{"id":1498,"date":"2020-09-03T15:11:46","date_gmt":"2020-09-03T07:11:46","guid":{"rendered":"http:\/\/home.wangjianshuo.com\/cn\/?p=1498"},"modified":"2020-09-03T15:11:46","modified_gmt":"2020-09-03T07:11:46","slug":"%e5%8d%b7%e7%a7%af%e7%bd%91%e7%bb%9c-cnn-%e5%ad%a6%e4%b9%a0%e7%ac%94%e8%ae%b0%e4%b9%8b%e4%ba%8c%ef%bc%9a%e7%bb%86%e8%8a%82%e6%b6%88%e5%a4%b1%ef%bc%8c%e6%8a%bd%e8%b1%a1%e7%94%9f%e6%88%90","status":"publish","type":"post","link":"https:\/\/home.wangjianshuo.com\/cn\/20200903_%e5%8d%b7%e7%a7%af%e7%bd%91%e7%bb%9c-cnn-%e5%ad%a6%e4%b9%a0%e7%ac%94%e8%ae%b0%e4%b9%8b%e4%ba%8c%ef%bc%9a%e7%bb%86%e8%8a%82%e6%b6%88%e5%a4%b1%ef%bc%8c%e6%8a%bd%e8%b1%a1%e7%94%9f%e6%88%90.htm","title":{"rendered":"\u5377\u79ef\u7f51\u7edc CNN \u5b66\u4e60\u7b14\u8bb0\u4e4b\u4e8c\uff1a\u7ec6\u8282\u6d88\u5931\uff0c\u62bd\u8c61\u751f\u6210"},"content":{"rendered":"<div id=\"wechat\">\n<p style=\"text-align: center;\"><img src=\"http:\/\/home.wangjianshuo.com\/cn\/2020\/09\/picture-30.jpg\"\/><\/p>\n<p style=\"text-align: left;\"><span style=\"font-size: 15px;color: rgb(136, 136, 136);\">\u9898\u56fe\uff1a\u9999\u6e2f\u7684\u8717\u5c45<\/span><span style=\"font-size: 15px;color: rgb(136, 136, 136);\"><\/span><\/p>\n<p style=\"text-align: left;\"><span style=\"color:#888888;\"><span style=\"caret-color: rgb(136, 136, 136);font-size: 15px;\">\u7ed8\u4e8e\uff1a2020\u5e743\u6708<\/span><\/span><\/p>\n<p style=\"text-align: left;\"><span style=\"font-size: 15px;color: rgb(136, 136, 136);\"><\/span><\/p>\n<p>\u4e0a\u4e00\u7bc7\uff1a<a data-itemshowtype=\"0\" data-linktype=\"2\" href=\"http:\/\/mp.weixin.qq.com\/s?__biz=MjM5NzI0Mjg0MA==&amp;mid=2652371437&amp;idx=1&amp;sn=57a20e87ac2aaed52adfe3b0e6a1e4d8&amp;chksm=bd3fab5a8a48224c62d0ac433d1c7e1336f41f01fdcd060c0123d93f2acabf24519951f77ebf&amp;scene=21#wechat_redirect\" tab=\"innerlink\" target=\"_blank\" rel=\"noopener noreferrer\">\u5377\u79ef\u7f51\u7edc CNN \u5b66\u4e60\u7b14\u8bb0\u4e4b\u4e00\uff1a\u6211\u4eec\u662f\u600e\u4e48\u8ba4\u8bc60\u7684<\/a><\/p>\n<p>\u6628\u5929\u7559\u4e86\u4e2a\u601d\u8003\u9898\uff0c\u4f60\u7684\u5927\u8111\u662f\u5982\u4f55\u8ba4\u8bc6\u90a3\u4e2a\u4e00\u5708\u7684\uff1f<\/p>\n<p>\u5728\u4f60\u7ffb\u770b\u00a0<a data-itemshowtype=\"0\" data-linktype=\"2\" href=\"http:\/\/mp.weixin.qq.com\/s?__biz=MjM5NzI0Mjg0MA==&amp;mid=2652371437&amp;idx=1&amp;sn=57a20e87ac2aaed52adfe3b0e6a1e4d8&amp;chksm=bd3fab5a8a48224c62d0ac433d1c7e1336f41f01fdcd060c0123d93f2acabf24519951f77ebf&amp;scene=21#wechat_redirect\" style=\"white-space: normal;\" tab=\"innerlink\" target=\"_blank\" rel=\"noopener noreferrer\">\u524d\u4e00\u7bc7\u6587\u7ae0<\/a>\u00a0\u4ee5\u524d\uff0c\u6211\u518d\u95ee\u4f60\u4e00\u4e2a\u95ee\u9898\uff0c\u4f60\u8fd8\u8bb0\u5f97\u4e0a\u4e00\u6b21\u6211\u4eec\u89c1\u5230\u7684 MNIST \u6570\u636e\u96c6\u91cc\u9762\u7684\u90a3\u4e2a 0 \u662f\u5982\u4e0b\u54ea\u4e00\u526f\u5417\uff1f\u6216\u8005\u964d\u4f4e\u70b9\u96be\u5ea6\uff0c\u548c\u5982\u4e0b\u54ea\u4e00\u4e2a 0 \u66f4\u63a5\u8fd1\u5462\uff1f<\/p>\n<p style=\"text-align: center;\"><img src=\"http:\/\/home.wangjianshuo.com\/cn\/2020\/09\/picture-31.jpg\"\/><\/p>\n<p style=\"text-align: center;\"><span style=\"font-size: 15px;color: rgb(136, 136, 136);\">\u8bad\u7ec3\u96c6\u91cc\u7684\u5404\u79cd\u00a00\u00a0\uff08\u751f\u6210\u8fd9\u4e2a\u56fe\u7684\u4ee3\u7801\u89c1\u00a0\u4ee3\u7801\u4e00\uff09<\/span><\/p>\n<p style=\"text-align: left;\">\u6216\u8bb8\u4f60\u548c\u6211\u4e00\u6837\uff0c\u8bb0\u4e0d\u592a\u6e05\u4e86\u3002<span style=\"text-align: left;\">\u6ca1\u5173\u7cfb\uff0c\u8fd9\u6b63\u662f\u6211\u8981\u8bb2\u7684\uff0c\u5c31\u662f\u795e\u7ecf\u7f51\u7edc\u5728<\/span><span style=\"text-align: left;\">\u5904\u7406\u4fe1\u606f\u7684\u65f6\u5019\uff0c\u5f53\u5f97\u5230\u9ad8\u4e00\u5c42\u7684\u4fe1\u606f\u7684\u65f6\u5019\u5c31\u4f1a\u5ffd\u7565\u5e95\u5c42\u7684\u4fe1\u606f\uff0c\u6545\u610f\u7684\u9057\u5fd8\u3002\u7136\u540e\u57fa\u4e8e\u8fd9\u4e00\u5c42\u7684\u4fe1\u606f\u518d\u5f80\u4e0a\u63a8\u5bfc\u4e00\u5c42\u66f4\u9ad8\u5c42\u6b21\u7684\u4fe1\u606f\uff0c\u518d\u628a\u4e0b\u9762\u7684\u4fe1\u606f\u9057\u5fd8\u3002\u5982\u6b64\u5f88\u591a\u5c42\uff0c\u76f4\u5230\u5f97\u5230\u6700\u7ec8\u7684\u7ed3\u8bba\u3002\u8fd9\u662f\u4e00\u4e2a\u4e0d\u65ad\u4e22\u5931\u4fe1\u606f\uff0c\u5f62\u6210\u4e00\u4e2a\u65b0\u7684\u4fe1\u606f\u7684\u8fc7\u7a0b\u3002<\/span><\/p>\n<p style=\"text-align: left;\"><span style=\"text-align: left;\"><\/span><\/p>\n<p style=\"text-align: left;\"><span style=\"text-align: left;\">\u4ee5\u4e0a\u9762\u7684\u56fe\u50cf\u4e3a\u4f8b\uff0c\u539f\u59cb\u7684\u4fe1\u606f\u91cf\u5176\u5b9e\u975e\u5e38\u5927\uff0c\u90a3\u662f\u4e00\u4e2a 28 x 28 x 256 \uff08\u4e5f\u5c31\u662f 28\u884c\uff0c28\u5217\uff0c\u6bcf\u4e2a\u70b9\u6709256\u79cd\u53ef\u80fd\u6027\u7684\u56fe\uff09\u7684\u6570\u636e\uff0c\u4e5f\u5c31\u662f20\u4e07\u79cd\u53ef\u80fd\u6027\u3002\u800c\u7ecf\u8fc7\u6211\u4eec\u7684\u4eba\u8089\u795e\u7ecf\u7f51\u7edc\u5904\u7406\u4ee5\u540e\u5f62\u6210\u4e86 0 &#8211; 9 \u8fd910\u4e2a\u6570\u5b57\u4e2d\u95f4\u7684\u4e00\u4e2a\u6570\u5b57\uff08 0\u00a0\uff09\uff0c\u4ece 20 \u4e07\u79cd\u53ef\u80fd\u6027\u5230 10\u00a0\u79cd\u53ef\u80fd\u6027\u8fd9\u4e2a\u8fc7\u7a0b\u5c31\u662f\u795e\u7ecf\u7f51\u7edc\u7684\u529f\u52b3\u3002<\/span>\u5426\u5219\uff0c\u4eba\u773c\u775b\u6709\u516d\u767e\u4e07\u4ee5\u4e0a\u4e2a\u89c6\u9525\uff0c\u773c\u775b\u770b\u5230\u7684\u8fd8\u662f\u8fde\u7eed\u7684\u89c6\u9891\u7684\u4fe1\u606f\uff0c\u8981\u662f\u6211\u4eec\u7684\u5e95\u5c42\u795e\u7ecf\u5143\u4e0d\u5ffd\u7565\u6389\u8fd9\u4e9b\u4fe1\u606f\u800c\u76f4\u63a5\u90fd\u4f20\u5bfc\u5230\u4e0a\u4e00\u5c42\u795e\u7ecf\u5143\u7684\u8bdd\uff0c\u6211\u4eec\u8111\u5b50\u91cc\u9762\u5c31\u4f1a\u5806\u6ee1\u4e86\u6240\u6709\u7684\u7ec6\u8282\uff0c\u5c31\u65e0\u6cd5\u601d\u8003\u66f4\u62bd\u8c61\u7684\u4e1c\u897f\u4e86\u3002<\/p>\n<p style=\"text-align: left;\"><span style=\"text-align: left;\"><\/span><\/p>\n<p style=\"text-align: left;\"><strong>\u5c0f\u91cf\u7ef4\u5ea6\u7684\u5927\u91cf\u4fe1\u606f\u00a0&#8211;&gt;\u00a0\u5927\u91cf\u7ef4\u5ea6\u7684\u5c0f\u91cf\u4fe1\u606f<\/strong><\/p>\n<p style=\"text-align: left;\"><span style=\"text-align: left;\"><\/span><\/p>\n<p style=\"text-align: left;\">\u867d\u7136\u56fe\u50cf\u7684\u7ec6\u8282\u4e22\u5931\u4e86\uff0c\u4f46\u66f4\u9ad8\u4e00\u5c42\u7684\u8ba4\u77e5\u5728\u5927\u8111\u91cc\u9762\u8bde\u751f\u4e86\u3002\u6bd4\u5982\u770b\u4e86\u8fd9\u4e2a\u56fe\u50cf\uff0c\u6211\u95ee\u4f60\u8fd9\u51e0\u4e2a\u95ee\u9898\uff1a<\/p>\n<section class=\"code-snippet__fix code-snippet__js\">\n<ul class=\"code-snippet__line-index code-snippet__js\">\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<\/ul>\n<pre class=\"code-snippet__js\"><code><span class=\"code-snippet_outer\">\u8fd9\u662f\u4e00\u4e2a\u9ed1\u767d\u56fe\u50cf\u5417\uff1f\u3010\u662f\u3011<\/span><\/code><code><span class=\"code-snippet_outer\">\u56fe\u50cf\u91cc\u9762\u6709\u4e00\u4e2a\u5708\u5417\uff1f\u3010\u662f\u3011<\/span><\/code><code><span class=\"code-snippet_outer\">\u56fe\u50cf\u91cc\u9762\u7ebf\u6761\u8fde\u7eed\u5417\uff1f\u3010\u662f\u3011<\/span><\/code><code><span class=\"code-snippet_outer\">\u8fd9\u662f\u4e2a\u6b63\u65b9\u5f62\u56fe\u7247\u5417\uff1f\u3010\u662f\u3011<\/span><\/code><\/pre>\n<\/section>\n<p style=\"white-space: normal;text-align: left;\">\u8fd9\u6837\u7684\u95ee\u9898\u6211\u53ef\u4ee5\u95ee\u51e0\u5341\u4e2a\uff0c\u5bf9\u4e8e\u8fd9\u4e9b\u95ee\u9898\u6211\u4eec\u7684\u795e\u7ecf\u7f51\u7edc\u90fd\u53ef\u4ee5\u56de\u7b54\uff0c\u4f46\u8fd9\u4e9b\u95ee\u9898\u7684\u7b54\u6848\u90fd\u6ca1\u6709\u76f4\u63a5\u5199\u5728\u539f\u59cb\u7684 28 x 28 \u7684\u6570\u636e\u91cc\u9762\u7684\uff0c\u800c\u662f\u6211\u4eec\u901a\u8fc7\u5e95\u5c42\u7684\u4fe1\u606f\uff0c\u4e00\u5c42\u4e00\u5c42\u7684\u62bd\u8c61\u51fa\u6765\u7684\u3002<\/p>\n<p style=\"white-space: normal;text-align: left;\">\u8bf7\u6ce8\u610f\uff0c\u4f60\u6709\u6ca1\u6709\u53d1\u73b0\u6700\u65e9\u7684\u6570\u636e\u662f\u5728\u4e00\u4e2a\u7ef4\u5ea6\uff08\u9ed1\u6216\u767d\u7684\u7a0b\u5ea6\uff09\u4e0a\u7684\u5927\u91cf\u6570\u636e\uff0c\u4e00\u6b65\u6b65\u7684\u53d8\u6210\u5728\u591a\u4e2a\u7ef4\u5ea6\u7684\u5c11\u91cf\u6570\u636e\u3002\u201c\u8fd9\u662f\u9ed1\u767d\u56fe\u50cf\u5417\u201d\u5c31\u662f\u4e00\u4e2a\u7ef4\u5ea6\uff0c\u8fd9\u4e2a\u7ef4\u5ea6\u7684\u6570\u636e\u53ea\u6709 \u4e00\u6bd4\u7279\uff08\u662f\/\u4e0d\u662f\uff09\uff1b\u56fe\u50cf\u91cc\u9762\u6709\u4e00\u4e2a\u5708\u5417\uff1f\u662f\u53e6\u5916\u4e00\u4e2a\u7ef4\u5ea6\uff0c\u8fd9\u6837\u7684\u7ef4\u5ea6\u5f88\u591a\u5f88\u591a\u3002<\/p>\n<p style=\"white-space: normal;text-align: left;\">\u4eba\u5de5\u667a\u80fd\u5c31\u662f\u4ece\u7070\u5ea6\u6216\u8005\u5f69\u8272\u7684\u56fe\u50cf\u91cc\u9762\uff0c\u5148\u8ba4\u8bc6\u8fb9\u754c\uff0c\u7136\u540e\u8ba4\u8bc6\u7ebf\uff0c\u7136\u540e\u8ba4\u8bc6\u5f62\u72b6\uff0c\u7136\u540e\u518d\u5728\u8fd9\u4e9b\u5706\u548c\u65b9\u7684\u7ec4\u5408\u518d\u8ba4\u8bc6\u4eba\u8138\uff0c\u7136\u540e\u518d\u8ba4\u8bc6\u4eba\uff0c\u7136\u540e\u518d\u4ece\u4eba\u8ba4\u8bc6\u573a\u666f\u7b49\u7b49\u3002\u8fd9\u6837\u4e00\u5c42\u5c42\u7684\u5b66\u4e60\u7684\u3002\u56e0\u4e3a\u5b66\u4e9b\u7684\u5c42\u6b21\u5f88\u591a\uff0c\u5f88\u6df1\uff0c\u6211\u4eec\u5e38\u628a\u8fd9\u79cd\u6a21\u578b\u53eb\u505a\u6df1\u5ea6\u5b66\u4e60 &#8211; \u6df1\u5ea6\u5c31\u662f\u5f88\u591a\u5c42\u7684\u610f\u601d\u3002<\/p>\n<p style=\"white-space: normal;text-align: left;\"><strong>\u4eba\u8111\u662f\u600e\u4e48\u8ba4\u8bc6 0 \u7684\uff1f<\/strong><\/p>\n<p style=\"text-align: left;\">\u73b0\u5728\u6211\u4eec\u51c6\u5907\u6765\u89e3\u51b3\u8fd9\u4e2a\u95ee\u9898\u3002\u76f4\u63a5\u4ece\u539f\u59cb\u7684\u6570\u636e\u4e2d\u770b\u51fa\u6765\u6700\u7ec8\u7684\u6570\u5b57\u4e0d\u5bb9\u6613\uff0c\u4f46\u5982\u679c\u8ba1\u7b97\u673a\u628a\u539f\u59cb\u6570\u636e\u7b80\u5316\u6210\u5982\u4e0b\u8fd916\u4e2a\u95ee\u9898\u7684\u7b54\u6848\uff0c\u4f60\u662f\u4e0d\u662f\u89c9\u5f97\u5bb9\u6613\u4e00\u4e9b\uff1f\u4e3a\u4e86\u7b80\u5316\u6211\u7528\u00a00.0\u00a0\u4ee3\u8868\u6ca1\u6709\uff0c1.0 \u4ee3\u8868\u6709\u3002\uff08\u5b9e\u9645\u60c5\u51b5\u53ef\u80fd\u4e0d\u50cf 0 \u548c 1 \u8fd9\u4e48\u7edd\u5bf9\uff0c\u800c\u662f\u5b83\u4eec\u4e4b\u95f4\u7684\u4e00\u4e2a\u6570\u5b57\uff09<\/p>\n<section class=\"code-snippet__fix code-snippet__js\">\n<ul class=\"code-snippet__line-index code-snippet__js\">\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<\/ul>\n<pre class=\"code-snippet__js\" data-lang=\"css\"><code><span class=\"code-snippet_outer\">\u8fd9\u4e2a\u56fe\u7247\u7684\u5de6\u4e0a\u89d2\u533a\u57df\uff1a<\/span><\/code><code><span class=\"code-snippet_outer\">\u3000\u3000\u3000\u6709\u3000\\\u3000\u00a0\u53cd\u659c\u7ebf\u5417\uff1f\u3000\u3000\u3000\u30100<span class=\"code-snippet__selector-class\">.0<\/span>\u3011<\/span><\/code><code><span class=\"code-snippet_outer\">\u3000\u3000\u3000\u6709\u3000\uff5c\u3000\u5782\u76f4\u7684\u7ebf\u6761\u5417\uff1f\u3000\u30101<span class=\"code-snippet__selector-class\">.0<\/span>\u3011\u3000<\/span><\/code><code><span class=\"code-snippet_outer\">\u3000\u3000\u3000\u6709\u3000\/\u3000\u00a0\u6b63\u659c\u7ebf\u5417\uff1f\u3000\u3000\u3000\u30101<span class=\"code-snippet__selector-class\">.0<\/span>\u3011<\/span><\/code><code><span class=\"code-snippet_outer\">\u3000\u3000\u3000\u6709\u00a0\u2013\u3000\u3000\u6a2a\u7ebf\u5417\uff1f\u3000\u3000\u3000\u3000\u30101<span class=\"code-snippet__selector-class\">.0<\/span>\u3011<\/span><\/code><code><span class=\"code-snippet_outer\">\u8fd9\u4e2a\u56fe\u7247\u7684\u53f3\u4e0a\u89d2\u533a\u57df\uff1a<\/span><\/code><code><span class=\"code-snippet_outer\">\u3000\u3000\u3000\u6709\u3000\\\u3000 \u53cd\u659c\u7ebf\u5417\uff1f\u3000\u3000\u3000\u30101<span class=\"code-snippet__selector-class\">.0<\/span>\u3011<\/span><\/code><code><span class=\"code-snippet_outer\">\u3000\u3000\u3000\u6709\u3000\uff5c\u3000\u5782\u76f4\u7684\u7ebf\u6761\u5417\uff1f\u3000\u30101<span class=\"code-snippet__selector-class\">.0<\/span>\u3011\u3000<\/span><\/code><code><span class=\"code-snippet_outer\">\u3000\u3000\u3000\u6709\u3000\/\u3000 \u6b63\u659c\u7ebf\u5417\uff1f\u3000\u3000\u3000\u30100<span class=\"code-snippet__selector-class\">.0<\/span>\u3011<\/span><\/code><code><span class=\"code-snippet_outer\">\u3000\u3000\u3000\u6709\u00a0\u2013\u3000\u3000\u6a2a\u7ebf\u5417\uff1f\u3000\u3000\u3000\u3000\u30101<span class=\"code-snippet__selector-class\">.0<\/span>\u3011<\/span><\/code><code><span class=\"code-snippet_outer\">\u8fd9\u4e2a\u56fe\u7247\u7684\u5de6\u4e0b\u89d2\u533a\u57df\uff1a<\/span><\/code><code><span class=\"code-snippet_outer\">\u3000\u3000\u3000\u6709\u3000\\\u3000\u00a0\u53cd\u659c\u7ebf\u5417\uff1f\u3000\u00a0\u00a0\u00a0\u30101<span class=\"code-snippet__selector-class\">.0<\/span>\u3011<\/span><\/code><code><span class=\"code-snippet_outer\">\u3000\u3000\u3000\u6709\u3000\uff5c\u3000\u5782\u76f4\u7684\u7ebf\u6761\u5417\uff1f\u3000\u30101<span class=\"code-snippet__selector-class\">.0<\/span>\u3011\u3000<\/span><\/code><code><span class=\"code-snippet_outer\">\u3000\u3000\u3000\u6709\u3000\/\u3000 \u6b63\u659c\u7ebf\u5417\uff1f\u3000\u3000\u3000\u30100<span class=\"code-snippet__selector-class\">.0<\/span>\u3011<\/span><\/code><code><span class=\"code-snippet_outer\">\u3000\u3000\u3000\u6709\u00a0\u2013\u3000\u3000\u6a2a\u7ebf\u5417\uff1f\u3000\u3000\u3000\u3000\u30101<span class=\"code-snippet__selector-class\">.0<\/span>\u3011<\/span><\/code><code><span class=\"code-snippet_outer\">\u8fd9\u4e2a\u56fe\u7247\u7684\u53f3\u4e0b\u89d2\u533a\u57df\uff1a<\/span><\/code><code><span class=\"code-snippet_outer\">\u3000\u3000\u3000\u6709\u3000\\\u3000 \u53cd\u659c\u7ebf\u5417\uff1f\u3000\u3000\u3000\u30100<span class=\"code-snippet__selector-class\">.0<\/span>\u3011<\/span><\/code><code><span class=\"code-snippet_outer\">\u3000\u3000\u3000\u6709\u3000\uff5c\u3000\u5782\u76f4\u7684\u7ebf\u6761\u5417\uff1f\u3000\u30101<span class=\"code-snippet__selector-class\">.0<\/span>\u3011\u3000<\/span><\/code><code><span class=\"code-snippet_outer\">\u3000\u3000\u3000\u6709\u3000\/\u3000 \u6b63\u659c\u7ebf\u5417\uff1f\u3000\u3000\u3000\u30101<span class=\"code-snippet__selector-class\">.0<\/span>\u3011<\/span><\/code><code><span class=\"code-snippet_outer\">\u3000\u3000\u3000\u6709\u00a0\u2013\u3000\u3000\u6a2a\u7ebf\u5417\uff1f\u3000\u3000\u3000\u3000\u30101<span class=\"code-snippet__selector-class\">.0<\/span>\u3011<\/span><\/code><\/pre>\n<\/section>\n<p style=\"text-align: left;\">\u5982\u679c\u4f60\u77e5\u9053\u5982\u4e0a 16 \u4e2a\u95ee\u9898\u7684\u7b54\u6848\uff0c\u5373\u4f7f\u4e0d\u770b\u5230\u539f\u59cb\u7684\u6570\u636e\uff0c\u4e5f\u80fd\u8111\u8865\u51fa\u6765\u4e00\u4e2a 0 \u7684\u6837\u5b50 \uff08\u5c31\u662f\u5de6\u4e0a\u89d2\u6709\u6a2a\u7ebf\uff0c\u7136\u540e\u8f6c\u5230\u53f3\u4e0a\u5230\u5de6\u4e0b\u7684\u659c\u7ebf\uff0c\u7136\u540e\u662f\u7ad6\u7ebf\uff0c\u7136\u540e\u63a5\u5de6\u4e0b\u89d2\u7684\u56db\u79cd\u7ebf\u6bb5\uff09\u3002<\/p>\n<p style=\"text-align: left;\">\u518d\u770b\u8fd9\u5f20\u56fe\u5462\uff1f<\/p>\n<section class=\"code-snippet__fix code-snippet__js\">\n<ul class=\"code-snippet__line-index code-snippet__js\">\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<\/ul>\n<pre class=\"code-snippet__js\" data-lang=\"css\"><code><span class=\"code-snippet_outer\">\u8fd9\u4e2a\u56fe\u7247\u7684\u5de6\u4e0a\u89d2\u533a\u57df\uff1a<\/span><\/code><code><span class=\"code-snippet_outer\"> 1\u3000\u3000\u3000\u6709\u3000\\\u3000\u00a0\u53cd\u659c\u7ebf\u5417\uff1f\u00a0\u00a0\u00a0\u30100<span class=\"code-snippet__selector-class\">.0<\/span>\u3011<\/span><\/code><code><span class=\"code-snippet_outer\"> 2\u3000\u3000\u3000\u6709\u3000\uff5c\u3000\u5782\u76f4\u7684\u7ebf\u6761\u5417\uff1f\u30101<span class=\"code-snippet__selector-class\">.0<\/span>\u3011\u3000<\/span><\/code><code><span class=\"code-snippet_outer\"> 3\u3000\u3000\u3000\u6709\u3000\/\u3000\u00a0\u6b63\u659c\u7ebf\u5417\uff1f\u00a0\u00a0\u00a0\u30100<span class=\"code-snippet__selector-class\">.0<\/span>\u3011<\/span><\/code><code><span class=\"code-snippet_outer\"> 4\u3000\u3000\u3000\u6709\u00a0\u2013\u3000\u3000\u6a2a\u7ebf\u5417\uff1f\u3000\u3000\u3000\u30100<span class=\"code-snippet__selector-class\">.0<\/span>\u3011<\/span><\/code><code><span class=\"code-snippet_outer\">\u8fd9\u4e2a\u56fe\u7247\u7684\u53f3\u4e0a\u89d2\u533a\u57df\uff1a<\/span><\/code><code><span class=\"code-snippet_outer\"> 5\u3000\u3000\u3000\u6709\u3000\\\u3000\u00a0\u53cd\u659c\u7ebf\u5417\uff1f\u00a0\u00a0\u00a0\u30100<span class=\"code-snippet__selector-class\">.0<\/span>\u3011<\/span><\/code><code><span class=\"code-snippet_outer\"> 6\u3000\u3000\u3000\u6709\u3000\uff5c\u3000\u5782\u76f4\u7684\u7ebf\u6761\u5417\uff1f\u30101<span class=\"code-snippet__selector-class\">.0<\/span>\u3011\u3000<\/span><\/code><code><span class=\"code-snippet_outer\">\u00a07\u3000\u3000\u3000\u6709\u3000\/\u3000\u00a0\u6b63\u659c\u7ebf\u5417\uff1f\u3000\u3000\u30100<span class=\"code-snippet__selector-class\">.0<\/span>\u3011<\/span><\/code><code><span class=\"code-snippet_outer\"> 8\u3000\u3000\u3000\u6709\u00a0\u2013\u3000\u3000\u6a2a\u7ebf\u5417\uff1f\u3000\u3000\u3000\u30100<span class=\"code-snippet__selector-class\">.0<\/span>\u3011<\/span><\/code><code><span class=\"code-snippet_outer\">\u8fd9\u4e2a\u56fe\u7247\u7684\u5de6\u4e0b\u89d2\u533a\u57df\uff1a<\/span><\/code><code><span class=\"code-snippet_outer\"> 9\u3000\u3000\u3000\u6709\u3000\\\u3000\u00a0\u53cd\u659c\u7ebf\u5417\uff1f\u00a0\u00a0\u00a0\u30100<span class=\"code-snippet__selector-class\">.0<\/span>\u3011<\/span><\/code><code><span class=\"code-snippet_outer\">10\u3000\u3000\u3000\u6709\u3000\uff5c\u3000\u5782\u76f4\u7684\u7ebf\u6761\u5417\uff1f\u30101<span class=\"code-snippet__selector-class\">.0<\/span>\u3011\u3000<\/span><\/code><code><span class=\"code-snippet_outer\">11\u3000\u3000\u3000\u6709\u3000\/\u3000\u00a0\u6b63\u659c\u7ebf\u5417\uff1f\u3000\u3000\u30100<span class=\"code-snippet__selector-class\">.0<\/span>\u3011<\/span><\/code><code><span class=\"code-snippet_outer\">12\u3000\u3000\u3000\u6709\u00a0\u2013\u3000\u3000\u6a2a\u7ebf\u5417\uff1f\u3000\u3000\u3000\u30100<span class=\"code-snippet__selector-class\">.0<\/span>\u3011<\/span><\/code><code><span class=\"code-snippet_outer\">\u8fd9\u4e2a\u56fe\u7247\u7684\u53f3\u4e0b\u89d2\u533a\u57df\uff1a<\/span><\/code><code><span class=\"code-snippet_outer\">13\u3000\u3000\u3000\u6709\u3000\\\u3000\u00a0\u53cd\u659c\u7ebf\u5417\uff1f\u00a0\u00a0\u00a0\u30100<span class=\"code-snippet__selector-class\">.0<\/span>\u3011<\/span><\/code><code><span class=\"code-snippet_outer\">14\u3000\u3000\u3000\u6709\u3000\uff5c\u3000\u5782\u76f4\u7684\u7ebf\u6761\u5417\uff1f\u30101<span class=\"code-snippet__selector-class\">.0<\/span>\u3011\u3000<\/span><\/code><code><span class=\"code-snippet_outer\">15\u3000\u3000\u3000\u6709\u3000\/\u3000\u00a0\u6b63\u659c\u7ebf\u5417\uff1f\u00a0\u00a0\u00a0\u30100<span class=\"code-snippet__selector-class\">.0<\/span>\u3011<\/span><\/code><code><span class=\"code-snippet_outer\">16\u3000\u3000\u3000\u6709\u00a0\u2013\u3000\u3000\u6a2a\u7ebf\u5417\uff1f\u3000\u3000\u3000\u30100<span class=\"code-snippet__selector-class\">.0<\/span>\u3011<\/span><\/code><\/pre>\n<\/section>\n<p style=\"text-align: left;\">\u6211\u731c\u8fd9\u4e2a\u6570\u662f 1 \uff0c\u548c\u4f60\u731c\u7684\u4e00\u6837\u5417\uff1f\u56e0\u4e3a\u4ece\u6570\u636e\u4e0a\u770b\uff0c\u53ea\u6709\u7ad6\u7ebf\uff0c\u6ca1\u6709\u659c\u7ebf\u548c\u6a2a\u7ebf\u3002\u3002\u3002\u3002<\/p>\n<p style=\"text-align: left;\">\u5230\u4e86\u8fd9\u4e00\u6b65\u5f53\u7136\u600e\u4e48\u505a\u90fd\u884c\uff0c\u6700\u571f\u7684\u4e5f\u662f\u6700\u4e0d\u53ef\u6269\u5c55\u7684\uff0c\u5c31\u662f\u786c\u5199\u7a0b\u5e8f\uff1a<\/p>\n<section class=\"code-snippet__fix code-snippet__js\">\n<ul class=\"code-snippet__line-index code-snippet__js\">\n<li><\/li>\n<li><\/li>\n<\/ul>\n<pre class=\"code-snippet__js\"><code><span class=\"code-snippet_outer\">\u5982\u679c 1\uff0c9\uff0c14\uff0c17 \u8fd9\u56db\u4e2a\u95ee\u9898\u7684\u7b54\u6848\u662f 0 \u5176\u4ed6\u95ee\u9898\u7b54\u6848\u90fd\u662f 1 \u7684\u8bdd\uff0c\u7ed3\u679c\u662f 0<\/span><\/code><code><span class=\"code-snippet_outer\">\u5982\u679c\u6240\u6709\u7684\u7ad6\u7ebf\u7684\u95ee\u9898\u90fd\u662f\u00a01\uff0c\u00a0\u4f46\u662f\u5176\u4ed6\u7684\u95ee\u9898\u7b54\u6848\u90fd\u662f\u00a00\u00a0\u7684\u8bdd\uff0c\u00a0\u7ed3\u679c\u662f\u00a01<\/span><\/code><\/pre>\n<\/section>\n<p style=\"text-align: left;\">\u800c\u5b9e\u9645\u4e0a\uff0c\u5927\u5bb6\u8fd8\u662f\u7528\u6570\u5b66\u4e0a\u9762\u7684\u7ebf\u6027\u56de\u5f52\u6765\u89e3\u51b3\u8fd9\u6837\u7684\u95ee\u9898\u3002\u6211\u4eec\u5047\u8bbe\u6240\u6709\u7684\u95ee\u9898\u90fd\u662f Y = a * X + b \u8fd9\u6837\u7684\u7ed3\u6784\uff0c\u53ea\u8981\u6709\u8db3\u591f\u7684\uff08X\uff0cY\uff09\u6570\u636e\u53ef\u4ee5\u8ba9\u6211\u4eec\u8bad\u7ec3\uff0c\u627e\u5230 a \u548c b \u8fd9\u4e24\u4e2a\u53c2\u6570\u5c31\u53ef\u4ee5\u3002\u4e5f\u5c31\u662f\u8bf4\uff0c\u8f93\u5165\u662f 16 \u4e2a\u6570\u5b57 \uff08x1, x2, \u00a0x3, x4, x5, &#8230;, x16\uff09\uff0c\u5206\u522b\u4ee3\u8868\u5bf9\u4e8e\u5982\u4e0a 1 \u5230 16 \u53f7\u95ee\u9898\u7684\u7b54\u6848\u3002\u8f93\u51fa\u662f 10 \u4e2a\u6570\u5b57 \uff08y1, y2, y3, &#8230;., y10) \u5206\u522b\u4ee3\u8868\u8fd9\u4e2a\u6570\u5b57\u662f 0 &#8211; 9 \u7684\u53ef\u80fd\u6027\u3002\u5f53\u6211\u4eec\u6709\u4e8660000\u591a\u7ec4\u8fd9\u6837\u7684\u5bf9\u5e94\u4ee5\u540e\uff0c\u5c31\u53ef\u4ee5\u627e\u5230\u89c4\u5f8b\uff0c\u518d\u7ed9\u6211\u4e00\u7ec4 x\uff0c\u6211\u5c31\u53ef\u4ee5\u8f93\u51fa\u4e00\u7ec4 y \u3002\u8fd9\u90e8\u5206\u662f\u6700\u57fa\u7840\u7684\u7ebf\u6027\u4ee3\u6570\uff0c\u7528 Excel \u5c31\u53ef\u4ee5\u8f7b\u677e\u641e\u5b9a\u7684\uff0c\u8fd8\u6ca1\u6709\u6d89\u53ca\u6700\u8fd1\u51fa\u73b0\u7684\u4eba\u5de5\u667a\u80fd\u6216\u8005\u5377\u79ef\u795e\u7ecf\u7f51\u7edc\u3002\u800c\u63a5\u4e0b\u6765\u6709\u8da3\u7684\u90e8\u5206\uff0c\u5c31\u662f\u5982\u679c\u4ece\u56fe\u50cf\u4fe1\u606f\u8f6c\u6362\u5230\u8fd9\u6837\u7684 16 \u4e2a\u95ee\u9898\u7b54\u6848\u3002\u8fd9\u662f\u540e\u9762\u7684\u6587\u7ae0\u9700\u8981\u89e3\u51b3\u7684\u95ee\u9898\u3002<\/p>\n<p style=\"white-space: normal;text-align: left;\"><strong>\u591a\u5c42\u7684\u795e\u7ecf\u7f51\u7edc<\/strong><\/p>\n<p style=\"white-space: normal;text-align: left;\">\u5f53\u7136\u5230\u8fd9\u91cc\u6211\u4eec\u8fd8\u8fdc\u8fdc\u6ca1\u6709\u56de\u7b54\u00a0<a data-itemshowtype=\"0\" data-linktype=\"2\" href=\"http:\/\/mp.weixin.qq.com\/s?__biz=MjM5NzI0Mjg0MA==&amp;mid=2652371437&amp;idx=1&amp;sn=57a20e87ac2aaed52adfe3b0e6a1e4d8&amp;chksm=bd3fab5a8a48224c62d0ac433d1c7e1336f41f01fdcd060c0123d93f2acabf24519951f77ebf&amp;scene=21#wechat_redirect\" tab=\"innerlink\" target=\"_blank\" rel=\"noopener noreferrer\">\u6211\u4eec\u662f\u600e\u4e48\u8ba4\u8bc60\u7684<\/a>\u00a0\u3002\u4e0a\u9762\u7684\u90e8\u5206\u5176\u5b9e\u5e0c\u671b\u5728\u6211\u5168\u90e8\u8bb2\u5b8c\u4e86\u4ee5\u540e\u518d\u56de\u8fc7\u5934\u6765\u770b\u624d\u6709\u53ef\u80fd\u770b\u61c2\u3002<\/p>\n<p style=\"white-space: normal;text-align: left;\">\u4f46\u6211\u4eec\u5148\u6253\u4f4f\uff0c\u770b\u4e00\u6bb5\u4ee3\u7801\uff0c\u7136\u540e\u7528\u540e\u9762\u7684\u51e0\u6b21\u6587\u7ae0\u901a\u8fc7\u4ece\u4ee3\u7801\u7684\u89d2\u5ea6\uff0c\u6765\u7ec6\u81f4\u7684\u628a\u521a\u624d\u4e24\u6bb5\u8bf4\u7684\u6bcf\u4e00\u4e2a\u7ec6\u8282\u4e86\u89e3\u6e05\u695a\u3002<\/p>\n<section class=\"code-snippet__fix code-snippet__js\">\n<ul class=\"code-snippet__line-index code-snippet__js\">\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<\/ul>\n<pre class=\"code-snippet__js\" data-lang=\"cs\"><code><span class=\"code-snippet_outer\"><span class=\"code-snippet__meta\"># \u5bfc\u5165\u8457\u540d\u7684 tensorflow\u3002\u5728\u547d\u4ee4\u884c\u4e0b\u7528 pip3 install tensorflow \u5b89\u88c5<\/span><\/span><\/code><code><span class=\"code-snippet_outer\">import tensorflow <span class=\"code-snippet__keyword\">as<\/span> tf<\/span><\/code><code><span class=\"code-snippet_outer\">import numpy <span class=\"code-snippet__keyword\">as<\/span> np<\/span><\/code><code><span class=\"code-snippet_outer\"><span class=\"code-snippet__keyword\">from<\/span> tensorflow.keras.models import Sequential<\/span><\/code><code><span class=\"code-snippet_outer\"><span class=\"code-snippet__keyword\">from<\/span> tensorflow.keras.layers import Conv2D, Flatten, MaxPooling2D, Dense, BatchNormalization<\/span><\/code><code><span class=\"code-snippet_outer\"><span class=\"code-snippet__meta\"># \u4ece\u4e92\u8054\u7f51\u4e0a\u4e0b\u8f7d mnist \u6570\u636e\u96c6\u5230\u672c\u5730 ~\/.keras\/datasets\/mnist.npz<\/span><\/span><\/code><code><span class=\"code-snippet_outer\"><span class=\"code-snippet__meta\"># x_train, y_train \u5206\u522b\u662f\u662f60000\u4e2a\u8bad\u7ec3\u56fe\u50cf\u548c\u7b54\u6848<\/span><\/span><\/code><code><span class=\"code-snippet_outer\"><span class=\"code-snippet__meta\"># x_test, y_test \u5206\u522b\u662f10000\u4e2a\u6d4b\u8bd5\u56fe\u50cf\u548c\u7b54\u6848<\/span><\/span><\/code><code><span class=\"code-snippet_outer\"><span class=\"code-snippet__meta\"># \u8bad\u7ec3\u7684\u7b97\u662f\u65e5\u5e38\u4e60\u9898\uff0c\u6d4b\u8bd5\u7684\u624d\u662f\u9ad8\u8003\u9898\u3002\u4e3a\u4e86\u8ba1\u7b97\u673a\u9632\u6b62\u4f5c\u5f0a\uff0c\u8ba1\u7b97\u673a\u8bfb\u4e66\u7684\u65f6\u5019\u662f\u4e0d\u80fd\u770b\u5230\u9ad8\u8003\u8bd5\u5377\u7684<\/span><\/span><\/code><code><span class=\"code-snippet_outer\">(x_train, y_train), (x_test, y_test) = tf.keras.datasets.mnist.load_data()<\/span><\/code><code><span class=\"code-snippet_outer\"><\/span><\/code><code><span class=\"code-snippet_outer\"><span class=\"code-snippet__meta\">#\u00a0\u4e0b\u9762\u4e24\u53e5\u505a\u4e00\u4e2a\u975e\u5e38\u7b80\u5355\u7684\u9884\u5904\u7406\uff0c\u5c31\u662f\u5148\u628a\u00a060000\u00a0x\u00a028\u00a0x\u00a028\u00a0\u7684\u4e09\u7ef4\u6570\u636e\u6269\u5c55\u6210\u00a060000\u00a0x\u00a028\u00a0x\u00a028\u00a0x\u00a01\u7684\u56db\u7ef4\u6570\u636e<\/span><\/span><\/code><code><span class=\"code-snippet_outer\"><span class=\"code-snippet__meta\">#\u00a0\u8fd9\u4e2a\u5bf9\u4e8e\u6570\u636e\u6ca1\u6709\u4efb\u4f55\u53d8\u5316\uff0c\u53ea\u662f\u548cTensorflow\u8981\u6c42\u7684\u8f93\u5165\u4fdd\u6301\u4e00\u81f4\u800c\u5df2<\/span><\/span><\/code><code><span class=\"code-snippet_outer\"><span class=\"code-snippet__meta\">#\u00a0\u7136\u540e\u518d\u90fd\u9664\u4ee5\u00a0255\uff0c\u4f7f\u5f97\u6570\u636e\u90fd\u843d\u57280-1\u4e4b\u95f4\uff0c\u4ec5\u4ec5\u63d0\u9ad8\u4e9b\u8ba1\u7b97\u6548\u7387\uff0c\u4e0d\u600e\u4e48\u5f71\u54cd\u7ed3\u679c\u7684<\/span><\/span><\/code><code><span class=\"code-snippet_outer\">x_train = x_train.astype(np.float32).reshape(*x_train.shape, <span class=\"code-snippet__number\">1<\/span>)<\/span><\/code><code><span class=\"code-snippet_outer\">x_train\u00a0\/=\u00a0<span class=\"code-snippet__number\">255<\/span><\/span><\/code><code><span class=\"code-snippet_outer\"><\/span><\/code><code><span class=\"code-snippet_outer\"><span class=\"code-snippet__meta\">#\u00a0\u7fe0\u82b1\uff0c\u4e0a\u6a21\u578b\uff01\u5982\u4e0b\u5c31\u662f\u4f20\u8bf4\u4e2d\u7684\u5377\u79ef\u795e\u7ecf\u7f51\u7edc<\/span><\/span><\/code><code><span class=\"code-snippet_outer\">model = Sequential()<\/span><\/code><code><span class=\"code-snippet_outer\">model.<span class=\"code-snippet__keyword\">add<\/span>(Conv2D(<span class=\"code-snippet__number\">32<\/span>, kernel_size=(<span class=\"code-snippet__number\">3<\/span>, <span class=\"code-snippet__number\">3<\/span>), activation=<span class=\"code-snippet__string\">'relu'<\/span>, input_shape=(<span class=\"code-snippet__number\">28<\/span>, <span class=\"code-snippet__number\">28<\/span>, <span class=\"code-snippet__number\">1<\/span>)))<\/span><\/code><code><span class=\"code-snippet_outer\">model.<span class=\"code-snippet__keyword\">add<\/span>(MaxPooling2D(pool_size=(<span class=\"code-snippet__number\">2<\/span>, <span class=\"code-snippet__number\">2<\/span>)))<\/span><\/code><code><span class=\"code-snippet_outer\">model.<span class=\"code-snippet__keyword\">add<\/span>(Flatten())<\/span><\/code><code><span class=\"code-snippet_outer\">model.<span class=\"code-snippet__keyword\">add<\/span>(Dense(<span class=\"code-snippet__number\">10<\/span>, activation=<span class=\"code-snippet__string\">'softmax'<\/span>))<\/span><\/code><code><span class=\"code-snippet_outer\">model.compile(optimizer=<span class=\"code-snippet__string\">'adam'<\/span>, metrics=[<span class=\"code-snippet__string\">'accuracy'<\/span>], loss=<span class=\"code-snippet__string\">'sparse_categorical_crossentropy'<\/span>)<\/span><\/code><code><span class=\"code-snippet_outer\">model.fit(x_train,\u00a0y_train,\u00a0epochs=<span class=\"code-snippet__number\">10<\/span>)<\/span><\/code><\/pre>\n<\/section>\n<p style=\"white-space: normal;\">\u5927\u5bb6\u53ef\u4ee5\u5148\u770b\u770b\u8fd9\u4e2a Python \u4ee3\u7801\u3002\u8fd9\u662f\u4e00\u4e2a\u5f88\u7b80\u5355\u7684\u8bad\u7ec3\u5377\u79ef\u7f51\u7edc\u6a21\u578b\u7684\u4ee3\u7801\uff0c19\u884c\u4ee5\u524d\u7684\u90fd\u8fd8\u662f\u4e00\u4e9b\u9884\u5904\u7406\u7684\u4ee3\u7801\uff0c\u771f\u6b63\u6709\u7528\u7684\u5c31\u53ea\u670919\u523025\u884c\u8fd97\u884c\u4ee3\u7801\u3002\u800c\u8fd9\u4e2a\u4ee3\u7801\u5c31\u662f\u6784\u5efa\u4e86\u4e00\u4e2a\u4ec5\u4ec5\u56db\u5c42\u7684\u795e\u7ecf\u7f51\u7edc\uff0c\u5206\u522b\u662f<\/p>\n<ul class=\"list-paddingleft-2\" style=\"list-style-type: circle;\">\n<li>\n<p style=\"white-space: normal;\">\u7b2c\u4e00\u5c42\uff1a\u5377\u79ef\u5c42 Conv2D<\/p>\n<\/li>\n<li>\n<p style=\"white-space: normal;\">\u7b2c\u4e8c\u5c42\uff1a\u6c60\u5316\u5c42 MaxPooling<\/p>\n<\/li>\n<li>\n<p style=\"white-space: normal;\">\u7b2c\u4e09\u5c42\uff1a\u538b\u5e73\u5c42 Flatten<\/p>\n<\/li>\n<li>\n<p style=\"white-space: normal;\">\u7b2c\u56db\u5c42\uff1a\u5168\u8fde\u63a5\u5c42 Dense<\/p>\n<\/li>\n<\/ul>\n<p style=\"white-space: normal;\">\u4e5f\u4e0d\u8981\u88ab\u8fd9\u4e9b\u540d\u5b57\u5413\u5012\u3002\u6bcf\u4e00\u5c42\uff0c\u5373\u4fbf\u662f\u4e0d\u7528\u4efb\u4f55\u7c7b\uff0c\u4ec5\u4ec5\u662f\u7528\u539f\u751f\u7684Python\u4ee3\u7801\uff0c\u4e5f\u90fd\u662f\u51e0\u884c\u641e\u5b9a\u7684\u7b80\u5355\u7684\u8fd0\u7b97\u3002\u6240\u4ee5\u4eba\u5de5\u667a\u80fd\u7684\u4ee3\u7801\u662f\u5982\u6b64\u4e4b\u7b80\u6d01\u4f18\u7f8e\uff0c\u5374\u8fbe\u5230\u4e86\u4ee5\u524d\u590d\u6742\u7684\u7cfb\u7edf\u51e0\u5341\u5e74\u6ca1\u6709\u8fbe\u5230\u7684\u9ad8\u5ea6\u3002<\/p>\n<p style=\"white-space: normal;\">\u5982\u4e0a\u4ee3\u7801\u6267\u884c\u4e00\u904d\uff0c\u4e00\u4e24\u5206\u949f\u4ee5\u540e\uff0c\u5c31\u4f1a\u5f97\u5230\u4e00\u4e2a\u8bad\u7ec3\u597d\u7684\u6a21\u578b\uff0c\u6309\u7167\u8f93\u51fa\u6211\u4eec\u770b\u5230\uff0c\u5b83\u7684\u8bc6\u522b\u51c6\u786e\u7387\u9ad8\u8fbe 99.73%<\/p>\n<p style=\"font-stretch: normal;font-size: 18px;line-height: normal;font-family: Courier;color: rgb(59, 35, 34);background-color: rgb(215, 211, 183);\"><span style=\"font-variant-ligatures: no-common-ligatures;\"><\/span><\/p>\n<section class=\"code-snippet__fix code-snippet__js\">\n<ul class=\"code-snippet__line-index code-snippet__js\">\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<\/ul>\n<pre class=\"code-snippet__js\" data-lang=\"diff\"><code><span class=\"code-snippet_outer\">Epoch 10\/10<\/span><\/code><code><span class=\"code-snippet_outer\">1875\/1875 [<span class=\"code-snippet__comment\">==============================]<\/span><\/span><\/code><code><span class=\"code-snippet_outer\">\u00a0-\u00a010s\u00a05ms\/step\u00a0-\u00a0loss:\u00a00.0083\u00a0-\u00a0accuracy:\u00a00.9973<\/span><\/code><\/pre>\n<\/section>\n<p>\u6211\u4e0d\u5f97\u4e0d\u627f\u8ba4\uff0c\u8fd9\u4e2a\u8bc6\u522b\u7387\u5df2\u7ecf\u8fdc\u9ad8\u4e8e\u6211\u8fd9\u4e2a\u4eba\u8089\u795e\u7ecf\u7f51\u7edc\u4e86\uff0c\u6bd5\u7adf\u5f88\u591a\u7684\u6570\u5b57\u6211\u770b\u534a\u5929\u4e5f\u62ff\u4e0d\u51c6\u662f\u591a\u5c11\u3002\u00a0&#x1f600;\u00a0<\/p>\n<p>\u4e0b\u4e00\u7bc7\u5c31\u8981\u5f00\u59cb\u6b63\u513f\u516b\u7ecf\u7684\u89e3\u91ca\uff0c\u4eba\u5de5\u667a\u80fd\u7f51\u7edc\u7684\u5de5\u4f5c\u7ec6\u8282\u4e86\u3002\u540e\u53f0\u56de\u590d\u00a0\u201c AI \u201d \u53ef\u4ee5\u83b7\u5f97\u5168\u7cfb\u5217\u6587\u7ae0\u548c\u4ee3\u7801<\/p>\n<p>\u540e\u6ce8\uff1a\u4ee3\u7801\u4e00\uff1a\u753b\u51fa\u5305\u62ec\u7b2c 1001 \u4e2a\u56fe\u50cf\u7684 64 \u5404\u79cd 0 \u7684\u4ee3\u7801<\/p>\n<section class=\"code-snippet__fix code-snippet__js\">\n<ul class=\"code-snippet__line-index code-snippet__js\">\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<li><\/li>\n<\/ul>\n<pre class=\"code-snippet__js\" data-lang=\"python\"><code><span class=\"code-snippet_outer\"><span class=\"code-snippet__keyword\">import<\/span> tensorflow <span class=\"code-snippet__keyword\">as<\/span> tf<\/span><\/code><code><span class=\"code-snippet_outer\"><span class=\"code-snippet__comment\"># \u753b\u56fe\u5de5\u5177\uff0c\u4eba\u5de5\u667a\u80fd\u5b66\u4e60\u5fc5\u5907<\/span><\/span><\/code><code><span class=\"code-snippet_outer\"><span class=\"code-snippet__keyword\">import<\/span> matplotlib.pyplot <span class=\"code-snippet__keyword\">as<\/span> plt<\/span><\/code><code><span class=\"code-snippet_outer\"><span class=\"code-snippet__comment\"># \u4ece\u4e92\u8054\u7f51\u4e0a\u4e0b\u8f7d mnist \u6570\u636e\u96c6\u5230\u672c\u5730 ~\/.keras\/datasets\/mnist.npz<\/span><\/span><\/code><code><span class=\"code-snippet_outer\"><span class=\"code-snippet__comment\"># x_train, y_train \u5206\u522b\u662f\u662f60000\u4e2a\u8bad\u7ec3\u56fe\u50cf\u548c\u7b54\u6848<\/span><\/span><\/code><code><span class=\"code-snippet_outer\"><span class=\"code-snippet__comment\"># x_test, y_test \u5206\u522b\u662f10000\u4e2a\u6d4b\u8bd5\u56fe\u50cf\u548c\u7b54\u6848<\/span><\/span><\/code><code><span class=\"code-snippet_outer\"><span class=\"code-snippet__comment\"># \u8bad\u7ec3\u7684\u7b97\u662f\u65e5\u5e38\u4e60\u9898\uff0c\u6d4b\u8bd5\u7684\u624d\u662f\u9ad8\u8003\u9898\u3002\u4e3a\u4e86\u8ba1\u7b97\u673a\u9632\u6b62\u4f5c\u5f0a\uff0c\u8ba1\u7b97\u673a\u8bfb\u4e66\u7684\u65f6\u5019\u662f\u4e0d\u80fd\u770b\u5230\u9ad8\u8003\u8bd5\u5377\u7684<\/span><\/span><\/code><code><span class=\"code-snippet_outer\">(x_train, y_train), (x_test, y_test) = tf.keras.datasets.mnist.load_data()<\/span><\/code><code><span class=\"code-snippet_outer\"><\/span><\/code><code><span class=\"code-snippet_outer\"><span class=\"code-snippet__comment\"># \u51c6\u5907\u53d6\u51fa\u676564\u4e2a\u56fe\u50cf\u753b\u51fa\u6765\u770b\u770b<\/span><\/span><\/code><code><span class=\"code-snippet_outer\">c = <span class=\"code-snippet__number\">0<\/span><\/span><\/code><code><span class=\"code-snippet_outer\"><span class=\"code-snippet__keyword\">for<\/span> index <span class=\"code-snippet__keyword\">in<\/span> range(<span class=\"code-snippet__number\">850<\/span>, <span class=\"code-snippet__number\">2000<\/span>):<\/span><\/code><code><span class=\"code-snippet_outer\">    <span class=\"code-snippet__comment\"># \u4e0d\u662f 0 \u5c31\u7b97\u4e86<\/span><\/span><\/code><code><span class=\"code-snippet_outer\">    <span class=\"code-snippet__keyword\">if<\/span> y_train[index] != <span class=\"code-snippet__number\">0<\/span>:<\/span><\/code><code><span class=\"code-snippet_outer\">        <span class=\"code-snippet__keyword\">continue<\/span><\/span><\/code><code><span class=\"code-snippet_outer\"><\/span><\/code><code><span class=\"code-snippet_outer\">    c += <span class=\"code-snippet__number\">1<\/span><\/span><\/code><code><span class=\"code-snippet_outer\">    <span class=\"code-snippet__keyword\">if<\/span> c &gt; <span class=\"code-snippet__number\">64<\/span>:<\/span><\/code><code><span class=\"code-snippet_outer\">        <span class=\"code-snippet__keyword\">continue<\/span><\/span><\/code><code><span class=\"code-snippet_outer\"><\/span><\/code><code><span class=\"code-snippet_outer\">    <span class=\"code-snippet__comment\"># \u5bfc\u5165\u7ed8\u56fe\u5de5\u5177 matplotlib\u3002\u5728\u547d\u4ee4\u884c\u4e0b\u7528 pip3 install matplotlib \u5b89\u88c5<\/span><\/span><\/code><code><span class=\"code-snippet_outer\">    <span class=\"code-snippet__comment\"># \u628a\u8fd9\u4e2a28x28\u7684\u77e9\u9635\u76f4\u63a5\u7ed9 Matplot \u753b\u7070\u5ea6\u56fe<\/span><\/span><\/code><code><span class=\"code-snippet_outer\">    ax = plt.subplot(<span class=\"code-snippet__number\">8<\/span>, <span class=\"code-snippet__number\">8<\/span>, c)<\/span><\/code><code><span class=\"code-snippet_outer\">    ax.set_xticks([])<\/span><\/code><code><span class=\"code-snippet_outer\">    ax.set_yticks([])<\/span><\/code><code><span class=\"code-snippet_outer\">    plt.imshow(x_train[index], cmap=<span class=\"code-snippet__string\">\"gray\"<\/span>)<\/span><\/code><code><span class=\"code-snippet_outer\"><span class=\"code-snippet__comment\"># \u663e\u793a\u51fa\u6765<\/span><\/span><\/code><code><span class=\"code-snippet_outer\">plt.show()<\/span><\/code><\/pre>\n<\/section>\n<\/div>\n","protected":false},"excerpt":{"rendered":"<p>\u9898\u56fe\uff1a\u9999\u6e2f\u7684\u8717\u5c45 \u7ed8\u4e8e\uff1a2020\u5e743\u6708 \u4e0a\u4e00\u7bc7\uff1a\u5377\u79ef\u7f51\u7edc CNN \u5b66\u4e60\u7b14\u8bb0\u4e4b\u4e00\uff1a\u6211\u4eec\u662f\u600e\u4e48\u8ba4\u8bc60\u7684 \u6628\u5929\u7559\u4e86\u4e2a &hellip; <a href=\"https:\/\/home.wangjianshuo.com\/cn\/20200903_%e5%8d%b7%e7%a7%af%e7%bd%91%e7%bb%9c-cnn-%e5%ad%a6%e4%b9%a0%e7%ac%94%e8%ae%b0%e4%b9%8b%e4%ba%8c%ef%bc%9a%e7%bb%86%e8%8a%82%e6%b6%88%e5%a4%b1%ef%bc%8c%e6%8a%bd%e8%b1%a1%e7%94%9f%e6%88%90.htm\" class=\"more-link\">\u7ee7\u7eed\u9605\u8bfb<span class=\"screen-reader-text\">\u5377\u79ef\u7f51\u7edc CNN \u5b66\u4e60\u7b14\u8bb0\u4e4b\u4e8c\uff1a\u7ec6\u8282\u6d88\u5931\uff0c\u62bd\u8c61\u751f\u6210<\/span> <span class=\"meta-nav\">&rarr;<\/span><\/a><\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[1],"tags":[],"_links":{"self":[{"href":"https:\/\/home.wangjianshuo.com\/cn\/wp-json\/wp\/v2\/posts\/1498"}],"collection":[{"href":"https:\/\/home.wangjianshuo.com\/cn\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/home.wangjianshuo.com\/cn\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/home.wangjianshuo.com\/cn\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/home.wangjianshuo.com\/cn\/wp-json\/wp\/v2\/comments?post=1498"}],"version-history":[{"count":1,"href":"https:\/\/home.wangjianshuo.com\/cn\/wp-json\/wp\/v2\/posts\/1498\/revisions"}],"predecessor-version":[{"id":1499,"href":"https:\/\/home.wangjianshuo.com\/cn\/wp-json\/wp\/v2\/posts\/1498\/revisions\/1499"}],"wp:attachment":[{"href":"https:\/\/home.wangjianshuo.com\/cn\/wp-json\/wp\/v2\/media?parent=1498"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/home.wangjianshuo.com\/cn\/wp-json\/wp\/v2\/categories?post=1498"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/home.wangjianshuo.com\/cn\/wp-json\/wp\/v2\/tags?post=1498"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}