u6a21\u578b\uff0c\u6240\u4ee5\u76f4\u63a5\u5bfc\u5165\u5373\u53ef\u4f7f\u7528\u3002<br \/>\n\u5173\u4e8e UUID\uff0c\u53ef\u4ee5\u4f7f\u7528\u51fd\u6570 <code>ResourceObject<\/code> \u67e5\u8be2\uff0c\u4e5f\u53ef\u4ee5\u5728\u5171\u4eab\u7684\u8d44\u6e90\u4e2d\u627e\u5230 <code>__net-module-index.txt<\/code> \u91cc\u9762\u5b58\u50a8\u4e86\u8fd9\u4e9b\u6a21\u578b\u7684\u5bf9\u5e94\u5173\u7cfb\u3002<br \/>\n\u4e5f\u53ef\u4ee5\u4f7f\u7528\u4e0b\u9762\u7684\u5b9a\u4e49\u7684\u51fd\u6570\u6765\u5bfc\u5165\u6a21\u578b\uff08\u6a21\u578b\u7528\u5230\u4e86\u51fd\u6570 <code>ResourceObject<\/code>\uff0c\u9700\u8981\u7f51\u7edc\uff09<\/p>\n<pre><code class=\"language-mathematica \">netModule[dir_,name_]:=Import[FileNameJoin[{dir,ResourceObject[name][\"UUID\"]&lt;&gt;\".wlnet\"}]];\nnet=netModule[\"\u5b58\u50a8\u6a21\u578b\u7684\u76ee\u5f55\",\"LeNet Trained on MNIST Data\"]\n<\/code><\/pre>\n<p>\u6216\u8005\u4f7f\u7528\u4e0b\u9762\u5b9a\u4e49\u7684\u51fd\u6570\u5bfc\u5165\u6a21\u578b\uff08\u4f7f\u7528\u6587\u4ef6 <code>__net-module-index.mx<\/code>\uff0c\u4e0d\u9700\u8981\u7f51\u7edc\uff09<\/p>\n<pre><code class=\"language-mathematica \">&lt;&lt; FileNameJoin[{\"\u5b58\u50a8\u6a21\u578b\u7684\u76ee\u5f55\",\"__net-module-index.mx\"}];\nnetModule[dir_, name_] := Import[FileNameJoin[{dir, nets[name]}]];\nnetModule[\"\u5b58\u50a8\u6a21\u578b\u7684\u76ee\u5f55\",\"LeNet Trained on MNIST Data\"]\n<\/code><\/pre>\n<h2><span id=\"22\">2.2 \u4fee\u6539\u6a21\u578b\u4e0e\u4f7f\u7528\u65b9\u6cd5<\/span><\/h2>\n<p>\u4f7f\u7528\u9884\u8bad\u7ec3\u6a21\u578b\u53ef\u4ee5\u8282\u7701\u8bad\u7ec3\u65f6\u95f4\uff0c\u66f4\u4eba\u5174\u594b\u7684\u662f\uff0c\u4f60\u53ef\u4ee5\u81ea\u7531\u4fee\u6539\u9884\u8bad\u7ec3\u8fc7\u7684\u7f51\u7edc\u3002<br \/>\n\u6bd4\u5982\uff0c\u4f7f\u7528\u7f51\u7edc <a href=\"https:\/\/resources.wolframcloud.com\/NeuralNetRepository\/resources\/Inception-V1-Trained-on-Extended-Salient-Object-Subitizing-Data\">Inception V1 Trained on Extended Salient Object Subitizing Data<\/a> \u53ef\u4ee5\u8bc6\u522b\u56fe\u50cf\u4e2d\u7a81\u51fa\u5bf9\u8c61\u7684\u6570\u91cf\u3002<br \/>\n\u6709\u65f6\u5019\u53ea\u662f\u60f3\u77e5\u9053\u56fe\u50cf\u4e2d\u6709\u6ca1\u6709\u76ee\u6807\u5bf9\u8c61\uff0c\u5373\u4e8c\u5206\u7c7b\u95ee\u9898\uff0c\u90a3\u4e48\u53ef\u4ee5\u901a\u8fc7\u4e0b\u9762\u7684\u65b9\u5f0f\u4fee\u6539\u7f51\u7edc\u7684\u6700\u540e\u51e0\u5c42\u548c\u8f93\u51fa\u5c42\uff1a<\/p>\n<pre><code class=\"language-mathematica \">(*\u5bfc\u5165\u6a21\u578b\uff0c\u51fd\u6570\u5b9a\u4e49\u53c2\u89c1\u4e0a\u9762*)\nnet=netModule[\"\u5b58\u50a8\u6a21\u578b\u7684\u76ee\u5f55\",\"Inception V1 Trained on Extended Salient Object Subitizing Data\"]\n\n(*\u5220\u9664\u6700\u540e\u4e24\u5c42*)\nnetTemp=NetDrop[net,{\"loss3_classifier_sos\",\"loss3_loss3\"}]\n\n(*\u8ffd\u52a0\u4e24\u5c42\uff0c\u5e76\u8fdb\u884c\u521d\u4f7f\u5316\uff0c\u6700\u540e\u5c06\u8f93\u51fa\u8bbe\u7f6e\u4e3a\u4e8c\u5206\u7c7b*)\nnet=NetAppend[netTemp,\n{\n\"loss3_classifier_sos\"-&gt;NetInitialize[LinearLayer[2,\"Input\"-&gt;NetInformation[netTemp,\"OutputPorts\"][\"Output\"]]],\n\"loss3\"-&gt;SoftmaxLayer[]\n},\n\"Output\"-&gt;NetDecoder[{\"Class\",{1,0}}]\n]\n<\/code><\/pre>\n<p><center><br \/>\n<a href=\"https:\/\/mmaqa.com\/blog\/wp-content\/uploads\/2018\/08\/TIM%E5%9B%BE%E7%89%8720180816224544.png\"><img src=\"https:\/\/mmaqa.com\/blog\/wp-content\/uploads\/2018\/08\/TIM%E5%9B%BE%E7%89%8720180816224544-228x300.png\" alt=\"\" \/><\/a><\/center><\/p>\n<p>\u8fd0\u884c\u4e0a\u9762\u7684\u4ee3\u7801\u5373\u53ef\u770b\u5230\u4fee\u6539\u540e\u7684\u7f51\u7edc\uff08Mathematica \u795e\u7ecf\u7f51\u7edc\u5bf9\u8c61\uff09\uff0c\u5982\u679c\u60f3\u770b\u6574\u4f53\u7684\u7ed3\u6784\u56fe\u53ef\u4ee5\u4f7f\u7528\u51fd\u6570 <code>NetInformation<\/code> \uff1a<\/p>\n<pre><code class=\"language-mathematica \">NetInformation[net, \"FullSummaryGraphic\"]\n<\/code><\/pre>\n<p><center><br \/>\n<a href=\"https:\/\/mmaqa.com\/blog\/wp-content\/uploads\/2018\/08\/TIM%E5%9B%BE%E7%89%8720180816224714.png\"><img src=\"https:\/\/mmaqa.com\/blog\/wp-content\/uploads\/2018\/08\/TIM%E5%9B%BE%E7%89%8720180816224714-1024x63.png\" alt=\"\" \/><\/a><\/center><\/p>\n<p>\u51fd\u6570 <code>NetInformation<\/code> \u53ef\u4ee5\u7ed9\u51fa\u7f51\u7edc\u7684\u5404\u79cd\u4fe1\u606f\uff0c\u67e5\u770b\u8be5\u7f51\u7edc\u652f\u6301\u7684\u5c5e\u6027\u53ef\u4ee5\u8fd0\u884c\uff1a<\/p>\n<pre><code class=\"language-mathematica \">NetInformation[net, \"Properties\"]\n<\/code><\/pre>\n","protected":false},"excerpt":{"rendered":"<p>\u6587\u7ae0\u76ee\u5f55\u8d44\u6e90\u4e0b\u8f7d0. \u8bf4\u660e1. \u8d44\u6e90\u4e0b\u8f7d\u4e0e\u4fdd\u5b581.1 \u6765\u6e901.2 \u767b\u5f55 Wolfram \u8d26\u53f71.3 \u4e0b\u8f7d\u5e76\u4fdd\u5b58 [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":436,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"\u65e5\u5fd7\u5934\u56fe":"https:\/\/mmaqa.com\/blog\/wp-content\/uploads\/2018\/08\/TIM\u56fe\u724720180816230145.png","bigfa_ding":"1","\u6253\u8d4f":"checked"},"categories":[11],"tags":[18,17,16],"_links":{"self":[{"href":"https:\/\/mmaqa.com\/blog\/wp-json\/wp\/v2\/posts\/426"}],"collection":[{"href":"https:\/\/mmaqa.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/mmaqa.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/mmaqa.com\/blog\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/mmaqa.com\/blog\/wp-json\/wp\/v2\/comments?post=426"}],"version-history":[{"count":1,"href":"https:\/\/mmaqa.com\/blog\/wp-json\/wp\/v2\/posts\/426\/revisions"}],"predecessor-version":[{"id":584,"href":"https:\/\/mmaqa.com\/blog\/wp-json\/wp\/v2\/posts\/426\/revisions\/584"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/mmaqa.com\/blog\/wp-json\/wp\/v2\/media\/436"}],"wp:attachment":[{"href":"https:\/\/mmaqa.com\/blog\/wp-json\/wp\/v2\/media?parent=426"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/mmaqa.com\/blog\/wp-json\/wp\/v2\/categories?post=426"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/mmaqa.com\/blog\/wp-json\/wp\/v2\/tags?post=426"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}