{"id":78647,"date":"2026-02-27T22:32:13","date_gmt":"2026-02-27T14:32:13","guid":{"rendered":"https:\/\/www.wsisp.com\/helps\/78647.html"},"modified":"2026-02-27T22:32:13","modified_gmt":"2026-02-27T14:32:13","slug":"%e6%9c%ba%e5%99%a8%e5%ad%a6%e4%b9%a0%e4%b9%8b%e9%80%bb%e8%be%91%e5%9b%9e%e5%bd%92","status":"publish","type":"post","link":"https:\/\/www.wsisp.com\/helps\/78647.html","title":{"rendered":"\u673a\u5668\u5b66\u4e60\u4e4b\u903b\u8f91\u56de\u5f52"},"content":{"rendered":"<h3><\/h3>\n<h3>\u6982\u8981\u4ecb\u7ecd<\/h3>\n<p><img loading=\"lazy\" decoding=\"async\" alt=\"\" height=\"1136\" src=\"https:\/\/www.wsisp.com\/helps\/wp-content\/uploads\/2026\/02\/20260227143209-69a1aae98600a.png\" width=\"1641\" \/><\/p>\n<p>\u00a0\u00a0\u00a0\u00a0<span style=\"color:#ff9900\">\u00a0\u00a0<\/span><span style=\"color:#ad720d\">\u00a0\u00a0\u5bf9\u4e8e\u903b\u8f91\u56de\u5f52\u6211\u4eec\u53ef\u4ee5\u4ece\u4e00\u4e2a\u95ee\u9898\u5165\u624b,\u5230\u5e95\u4ec0\u4e48\u662f\u903b\u8f91\u56de\u5f52\u7684\u7b97\u6cd5\u601d\u60f3?<\/span><\/p>\n<p>\u00a0 \u00a0 \u00a0 \u00a0\u9762\u8bd5\u5b98\u95ee\u201c\u8bf7\u63cf\u8ff0\u903b\u8f91\u56de\u5f52\u7684\u7b97\u6cd5\u601d\u60f3\u201d\u8fd9\u4e2a\u95ee\u9898\u65f6&#xff0c;\u5176\u5b9e\u662f\u5728\u8003\u5bdf\u4f60\u5bf9\u8fd9\u4e2a\u57fa\u7840\u6a21\u578b\u7684<br \/>\n\u672c\u8d28\u7406\u89e3\u2014\u2014\u662f\u53ea\u4f1a\u8c03\u7528sklearn.learn.LogisticRegression&#xff0c;\u8fd8\u662f\u771f\u6b63\u660e\u767d\u5b83\u4e3a\u4ec0\u4e48\u53eb\u201c\u56de\u5f52\u201d\u5374\u505a\u5206\u7c7b&#xff0c;\u5b83\u7684\u6838\u5fc3\u5728\u7b97\u4ec0\u4e48\u3002<\/p>\n<p>\u00a0 \u00a0 \u00a0 \u201c\u903b\u8f91\u56de\u5f52\u867d\u7136\u540d\u5b57\u91cc\u6709\u2018\u56de\u5f52\u2019&#xff0c;\u4f46\u5b83\u5b9e\u9645\u4e0a\u662f\u4e00\u79cd\u7528\u4e8e\u89e3\u51b3\u4e8c\u5206\u7c7b\u95ee\u9898\u7684\u7ebf\u6027\u6a21\u578b\u3002\u5b83\u7684\u6838\u5fc3\u601d\u60f3\u662f&#xff1a;\u5148\u62df\u5408\u51b3\u7b56\u8fb9\u754c&#xff08;\u7ebf\u6027\u56de\u5f52\u7684\u6d3b\u513f&#xff09;&#xff0c;\u518d\u628a\u7ebf\u6027\u8f93\u51fa\u6620\u5c04\u52300\u52301\u4e4b\u95f4\u7684\u6982\u7387(Sigmoid\u51fd\u6570\u7684\u6d3b\u513f&#xff09;&#xff0c;\u6700\u540e\u6839\u636e\u6982\u7387\u8fdb\u884c\u5206\u7c7b\u3002\u201d<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" alt=\"\" height=\"258\" src=\"https:\/\/www.wsisp.com\/helps\/wp-content\/uploads\/2026\/02\/20260227143210-69a1aaeaa90d9.png\" width=\"772\" \/><\/p>\n<p><img loading=\"lazy\" decoding=\"async\" alt=\"\" height=\"366\" src=\"https:\/\/www.wsisp.com\/helps\/wp-content\/uploads\/2026\/02\/20260227143210-69a1aaeac93c1.png\" width=\"769\" \/><\/p>\n<p><span style=\"background-color:#ffd900\">\u6240\u4ee5&#xff0c;\u903b\u8f91\u56de\u5f52\u7684\u6838\u5fc3\u601d\u60f3\u53ef\u4ee5\u6982\u62ec\u4e3a&#xff1a;\u7ebf\u6027\u56de\u5f52 &#043; Sigmoid\u8f6c\u6362 &#043; \u6700\u5927\u4f3c\u7136\u4f30\u8ba1\u3002\u5b83\u7b80\u5355\u3001\u53ef\u89e3\u91ca\u6027\u5f3a\u3001\u8bad\u7ec3\u5feb&#xff0c;\u662f\u5f88\u591a\u590d\u6742\u6a21\u578b&#xff08;\u5982\u795e\u7ecf\u7f51\u7edc\u3001\u63a8\u8350\u7cfb\u7edf&#xff09;\u7684\u57fa\u7840\u7ec4\u4ef6\u3002<\/span><\/p>\n<p>\u903b\u8f91\u56de\u5f52\u6a21\u578b\u4ecb\u7ecd:<br \/>\n    \u6982\u8ff0:<br \/>\n        \u5c5e\u4e8e\u6709\u76d1\u7763\u5b66\u4e60, \u5373: \u6709\u7279\u5f81, \u6709\u6807\u7b7e, \u4e14\u6807\u7b7e\u662f\u79bb\u6563\u7684.<br \/>\n        \u4e3b\u8981\u9002\u7528\u4e8e: \u4e8c\u5206\u7c7b,\u662f\u5206\u7c7b\u7b97\u6cd5\u7684\u4e00\u79cd.<br \/>\n    \u539f\u7406:<br \/>\n        \u628a\u7ebf\u6027\u56de\u5f52\u5904\u7406\u540e\u7684\u9884\u6d4b\u503c -&gt; \u901a\u8fc7 Sigmoid\u6fc0\u6d3b\u51fd\u6570, \u6620\u5c04\u5230[0, 1] \u6982\u7387 -&gt; \u57fa\u4e8e\u81ea\u5b9a\u4e49\u7684\u9608\u503c, \u7ed3\u5408\u6982\u7387\u6765 \u5206\u7c7b.<br \/>\n\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a01. \u57fa\u4e8e\u7ebf\u6027\u56de\u5f52, \u7ed3\u5408\u7279\u5f81\u503c, \u8ba1\u7b97\u51fa\u6807\u7b7e\u503c.<br \/>\n\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a02. \u628a\u4e0a\u8ff0\u7b97\u51fa\u6765\u7684\u6807\u7b7e\u503c\u4f20\u7ed9 \u6fc0\u6d3b\u51fd\u6570(Sigmoid), \u6620\u5c04\u6210 [0, 1]\u533a\u95f4\u7684\u503c.<br \/>\n\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a03. \u7ed3\u5408\u624b\u52a8\u8bbe\u7f6e\u7684\u9608\u503c, \u6765\u5212\u5206\u533a\u95f4\u5373\u53ef.<br \/>\n    \u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u4f8b\u5982: \u9608\u503c &#061; 0.6, \u5219:<br \/>\n        \u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u7ed3\u679c &gt; 0.6        A\u7c7b<br \/>\n        \u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u5426\u5219              B\u7c7b<br \/>\n    \u635f\u5931\u51fd\u6570:<br \/>\n        \u6781\u5927\u4f3c\u7136\u4f30\u8ba1\u51fd\u6570\u7684 \u8d1f\u6570\u5f62\u5f0f,\u5148\u57fa\u4e8e \u6781\u5927\u4f3c\u7136\u51fd\u6570\u8ba1\u7b97, \u7136\u540e\u8f6c\u6210 \u5bf9\u6570\u4f3c\u7136\u51fd\u6570, \u7ed3\u5408\u68af\u5ea6\u4e0b\u964d, \u8ba1\u7b97\u6700\u5c0f\u503c\u5373\u53ef.<br \/>\n\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u603b\u7ed3:<br \/>\n    \u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a01. \u903b\u8f91\u56de\u5f52\u539f\u7406: \u628a\u7ebf\u6027\u56de\u5f52\u7684\u8f93\u51fa, \u4f5c\u4e3a\u903b\u8f91\u56de\u5f52\u7684\u8f93\u5165.<br \/>\n   \u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 2. \u9ed8\u8ba4\u60c5\u51b5\u4e0b: \u91c7\u7528\u6837\u672c\u5c11\u7684\u5f53\u505a\u6b63\u4f8b, \u5176\u5b83\u662f\u53cd\u4f8b(\u4e5f\u53eb: \u5047\u4f8b)<br \/>\n   \u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 3. (\u903b\u8f91\u56de\u5f52)\u635f\u5931\u51fd\u6570\u7684\u8bbe\u8ba1\u539f\u5219: \u771f\u5b9e\u4f8b\u5b50\u662f\u6b63\u4f8b\u7684\u60c5\u51b5\u4e0b, \u6982\u7387\u503c\u8d8a\u5927\u8d8a\u597d.<br \/>\n\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0<br \/>\n\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u7406\u89e3\u5206\u7c7b\u8bc4\u4f30\u65b9\u6cd5\u5e76\u8fdb\u884c\u8be6\u7ec6\u7684\u63cf\u8ff0:<br \/>\n\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u51c6\u786e\u7387:\u6240\u6709\u6837\u672c\u4e2d\u9884\u6d4b\u6b63\u786e\u7684\u6837\u672c\u6bd4\u4f8b(\u5305\u62ec\u6b63\u4f8b\u548c\u53cd\u4f8b)<br \/>\n\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u7cbe\u786e\u7387:\u9884\u6d4b\u4e3a\u6b63\u4f8b\u6837\u672c\u4e2d\u771f\u6b63\u4f8b\u6837\u672c\u7684\u6bd4\u4f8b,\u67e5\u51c6\u7387<br \/>\n\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u53ec\u56de\u7387:\u771f\u5b9e\u4e3a\u6b63\u4f8b\u7684\u6837\u672c\u4e2d,\u9884\u6d4b\u4e3a\u6b63\u4f8b\u6837\u672c\u7684\u6bd4\u4f8b,\u67e5\u5168\u7387<br \/>\n\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0f1-score:\u7cbe\u786e\u7387\u548c\u53ec\u56de\u7387\u7684\u7ec4\u5408<br \/>\n\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0ROC\u66f2\u7ebf\u548cAUC\u6307\u6807:<br \/>\n\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0ROC\u662f\u4ee5FPR(FP\/ALL_\u53cd\u4f8b)\u548cTPR(TP\/ALL_\u6b63\u4f8b)\u7ed8\u5236\u7684\u6a21\u578b\u8bc4\u4f30\u66f2\u7ebf<br \/>\n\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0AUC\u662fROC\u66f2\u7ebf\u4e0b\u9762\u79ef,\u53d6\u503c\u57280-1\u4e4b\u95f4,\u4e00\u822c\u662f\u5927\u4e8e0.5\u7684,<br \/>\n\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u8868\u793a\u6a21\u578b\u7684\u8bc4\u4f30\u80fd\u529b\u5982\u4f55,\u8d8a\u63a5\u8fd11\u8d8a\u4f18\u79c0.<br \/>\n\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0<\/p>\n<p>\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u673a\u5668\u5b66\u4e60\u9879\u76ee\u6d41\u7a0b<br \/>\n    \u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a01. \u51c6\u5907\u6570\u636e.<br \/>\n\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a02. \u6570\u636e\u7684\u9884\u5904\u7406.<br \/>\n\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a03. \u7279\u5f81\u5de5\u7a0b.<br \/>\n              \u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u7279\u5f81\u63d0\u53d6, \u7279\u5f81\u9884\u5904\u7406, \u7279\u5f81\u964d\u7ef4, \u7279\u5f81\u9009\u53d6, \u7279\u5f81\u7ec4\u5408<br \/>\n\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a04. \u6a21\u578b\u8bad\u7ec3.<br \/>\n\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0    5. \u6a21\u578b\u9884\u6d4b.<br \/>\n\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0    6. \u6a21\u578b\u8bc4\u4f30.<\/p>\n<h3>\u5165\u95e8\u6848\u4f8b<\/h3>\n<p>&#034;&#034;&#034;<br \/>\n\u6848\u4f8b:<br \/>\n    \u764c\u75c7\u9884\u6d4b\u6848\u4f8b, \u76ee\u7684: \u6f14\u793a\u903b\u8f91\u56de\u5f52\u76f8\u5173API.<\/p>\n<p>\u903b\u8f91\u56de\u5f52:<br \/>\n    \u6982\u8ff0:<br \/>\n        \u5b83\u5c5e\u4e8e\u5206\u7c7b\u7b97\u6cd5\u7684\u4e00\u79cd, \u4e00\u822c\u7528\u4e8e: \u4e8c\u5206\u6cd5.<br \/>\n    \u539f\u7406:<br \/>\n        1. \u57fa\u4e8e\u7ebf\u6027\u56de\u5f52, \u7ed3\u5408\u7279\u5f81\u503c, \u8ba1\u7b97\u51fa\u6807\u7b7e\u503c.<br \/>\n        2. \u628a\u4e0a\u8ff0\u7b97\u51fa\u6765\u7684\u6807\u7b7e\u503c\u4f20\u7ed9 \u6fc0\u6d3b\u51fd\u6570(Sigmoid), \u6620\u5c04\u6210 [0, 1]\u533a\u95f4\u7684\u503c.<br \/>\n        3. \u7ed3\u5408\u624b\u52a8\u8bbe\u7f6e\u7684\u9608\u503c, \u6765\u5212\u5206\u533a\u95f4\u5373\u53ef.<br \/>\n            \u4f8b\u5982: \u9608\u503c &#061; 0.6, \u5219:<br \/>\n                \u7ed3\u679c &gt; 0.6        A\u7c7b<br \/>\n                \u5426\u5219              B\u7c7b<br \/>\n    \u635f\u5931\u51fd\u6570:<br \/>\n        \u5148\u57fa\u4e8e \u6781\u5927\u4f3c\u7136\u51fd\u6570\u8ba1\u7b97, \u7136\u540e\u8f6c\u6210 \u5bf9\u6570\u4f3c\u7136\u51fd\u6570, \u7ed3\u5408\u68af\u5ea6\u4e0b\u964d, \u8ba1\u7b97\u6700\u5c0f\u503c\u5373\u53ef.<\/p>\n<p>    \u603b\u7ed3:<br \/>\n        1. \u903b\u8f91\u56de\u5f52\u539f\u7406: \u628a\u7ebf\u6027\u56de\u5f52\u7684\u8f93\u51fa, \u4f5c\u4e3a\u903b\u8f91\u56de\u5f52\u7684\u8f93\u5165.<br \/>\n        2. \u9ed8\u8ba4\u60c5\u51b5\u4e0b: \u91c7\u7528\u6837\u672c\u5c11\u7684\u5f53\u505a\u6b63\u4f8b, \u5176\u5b83\u662f\u53cd\u4f8b(\u4e5f\u53eb: \u5047\u4f8b)<br \/>\n        3. (\u903b\u8f91\u56de\u5f52)\u635f\u5931\u51fd\u6570\u7684\u8bbe\u8ba1\u539f\u5219: \u771f\u5b9e\u4f8b\u5b50\u662f\u6b63\u4f8b\u7684\u60c5\u51b5\u4e0b, \u6982\u7387\u503c\u8d8a\u5927\u8d8a\u597d.<\/p>\n<p>\u56de\u987e: \u673a\u5668\u5b66\u4e60\u7684\u5f00\u53d1\u6d41\u7a0b<br \/>\n    1. \u51c6\u5907\u6570\u636e.<br \/>\n    2. \u6570\u636e\u7684\u9884\u5904\u7406.<br \/>\n    3. \u7279\u5f81\u5de5\u7a0b.<br \/>\n        \u7279\u5f81\u63d0\u53d6, \u7279\u5f81\u9884\u5904\u7406, \u7279\u5f81\u964d\u7ef4, \u7279\u5f81\u9009\u53d6, \u7279\u5f81\u7ec4\u5408<br \/>\n    4. \u6a21\u578b\u8bad\u7ec3.<br \/>\n    5. \u6a21\u578b\u9884\u6d4b.<br \/>\n    6. \u6a21\u578b\u8bc4\u4f30.<br \/>\n&#034;&#034;&#034;<\/p>\n<p># \u5bfc\u5305<br \/>\nimport pandas as pd<br \/>\nimport numpy as np<br \/>\nfrom sklearn.linear_model import LogisticRegression<br \/>\nfrom sklearn.metrics import accuracy_score<br \/>\nfrom sklearn.model_selection import train_test_split<br \/>\nfrom sklearn.preprocessing import StandardScaler<\/p>\n<p># 1. \u51c6\u5907\u6570\u636e.<br \/>\ndata &#061; pd.read_csv(&#039;.\/data\/breast-cancer-wisconsin.csv&#039;)<br \/>\ndata.info()     # 699\u884c * 11\u5217, \u770b\u4e0d\u5230\u7a7a\u503c, \u56e0\u4e3a\u6709?\u6807\u8bb0.<\/p>\n<p># 2. \u6570\u636e\u7684\u9884\u5904\u7406.<br \/>\n# 2.1 \u7528 np.NaN\u6765\u66ff\u6362?<br \/>\ndata &#061; data.replace(&#039;?&#039;, np.nan)<br \/>\ndata.info()<\/p>\n<p># 2.2 \u56e0\u4e3a\u6709\u7f3a\u5931\u503c, \u4f46\u662f\u7f3a\u5931\u503c\u4e0d\u591a, \u6211\u4eec\u5220\u9664\u5373\u53ef.  \u6309\u884c\u5220\u9664.<br \/>\ndata.dropna(axis&#061;0, inplace&#061;True)   # axis&#061;0(\u9ed8\u8ba4), \u6309\u884c\u5220.<br \/>\ndata.info()<\/p>\n<p># 3. \u7279\u5f81\u5de5\u7a0b, \u7279\u5f81\u63d0\u53d6, \u7279\u5f81\u9884\u5904\u7406, \u7279\u5f81\u964d\u7ef4, \u7279\u5f81\u9009\u53d6, \u7279\u5f81\u7ec4\u5408<br \/>\n# 3.1 \u83b7\u53d6\u7279\u5f81\u503c \u548c \u76ee\u6807\u503c(\u6807\u7b7e\u503c).<br \/>\nx &#061; data.iloc[:, 1:-1]      # \u4ece\u7d22\u5f15\u4e3a1\u7684\u5217\u5f00\u59cb\u83b7\u53d6, \u76f4\u81f3 \u6700\u540e\u4e00\u5217(\u4e0d\u5305\u62ec).<br \/>\n# y &#061; data.iloc[:, -1]<br \/>\n# y &#061; data[&#039;Class&#039;]<br \/>\ny &#061; data.Class<\/p>\n<p># 3.2 \u67e5\u770b\u7ed3\u679c.<br \/>\nprint(len(x), len(y))<br \/>\nprint(x.head(10))<br \/>\nprint(y.head(10))<\/p>\n<p># 3.3 \u62c6\u5206\u8bad\u7ec3\u96c6 \u548c \u6d4b\u8bd5\u96c6.<br \/>\nx_train, x_test, y_train, y_test &#061; train_test_split(x, y, test_size&#061;0.2, random_state&#061;22)<\/p>\n<p># 3.4 \u6570\u636e\u96c6\u76f8\u5dee\u4e0d\u5927, \u53ef\u4ee5\u4e0d\u505a \u6807\u51c6\u5316\u5904\u7406, \u4f46\u662f\u4e3a\u4e86\u8ba9\u6b65\u9aa4\u66f4\u5b8c\u6574, \u6211\u4eec\u8fd8\u662f\u505a\u4e00\u4e0b.<br \/>\ntransfer &#061; StandardScaler()<br \/>\nx_train &#061; transfer.fit_transform(x_train)<br \/>\nx_test &#061; transfer.transform(x_test)<\/p>\n<p># 4. \u6a21\u578b\u8bad\u7ec3.<br \/>\n# 4.1 \u521b\u5efa\u6a21\u578b, \u903b\u8f91\u56de\u5f52\u6a21\u578b.<br \/>\nestimator &#061; LogisticRegression()<br \/>\n# 4.2 \u8bad\u7ec3\u6a21\u578b.<br \/>\nestimator.fit(x_train, y_train)<\/p>\n<p># 5. \u6a21\u578b\u9884\u6d4b.<br \/>\ny_predict &#061; estimator.predict(x_test)<br \/>\nprint(f&#039;\u9884\u6d4b\u503c: {y_predict}&#039;)<\/p>\n<p># 6. \u6a21\u578b\u8bc4\u4f30.<br \/>\nprint(f&#039;\u51c6\u786e\u7387: {estimator.score(x_test, y_test)}&#039;)    # 0.9854014598540146<br \/>\nprint(f&#039;\u51c6\u786e\u7387: {accuracy_score(y_test, y_predict)}&#039;)  # 0.9854014598540146<\/p>\n<p># \u81f3\u6b64, \u903b\u8f91\u56de\u5f52\u7684\u5165\u95e8API\u4ee3\u7801\u6211\u4eec\u5c31\u5199\u5b8c\u4e86, \u4f46\u662f\u6211\u4eec\u8fd9\u91cc\u505a\u7684\u662f\u764c\u75c7\u9884\u6d4b, \u601d\u8003: \u4ec5\u4ec5\u9760\u6b63\u786e\u7387, \u80fd\u8861\u91cf\u903b\u8f91\u56de\u5f52\u7ed3\u679c\u5417?<br \/>\n# \u80af\u5b9a\u662f\u4e0d\u53ef\u4ee5\u7684, \u56e0\u4e3a\u53ea\u77e5\u9053\u6b63\u786e\u7387, \u4e0d\u77e5\u9053\u5230\u5e95\u54ea\u4e9b\u662f\u9884\u6d4b\u6210\u529f\u4e86, \u54ea\u4e9b\u662f\u9884\u6d4b\u5931\u8d25\u4e86, \u6240\u4ee5\u4e3a\u4e86\u8fdb\u4e00\u6b65\u7684\u8bc4\u4f30, \u6211\u4eec\u9700\u8981\u52a0\u5165:<br \/>\n# \u6df7\u6dc6\u77e9\u9635, \u7cbe\u786e\u7387(\u638c\u63e1), \u53ec\u56de\u7387(\u638c\u63e1), F1\u503c(F1-score)(\u638c\u63e1),    ROC\u66f2\u7ebf(\u4e86\u89e3), AUC\u503c(\u4e86\u89e3).<\/p>\n<h3>\u6df7\u6dc6\u77e9\u9635-\u7cbe\u786e\u7387-\u53ec\u56de\u7387<\/h3>\n<p>&#034;&#034;&#034;<br \/>\n\u6848\u4f8b:<br \/>\n    \u6f14\u793a\u6df7\u6dc6\u77e9\u9635 \u548c \u7cbe\u786e\u7387, \u53ec\u56de\u7387, F1\u503c.<\/p>\n<p>\u56de\u987e: \u903b\u8f91\u56de\u5f52<br \/>\n    \u6982\u8ff0:<br \/>\n        \u5c5e\u4e8e\u6709\u76d1\u7763\u5b66\u4e60, \u5373: \u6709\u7279\u5f81, \u6709\u6807\u7b7e, \u4e14\u6807\u7b7e\u662f\u79bb\u6563\u7684.<br \/>\n        \u9002\u7528\u4e8e \u4e8c\u5206\u7c7b.<br \/>\n    \u8bc4\u4f30:<br \/>\n        \u7cbe\u786e\u7387, \u53ec\u56de\u7387, F1\u503c<\/p>\n<p>\u6df7\u6dc6\u77e9\u9635:<br \/>\n    \u6982\u8ff0:<br \/>\n        \u7528\u6765\u63cf\u8ff0 \u771f\u5b9e\u503c  \u548c \u9884\u6d4b\u503c\u4e4b\u95f4\u5173\u7cfb\u7684.<br \/>\n    \u56fe\u89e3:<br \/>\n                        \u9884\u6d4b\u6807\u7b7e(\u6b63\u4f8b)        \u9884\u6d4b\u6807\u7b7e(\u53cd\u4f8b)<br \/>\n        \u771f\u5b9e\u6807\u7b7e(\u6b63\u4f8b)      \u771f\u6b63\u4f8b(TP)           \u4f2a\u53cd\u4f8b(FN)<br \/>\n        \u771f\u5b9e\u6807\u7b7e(\u53cd\u4f8b)      \u4f2a\u6b63\u4f8b(FP)           \u771f\u53cd\u4f8b(TN)<br \/>\n    \u5355\u8bcd:<br \/>\n        True: \u771f,  False: \u5047(\u4f2a)<br \/>\n        Positive: \u6b63\u4f8b<br \/>\n        Negative: \u53cd\u4f8b<\/p>\n<p>    \u7ed3\u8bba:<br \/>\n        1. \u6a21\u62df\u4f7f\u7528 \u5206\u7c7b\u5c11\u7684 \u5145\u5f53 \u6b63\u4f8b.<br \/>\n        2. \u7cbe\u786e\u7387 &#061; \u771f\u6b63\u4f8b \u5728 \u9884\u6d4b\u6b63\u4f8b\u4e2d\u7684\u5360\u6bd4, \u5373: tp \/ (tp &#043; fp)<br \/>\n        3. \u53ec\u56de\u7387 &#061; \u771f\u6b63\u4f8b \u5728 \u771f\u6b63\u4f8b\u4e2d\u7684\u5360\u6bd4, \u5373: tp \/ (tp &#043; fn)<br \/>\n        4. F1\u503c &#061; 2 * (\u7cbe\u786e\u7387 * \u53ec\u56de\u7387) \/ (\u7cbe\u786e\u7387 &#043; \u53ec\u56de\u7387)<\/p>\n<p>    \u903b\u8f91\u56de\u5f52 \u8bc4\u4f30\u65b9\u5f0f:<br \/>\n    \u51c6\u786e\u7387:<br \/>\n        \u9884\u6d4b\u6b63\u786e\u7684 \/ \u6837\u672c\u603b\u6570, \u5373:  (tp &#043; tn) \/ \u6837\u672c\u603b\u6570<\/p>\n<p>    \u7cbe\u786e\u7387(\u67e5\u51c6\u7387, Precision):<br \/>\n        \u771f\u6b63\u4f8b \/ (\u771f\u6b63\u4f8b &#043; \u4f2a\u6b63\u4f8b), \u5373: tp \/ (tp &#043; fp)<br \/>\n        \u5927\u767d\u8bdd: \u771f\u6b63\u4f8b \u5728 \u9884\u6d4b\u4e3a\u6b63\u4f8b\u7684\u7ed3\u679c\u4e2d\u7684 \u5360\u6bd4.<\/p>\n<p>    \u53ec\u56de\u7387(\u67e5\u5168\u7387, Recall):<br \/>\n        \u771f\u6b63\u4f8b \/ (\u771f\u6b63\u4f8b &#043; \u4f2a\u53cd\u4f8b), \u5373: tp \/ (tp &#043; fn)<br \/>\n        \u5927\u767d\u8bdd: \u771f\u6b63\u4f8b \u5728 \u771f\u5b9e\u6b63\u4f8b\u6837\u672c\u4e2d\u7684 \u5360\u6bd4.<\/p>\n<p>    F1\u503c(F1-Score):<br \/>\n        2 * \u7cbe\u786e\u7387 * \u53ec\u56de\u7387 \/ (\u7cbe\u786e\u7387 &#043; \u53ec\u56de\u7387)<br \/>\n        \u9002\u7528\u4e8e: \u65e2\u8981\u8003\u8651\u7cbe\u786e\u7387, \u8fd8\u8981\u8003\u8651\u53ec\u56de\u7387\u7684\u60c5\u51b5.<br \/>\n&#034;&#034;&#034;<\/p>\n<p># \u5bfc\u5305<br \/>\nimport pandas as pd<br \/>\nfrom sklearn.metrics import confusion_matrix, precision_score, recall_score, f1_score    # \u6df7\u6dc6\u77e9\u9635, \u7cbe\u786e\u7387, \u53ec\u56de\u7387, F1\u503c<\/p>\n<p># \u9700\u6c42: \u5df2\u77e5\u670910\u4e2a\u6837\u672c, 6\u4e2a\u6076\u6027\u80bf\u7624(\u6b63\u4f8b), 4\u4e2a\u826f\u6027\u80bf\u7624(\u53cd\u4f8b).<br \/>\n# \u6a21\u578bA\u9884\u6d4b\u7ed3\u679c\u4e3a: \u9884\u6d4b\u5bf9\u4e863\u4e2a\u6076\u6027\u80bf\u7624, \u9884\u6d4b\u5bf9\u4e864\u4e2a\u826f\u6027\u80bf\u7624<br \/>\n# \u6a21\u578bB\u9884\u6d4b\u7ed3\u679c\u4e3a: \u9884\u6d4b\u5bf9\u4e866\u4e2a\u6076\u6027\u80bf\u7624, \u9884\u6d4b\u5bf9\u4e861\u4e2a\u826f\u6027\u80bf\u7624<br \/>\n# \u8bf7\u9488\u5bf9\u4e8e\u4e0a\u8ff0\u7684\u6570\u636e\u96c6, \u642d\u5efa \u6df7\u6dc6\u77e9\u9635, \u5e76\u5206\u522b\u8ba1\u7b97\u6a21\u578bA, \u6a21\u578bB\u7684 \u7cbe\u786e\u7387, \u53ec\u56de\u7387, F1\u503c.<\/p>\n<p># 1. \u5b9a\u4e49\u53d8\u91cf, \u8bb0\u5f55: \u6837\u672c\u6570\u636e<br \/>\ny_train &#061; [&#039;\u6076\u6027&#039;, &#039;\u6076\u6027&#039;, &#039;\u6076\u6027&#039;, &#039;\u6076\u6027&#039;, &#039;\u6076\u6027&#039;, &#039;\u6076\u6027&#039;,     &#039;\u826f\u6027&#039;, &#039;\u826f\u6027&#039;, &#039;\u826f\u6027&#039;, &#039;\u826f\u6027&#039;]<\/p>\n<p># 2. \u5b9a\u4e49\u53d8\u91cf, \u8bb0\u5f55: \u6a21\u578bA\u7684\u9884\u6d4b\u7ed3\u679c<br \/>\ny_pred_A &#061; [&#039;\u6076\u6027&#039;, &#039;\u6076\u6027&#039;, &#039;\u6076\u6027&#039;, &#039;\u826f\u6027&#039;, &#039;\u826f\u6027&#039;, &#039;\u826f\u6027&#039;,    &#039;\u826f\u6027&#039;, &#039;\u826f\u6027&#039;, &#039;\u826f\u6027&#039;, &#039;\u826f\u6027&#039;]<\/p>\n<p># 3. \u5b9a\u4e49\u53d8\u91cf, \u8bb0\u5f55: \u6a21\u578bB\u7684\u9884\u6d4b\u7ed3\u679c<br \/>\ny_pred_B &#061; [&#039;\u6076\u6027&#039;, &#039;\u6076\u6027&#039;, &#039;\u6076\u6027&#039;, &#039;\u6076\u6027&#039;, &#039;\u6076\u6027&#039;, &#039;\u6076\u6027&#039;,     &#039;\u826f\u6027&#039;, &#039;\u6076\u6027&#039;, &#039;\u6076\u6027&#039;, &#039;\u6076\u6027&#039;]<\/p>\n<p># 4. \u7528\u6807\u7b7e\u6807\u8bb0 \u6b63\u4f8b, \u53cd\u4f8b.<br \/>\nlabel &#061; [&#039;\u6076\u6027&#039;, &#039;\u826f\u6027&#039;] # \u6807\u7b7e<br \/>\ndf_label &#061; [&#039;\u6076\u6027(\u6b63\u4f8b)&#039;, &#039;\u826f\u6027(\u53cd\u4f8b)&#039;]<\/p>\n<p># 5. \u9488\u5bf9\u4e8e \u771f\u5b9e\u503c(y_train) \u548c \u6a21\u578bA\u7684\u9884\u6d4b\u7ed3\u679c(y_pred_A), \u642d\u5efa \u6df7\u6dc6\u77e9\u9635.<br \/>\ncm_A &#061; confusion_matrix(y_train, y_pred_A, labels&#061;label) # \u53c2\u65701: \u771f\u5b9e\u503c, \u53c2\u65702: \u9884\u6d4b\u503c, \u53c2\u65703: \u6807\u7b7e, \u9ed8\u8ba4: \u9ed8\u8ba4, \u4f1a\u7528 \u5206\u7c7b\u5c11\u7684 \u6837\u672c\u5f53\u505a \u6b63\u4f8b.<br \/>\nprint(f&#039;\u6df7\u6dc6\u77e9\u9635A: \\\\n {cm_A}&#039;)<\/p>\n<p># 6. \u4e3a\u4e86\u6d4b\u8bd5\u7ed3\u679c\u66f4\u597d\u770b, \u628a\u4e0a\u8ff0\u7684 \u6df7\u6dc6\u77e9\u9635 \u8f6c\u6362\u6210 DataFrame.<br \/>\ndf_A &#061; pd.DataFrame(cm_A, index&#061;df_label, columns&#061;df_label)<br \/>\nprint(f&#039;\u6df7\u6dc6\u77e9\u9635A\u7684 DataFrame\u5bf9\u8c61\u5f62\u5f0f: \\\\n {df_A}&#039;)<\/p>\n<p># 7. \u9488\u5bf9\u4e8e \u771f\u5b9e\u503c(y_train) \u548c \u6a21\u578bB\u7684\u9884\u6d4b\u7ed3\u679c(y_pred_B), \u642d\u5efa \u6df7\u6dc6\u77e9\u9635.<br \/>\ncm_B &#061; confusion_matrix(y_train, y_pred_B, labels&#061;label)<br \/>\nprint(f&#039;\u6df7\u6dc6\u77e9\u9635B: \\\\n {cm_B}&#039;)<\/p>\n<p># 8. \u4e3a\u4e86\u6d4b\u8bd5\u7ed3\u679c\u66f4\u597d\u770b, \u628a\u4e0a\u8ff0\u7684 \u6df7\u6dc6\u77e9\u9635 \u8f6c\u6362\u6210 DataFrame.<br \/>\ndf_B &#061; pd.DataFrame(cm_B, index&#061;df_label, columns&#061;df_label)<br \/>\nprint(f&#039;\u6df7\u6dc6\u77e9\u9635B\u7684 DataFrame\u5bf9\u8c61\u5f62\u5f0f: \\\\n {df_B}&#039;)<\/p>\n<p># 9. \u8ba1\u7b97A\u6a21\u578b\u7684 \u7cbe\u786e\u7387, \u53ec\u56de\u7387, F1\u503c.<br \/>\nprint(f&#039;\u6a21\u578bA \u7cbe\u786e\u7387: {precision_score(y_train, y_pred_A, pos_label&#061;&#039;\u6076\u6027&#039;)}&#039;)   # \u53c21: \u771f\u5b9e\u503c, \u53c22: \u9884\u6d4b\u503c, \u53c23: \u6b63\u4f8b\u7684\u6807\u7b7e<br \/>\nprint(f&#039;\u6a21\u578bA \u53ec\u56de\u7387: {recall_score(y_train, y_pred_A, pos_label&#061;&#034;\u6076\u6027&#034;)}&#039;)      # \u53c21: \u771f\u5b9e\u503c, \u53c22: \u9884\u6d4b\u503c, \u53c23: \u6b63\u4f8b\u7684\u6807\u7b7e<br \/>\nprint(f&#039;\u6a21\u578bA F1\u503c: {f1_score(y_train, y_pred_A, pos_label&#061;&#034;\u6076\u6027&#034;)}&#039;)           # \u53c21: \u771f\u5b9e\u503c, \u53c22: \u9884\u6d4b\u503c, \u53c23: \u6b63\u4f8b\u7684\u6807\u7b7e<\/p>\n<p># 10. \u8ba1\u7b97B\u6a21\u578b\u7684 \u7cbe\u786e\u7387, \u53ec\u56de\u7387, F1\u503c.<br \/>\nprint(f&#039;\u6a21\u578bB \u7cbe\u786e\u7387: {precision_score(y_train, y_pred_B, pos_label&#061;&#039;\u6076\u6027&#039;)}&#039;)   # \u53c21: \u771f\u5b9e\u503c, \u53c22: \u9884\u6d4b\u503c, \u53c23: \u6b63\u4f8b\u7684\u6807\u7b7e<br \/>\nprint(f&#039;\u6a21\u578bB \u53ec\u56de\u7387: {recall_score(y_train, y_pred_B, pos_label&#061;&#034;\u6076\u6027&#034;)}&#039;)      # \u53c21: \u771f\u5b9e\u503c, \u53c22: \u9884\u6d4b\u503c, \u53c23: \u6b63\u4f8b\u7684\u6807\u7b7e<br \/>\nprint(f&#039;\u6a21\u578bB F1\u503c: {f1_score(y_train, y_pred_B, pos_label&#061;&#034;\u6076\u6027&#034;)}&#039;)           # \u53c21: \u771f\u5b9e\u503c, \u53c22: \u9884\u6d4b\u503c, \u53c23: \u6b63\u4f8b\u7684\u6807\u7b7e<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" alt=\"\" height=\"288\" src=\"https:\/\/www.wsisp.com\/helps\/wp-content\/uploads\/2026\/02\/20260227143210-69a1aaeae2a11.png\" width=\"704\" \/><\/p>\n<h3>ROC\u66f2\u7ebf\u7684\u7ed8\u5236<\/h3>\n<p><img loading=\"lazy\" decoding=\"async\" alt=\"\" height=\"742\" src=\"https:\/\/www.wsisp.com\/helps\/wp-content\/uploads\/2026\/02\/20260227143211-69a1aaeb08a47.png\" width=\"1643\" \/><\/p>\n<p><span style=\"color:#ff9900\">FPR\u4f2a\u6b63\u7387:\u4f2a\u6b63\u4f8bFP \u5728 \u5168\u90e8\u5047\u4f8b \u7684\u5360\u6bd4<\/span><\/p>\n<p><span style=\"color:#ff9900\">TPR\u771f\u6b63\u7387:\u771f\u6b63\u4f8bTP \u5728 \u5168\u90e8\u6b63\u4f8b \u7684\u5360\u6bd4<\/span><\/p>\n<p><img loading=\"lazy\" decoding=\"async\" alt=\"\" height=\"807\" src=\"https:\/\/www.wsisp.com\/helps\/wp-content\/uploads\/2026\/02\/20260227143211-69a1aaeb8df6a.png\" width=\"959\" \/><\/p>\n<h3 style=\"background-color:transparent\">\u5ba2\u6237\u6d41\u5931\u6848\u4f8b\u5206\u6790<\/h3>\n<p>&#034;&#034;&#034;<br \/>\n\u6848\u4f8b:<br \/>\n    \u7535\u4fe1\u5ba2\u6237\u6d41\u5931\u5206\u6790.<\/p>\n<p>\u76ee\u7684:<br \/>\n    1. \u6f14\u793a\u903b\u8f91\u56de\u5f52\u7684\u76f8\u5173\u64cd\u4f5c, \u4e3b\u8981\u662f: \u4e8c\u5206\u6cd5(\u6d41\u5931, \u4e0d\u6d41\u5931)<br \/>\n    2. \u6f14\u793a\u903b\u8f91\u56de\u5f52\u7684\u8bc4\u4f30\u64cd\u4f5c, \u4e3b\u8981\u662f: \u6df7\u6dc6\u77e9\u9635, \u51c6\u786e\u7387, \u53ec\u56de\u7387, F1\u503c, ROC\u66f2\u7ebf, AUC\u503c, \u5206\u7c7b\u8bc4\u4f30\u62a5\u544a(\u4e86\u89e3)<br \/>\n&#034;&#034;&#034;<br \/>\nimport pandas as pd<br \/>\nimport matplotlib.pyplot as plt<br \/>\nimport seaborn as sns<br \/>\nfrom sklearn.linear_model import LogisticRegression<br \/>\n#                             \u51c6\u786e\u7387           \u7cbe\u786e\u7387            \u53ec\u56de\u7387        F1\u503c        roc\u66f2\u7ebf          \u5206\u7c7b\u8bc4\u4f30\u62a5\u544a<br \/>\nfrom sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score, roc_auc_score, classification_report<br \/>\nfrom sklearn.model_selection import train_test_split<\/p>\n<p># 1. \u5b9a\u4e49\u51fd\u6570, \u7528\u4e8e\u5b9e\u73b0: \u6570\u636e\u9884\u5904\u7406.<br \/>\ndef dm01_\u6570\u636e\u9884\u5904\u7406():<br \/>\n    # 1. \u8bfb\u53d6\u6570\u636e.<br \/>\n    data &#061; pd.read_csv(&#039;.\/data\/churn.csv&#039;)<br \/>\n    data.info()<\/p>\n<p>    # 2. \u56e0\u4e3a\u4e0a\u8ff0\u7684Churn, gender\u662f\u5b57\u7b26\u4e32\u7c7b\u578b, \u6211\u4eec\u5bf9\u5176\u505a\u70ed\u7f16\u7801(one-hot)\u5904\u7406.<br \/>\n    data &#061; pd.get_dummies(data)<br \/>\n    data.info()<br \/>\n    print(data.head(10))<\/p>\n<p>    # 3. \u5220\u9664\u5217, \u56e0\u4e3a\u70ed\u7f16\u7801\u4e4b\u540e, \u4f1a\u591a\u51fa\u4e00\u4e2a\u5217, \u6211\u4eec\u5220\u9664\u6389.<br \/>\n    data.drop([&#039;gender_Male&#039;, &#039;Churn_No&#039;], axis&#061;1, inplace&#061;True)<br \/>\n    print(data.head(10))<\/p>\n<p>    # 4. \u4fee\u6539\u5217\u540d.<br \/>\n    data.rename(columns&#061;{&#039;Churn_Yes&#039;:&#039;flag&#039;}, inplace&#061;True)<br \/>\n    print(data.head(10))<\/p>\n<p>    # 5. \u6211\u4eec\u67e5\u770b\u4e0b\u6570\u636e\u96c6\u4e2d, \u6807\u7b7e \u662f\u5426\u662f \u5747\u8861\u7684.<br \/>\n    print(data.flag.value_counts()) # False -&gt; \u4e0d\u6d41\u5931, True -&gt; \u6d41\u5931<\/p>\n<p># 2. \u5b9a\u4e49\u51fd\u6570, \u7528\u4e8e\u663e\u793a: \u6708\u5ea6\u4f1a\u5458\u7684\u6d41\u5931\u60c5\u51b5.<br \/>\ndef dm02_\u4f1a\u5458\u6d41\u5931\u53ef\u89c6\u5316\u60c5\u51b5():<br \/>\n    # 1. \u8bfb\u53d6\u6570\u636e.<br \/>\n    data &#061; pd.read_csv(&#039;.\/data\/churn.csv&#039;)<br \/>\n    # 2. \u5bf9\u4e0a\u8ff0\u7684\u6570\u636e\u505a \u70ed\u7f16\u7801\u5904\u7406.<br \/>\n    data &#061; pd.get_dummies(data)<br \/>\n    # 3. \u5220\u9664\u5217, \u56e0\u4e3a\u70ed\u7f16\u7801\u4e4b\u540e, \u4f1a\u591a\u51fa\u4e00\u4e2a\u5217, \u6211\u4eec\u5220\u9664\u6389.<br \/>\n    data.drop([&#039;gender_Male&#039;, &#039;Churn_No&#039;], axis&#061;1, inplace&#061;True)<br \/>\n    # 4. \u4fee\u6539\u5217\u540d.<br \/>\n    data.rename(columns&#061;{&#039;Churn_Yes&#039;:&#039;flag&#039;}, inplace&#061;True)<br \/>\n    # 5. \u67e5\u770b\u6570\u636e\u96c6\u7684\u5206\u5e03\u60c5\u51b5.<br \/>\n    print(data.flag.value_counts())<br \/>\n    print(data.columns) # \u67e5\u770b\u6240\u6709\u5217\u540d.<\/p>\n<p>    # 6. \u901a\u8fc7\u8ba1\u6570\u67f1\u72b6\u56fe, \u7ed8\u5236(\u6708\u5ea6)\u4f1a\u5458\u7684\u6d41\u5931\u60c5\u51b5.<br \/>\n    # \u53c2\u6570x\u610f\u601d\u662f: x\u8f74\u7684\u5217\u540d(\u662f\u5426\u662f\u6708\u5ea6\u4f1a\u5458, 0 -&gt; \u4e0d\u662f\u4f1a\u5458, 1 -&gt; \u662f\u4f1a\u5458)<br \/>\n    # \u53c2\u6570hue\u610f\u601d\u662f: \u6839\u636ehue\u7684\u503c, \u5c06\u6570\u636e\u8fdb\u884c\u5206\u7c7b(False -&gt; \u4e0d\u6d41\u5931, True -&gt; \u6d41\u5931)<br \/>\n    sns.countplot(data, x&#061;&#039;Contract_Month&#039;, hue&#061;&#039;flag&#039;)<br \/>\n    plt.show()<\/p>\n<p># 3. \u5b9a\u4e49\u51fd\u6570, \u7528\u4e8e\u5b9e\u73b0: \u903b\u8f91\u56de\u5f52\u6a21\u578b\u7684\u8bad\u7ec3\u548c\u8bc4\u4f30.<br \/>\ndef dm03_\u903b\u8f91\u56de\u5f52\u6a21\u578b\u8bad\u7ec3\u8bc4\u4f30():<br \/>\n    # 1. \u8bfb\u53d6\u6570\u636e.<br \/>\n    data &#061; pd.read_csv(&#039;.\/data\/churn.csv&#039;)<br \/>\n    # 2. \u5bf9\u4e0a\u8ff0\u7684\u6570\u636e\u505a \u70ed\u7f16\u7801\u5904\u7406.<br \/>\n    data &#061; pd.get_dummies(data)<br \/>\n    # 3. \u5220\u9664\u5217, \u56e0\u4e3a\u70ed\u7f16\u7801\u4e4b\u540e, \u4f1a\u591a\u51fa\u4e00\u4e2a\u5217, \u6211\u4eec\u5220\u9664\u6389.<br \/>\n    data.drop([&#039;gender_Male&#039;, &#039;Churn_No&#039;], axis&#061;1, inplace&#061;True)<br \/>\n    # 4. \u4fee\u6539\u5217\u540d.<br \/>\n    data.rename(columns&#061;{&#039;Churn_Yes&#039;:&#039;flag&#039;}, inplace&#061;True)<br \/>\n    # 5. \u67e5\u770b\u6570\u636e\u96c6, \u4ece\u4e2d\u7b5b\u9664: \u7279\u5f81\u5217 \u548c \u6807\u7b7e\u5217.<br \/>\n    # print(data.head(10))    # \u7279\u5f81\u5217: Contract_Month, PaymentElectronic, internet_other<br \/>\n    # print(data.columns)     # \u6807\u7b7e\u5217: flag<\/p>\n<p>    # 6. \u62c6\u5206\u8bad\u7ec3\u96c6\u548c\u6d4b\u8bd5\u96c6.<br \/>\n    x &#061; data[[&#039;Contract_Month&#039;, &#039;PaymentElectronic&#039;, &#039;internet_other&#039;]]<br \/>\n    y &#061; data[&#039;flag&#039;]<br \/>\n    # print(len(x), len(y))<br \/>\n    # print(x.head(10))<br \/>\n    # print(y.head(10))<br \/>\n    x_train, x_test, y_train, y_test &#061; train_test_split(x, y, test_size&#061;0.2, random_state&#061;22)<\/p>\n<p>    # 7. \u521b\u5efa\u903b\u8f91\u56de\u5f52\u6a21\u578b, \u5e76\u8bad\u7ec3.<br \/>\n    estimator &#061; LogisticRegression()<br \/>\n    estimator.fit(x_train, y_train)<\/p>\n<p>    # 8. \u6a21\u578b\u9884\u6d4b.<br \/>\n    y_predict &#061; estimator.predict(x_test)<br \/>\n    print(f&#039;\u9884\u6d4b\u503c\u4e3a: {y_predict}&#039;)<\/p>\n<p>    # 9. \u6a21\u578b\u8bc4\u4f30.<br \/>\n    # 9.1 \u51c6\u786e\u7387.<br \/>\n    print(f&#039;\u51c6\u786e\u7387: {estimator.score(x_test, y_test)}&#039;)<br \/>\n    print(f&#039;\u51c6\u786e\u7387: {accuracy_score(y_test, y_predict)}&#039;)  # \u771f\u5b9e\u503c, \u9884\u6d4b\u503c.<br \/>\n    print(&#039;-&#039; * 22)<br \/>\n    # 9.2 \u7cbe\u786e\u7387.<br \/>\n    print(f&#039;\u7cbe\u786e\u7387: {precision_score(y_test, y_predict)}&#039;)<br \/>\n    print(&#039;-&#039; * 22)<br \/>\n    # 9.3 \u53ec\u56de\u7387.<br \/>\n    print(f&#039;\u53ec\u56de\u7387: {recall_score(y_test, y_predict)}&#039;)<br \/>\n    print(&#039;-&#039; * 22)<br \/>\n    # 9.4 F1\u503c<br \/>\n    print(f&#039;F1\u503c: {f1_score(y_test, y_predict)}&#039;)<br \/>\n    print(&#039;-&#039; * 22)<br \/>\n    # 9.5 roc\u66f2\u7ebf<br \/>\n    print(f&#039;roc\u66f2\u7ebf: {roc_auc_score(y_test, y_predict)}&#039;)<br \/>\n    print(&#039;-&#039; * 22)<br \/>\n    # 9.6 \u5206\u7c7b\u8bc4\u4f30\u62a5\u544a<br \/>\n    # \u53c2\u6570macro avg\u610f\u601d\u662f: \u5b8f\u5e73\u5747, \u662f\u6307: \u6240\u6709\u7684\u5206\u7c7b\u5668, \u90fd\u6309\u7167 macro \u7684\u65b9\u5f0f, \u8ba1\u7b97\u5e73\u5747\u503c\/<br \/>\n    # \u4e0d\u8003\u8651\u6837\u672c\u7684\u6743\u91cd, \u76f4\u63a5\u5e73\u5747, \u8ddf\u6837\u672c\u7684\u6570\u91cf, \u6743\u91cd\u65e0\u5173, \u6240\u6709\u7279\u5f81\u6743\u91cd\u90fd\u4e00\u6837, \u9002\u5408\u4e8e \u6570\u636e\u96c6\u6bd4\u8f83\u5e73\u8861\u7684\u60c5\u51b5.<\/p>\n<p>    # \u53c2\u6570weighted avg\u610f\u601d\u662f: \u6743\u91cd\u5e73\u5747, \u662f\u6307: \u6240\u6709\u7684\u5206\u7c7b\u5668, \u90fd\u6309\u7167 weighted \u7684\u65b9\u5f0f, \u8ba1\u7b97\u5e73\u5747\u503c\/<br \/>\n    # \u8003\u8651\u6837\u672c\u7684\u6743\u91cd, \u6839\u636e\u6837\u672c\u7684\u6743\u91cd, \u8ba1\u7b97\u5e73\u5747\u503c, \u9002\u5408\u4e8e \u6570\u636e\u96c6\u6bd4\u8f83\u4e0d\u5e73\u8861\u7684\u60c5\u51b5.<br \/>\n    print(f&#039;\u5206\u7c7b\u8bc4\u4f30\u62a5\u544a: {classification_report(y_test, y_predict)}&#039;)<\/p>\n<p># 4. \u5728main\u51fd\u6570\u4e2d\u6d4b\u8bd5.<br \/>\nif __name__ &#061;&#061; &#039;__main__&#039;:<br \/>\n    # dm01_\u6570\u636e\u9884\u5904\u7406()<br \/>\n    # dm02_\u4f1a\u5458\u6d41\u5931\u53ef\u89c6\u5316\u60c5\u51b5()<br \/>\n    dm03_\u903b\u8f91\u56de\u5f52\u6a21\u578b\u8bad\u7ec3\u8bc4\u4f30()<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" alt=\"\" height=\"533\" src=\"https:\/\/www.wsisp.com\/helps\/wp-content\/uploads\/2026\/02\/20260227143211-69a1aaebd33f7.png\" width=\"868\" \/><\/p>\n<h4 style=\"background-color:transparent\">\u4e00\u3001\u5206\u7c7b\u8bc4\u4f30\u62a5\u544a\u53c2\u6570\u8be6\u89e3<\/h4>\n<p>\u9996\u5148&#xff0c;\u6211\u4eec\u6765\u9010\u4e00\u89e3\u8bfb\u62a5\u544a\u4e2d\u6bcf\u4e00\u5217\u548c\u6bcf\u4e00\u884c\u7684\u542b\u4e49\u3002<\/p>\n<h5>1. \u884c&#xff08;\u7c7b\u522b&#xff09;<\/h5>\n<p>\u62a5\u544a\u5c55\u793a\u4e86\u4e24\u884c&#xff0c;\u5bf9\u5e94\u6a21\u578b\u7684\u8f93\u51fa\u7c7b\u522b\u3002\u5728\u8fd9\u4e2a\u5ba2\u6237\u6d41\u5931\u6848\u4f8b\u4e2d&#xff1a;<\/p>\n<ul>\n<li>\n<p>False\u00a0(0)&#xff1a;\u4ee3\u8868\u8d1f\u7c7b&#xff0c;\u5373\u4e0d\u6d41\u5931\u7684\u5ba2\u6237\u3002<\/p>\n<\/li>\n<li>\n<p>True\u00a0(1)&#xff1a;\u4ee3\u8868\u6b63\u7c7b&#xff0c;\u5373\u6d41\u5931\u7684\u5ba2\u6237\u3002<\/p>\n<\/li>\n<\/ul>\n<h5>2. \u5217&#xff08;\u8bc4\u4f30\u6307\u6807&#xff09;<\/h5>\n<p><img loading=\"lazy\" decoding=\"async\" alt=\"\" height=\"300\" src=\"https:\/\/www.wsisp.com\/helps\/wp-content\/uploads\/2026\/02\/20260227143212-69a1aaec0b2ae.png\" width=\"1369\" \/><\/p>\n<p>\u5982\u679c\u6709\u7591\u60d1\u53ef\u4ee5\u53c2\u8003\u7740\u4e0b\u56fe\u6765\u89e3\u6790\u8bc4\u4f30\u6307\u6807<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" alt=\"\" height=\"667\" src=\"https:\/\/www.wsisp.com\/helps\/wp-content\/uploads\/2026\/02\/20260227143212-69a1aaec3544d.png\" width=\"1610\" \/><\/p>\n<hr \/>\n<h4>\u4e8c\u3001\u5e95\u5c42\u4e09\u884c\u6c47\u603b\u6307\u6807\u89e3\u8bfb<\/h4>\n<p>\u62a5\u544a\u5e95\u90e8\u8fd8\u6709\u4e09\u884c&#xff0c;\u63d0\u4f9b\u4e86\u4e0d\u540c\u7ef4\u5ea6\u7684\u603b\u4f53\u8bc4\u4ef7&#xff1a;<\/p>\n<h5 style=\"background-color:transparent\">1.\u00a0accuracy\u00a0(\u51c6\u786e\u7387)<\/h5>\n<ul>\n<li>\n<p>\u6570\u503c&#xff1a;0.76<\/p>\n<\/li>\n<li>\n<p>\u542b\u4e49&#xff1a;\u6240\u6709\u6d4b\u8bd5\u6837\u672c\u4e2d&#xff0c;\u9884\u6d4b\u6b63\u786e\u7684\u6bd4\u4f8b\u3002<\/p>\n<\/li>\n<li>\n<p>\u89e3\u8bfb&#xff1a;\u6574\u4f53\u6765\u770b&#xff0c;\u6a21\u578b\u5728 76% \u7684\u60c5\u51b5\u4e0b\u80fd\u6b63\u786e\u5224\u65ad\u5ba2\u6237\u662f\u5426\u6d41\u5931\u3002\u4f46\u8981\u6ce8\u610f&#xff0c;\u8fd9\u4e2a\u6307\u6807\u5728\u6837\u672c\u4e0d\u5e73\u8861\u65f6\u5bb9\u6613\u88ab\u591a\u6570\u7c7b&#xff08;\u4e0d\u6d41\u5931&#xff09;\u5e26\u504f\u3002<\/p>\n<\/li>\n<\/ul>\n<h5>2.\u00a0macro avg\u00a0(\u5b8f\u5e73\u5747)<\/h5>\n<ul>\n<li>\n<p>\u6570\u503c&#xff1a;precision 0.71, recall 0.65, f1-score 0.67<\/p>\n<\/li>\n<li>\n<p>\u542b\u4e49&#xff1a;\u5206\u522b\u8ba1\u7b97\u6bcf\u4e2a\u7c7b\u522b\u7684\u6307\u6807&#xff0c;\u7136\u540e\u76f4\u63a5\u53d6\u7b97\u672f\u5e73\u5747&#xff0c;\u4e0d\u8003\u8651\u6837\u672c\u591a\u5c11\u3002<\/p>\n<\/li>\n<li>\n<p>\u89e3\u8bfb&#xff1a;\u5b8f\u5e73\u5747\u628a\u201c\u6d41\u5931\u201d\u548c\u201c\u4e0d\u6d41\u5931\u201d\u4e24\u4e2a\u7c7b\u522b\u770b\u5f97\u540c\u7b49\u91cd\u8981\u30020.65 \u7684\u53ec\u56de\u7387\u8bf4\u660e&#xff0c;\u6a21\u578b\u5e73\u5747\u6765\u770b&#xff0c;\u627e\u51fa\u6bcf\u4e2a\u7c7b\u522b&#xff08;\u7279\u522b\u662f\u6d41\u5931\u5ba2\u6237&#xff09;\u7684\u80fd\u529b\u8f83\u5f31\u3002\u8fd9\u4e2a\u6307\u6807\u66f4\u80fd\u53cd\u6620\u6a21\u578b\u5728\u5c11\u6570\u7c7b&#xff08;\u6d41\u5931&#xff09;\u4e0a\u7684\u7cdf\u7cd5\u8868\u73b0\u3002<\/p>\n<\/li>\n<\/ul>\n<h5>3.\u00a0weighted avg\u00a0(\u52a0\u6743\u5e73\u5747)<\/h5>\n<ul>\n<li>\n<p>\u6570\u503c&#xff1a;precision 0.74, recall 0.76, f1-score 0.74<\/p>\n<\/li>\n<li>\n<p>\u542b\u4e49&#xff1a;\u5206\u522b\u8ba1\u7b97\u6bcf\u4e2a\u7c7b\u522b\u7684\u6307\u6807&#xff0c;\u7136\u540e\u6309\u8be5\u7c7b\u6837\u672c\u6570\u52a0\u6743\u5e73\u5747\u3002<\/p>\n<\/li>\n<li>\n<p>\u89e3\u8bfb&#xff1a;\u7531\u4e8e\u201c\u4e0d\u6d41\u5931\u201d\u7684\u6837\u672c&#xff08;1012\u4e2a&#xff09;\u8fdc\u591a\u4e8e\u201c\u6d41\u5931\u201d\u7684\u6837\u672c&#xff08;397\u4e2a&#xff09;&#xff0c;\u52a0\u6743\u5e73\u5747\u7684\u7ed3\u679c\u66f4\u504f\u5411\u4e8e\u201c\u4e0d\u6d41\u5931\u201d\u7c7b\u7684\u6307\u6807&#xff0c;\u56e0\u6b64\u770b\u8d77\u6765\u6bd4\u5b8f\u5e73\u5747\u66f4\u597d\u770b\u3002\u4f46\u5728\u4e1a\u52a1\u4e2d&#xff0c;\u5982\u679c\u66f4\u5173\u6ce8\u6d41\u5931\u5ba2\u6237&#xff0c;\u8fd9\u4e2a\u6307\u6807\u4f1a\u63a9\u76d6\u771f\u6b63\u7684\u95ee\u9898\u3002<\/p>\n<\/li>\n<\/ul>\n<hr \/>\n<h4>\u4e09\u3001\u7ed3\u5408\u4ee3\u7801\u548c\u4e1a\u52a1\u573a\u666f\u7684\u5206\u6790<\/h4>\n<h5>1. \u4e1a\u52a1\u80cc\u666f<\/h5>\n<p>\u8fd9\u662f\u4e00\u4e2a\u5ba2\u6237\u6d41\u5931\u9884\u6d4b\u6a21\u578b\u3002\u76ee\u6807\u662f\u63d0\u524d\u53d1\u73b0\u53ef\u80fd\u6d41\u5931\u7684\u5ba2\u6237&#xff08;True&#xff09;&#xff0c;\u4ee5\u4fbf\u8fd0\u8425\u5546\u91c7\u53d6\u63aa\u65bd&#xff08;\u5982\u53d1\u9001\u4f18\u60e0\u5238&#xff09;\u8fdb\u884c\u633d\u7559\u3002<\/p>\n<h5 style=\"background-color:transparent\">2. \u6a21\u578b\u8868\u73b0\u5206\u6790<\/h5>\n<p>\u4ece\u4e1a\u52a1\u89d2\u5ea6\u770b&#xff0c;\u6211\u4eec\u6700\u5173\u5fc3\u7684\u662f\u53ec\u56de\u7387 (Recall)&#xff0c;\u56e0\u4e3a\u6f0f\u6389\u4e00\u4e2a\u771f\u6b63\u8981\u6d41\u5931\u7684\u5ba2\u6237&#xff08;FN&#xff09;&#xff0c;\u5c31\u610f\u5473\u7740\u76f4\u63a5\u635f\u5931\u4e86\u4e00\u4e2a\u5ba2\u6237\u3002\u76f8\u6bd4\u4e4b\u4e0b&#xff0c;\u628a\u6ca1\u60f3\u6d41\u5931\u7684\u5ba2\u6237\u8bef\u5224\u4e3a\u6d41\u5931&#xff08;FP&#xff09;&#xff0c;\u6700\u591a\u53ea\u662f\u591a\u53d1\u4e00\u5f20\u4f18\u60e0\u5238\u7684\u6210\u672c\u3002<\/p>\n<ul>\n<li>\n<p>\u6a21\u578b\u77ed\u677f&#xff1a;\u6a21\u578b\u7684\u53ec\u56de\u7387&#xff08;\u6d41\u5931\u7c7b&#xff09;\u53ea\u6709 0.40\u3002\u8fd9\u610f\u5473\u7740\u5728\u6240\u6709\u771f\u5b9e\u6d41\u5931\u7684\u5ba2\u6237\u4e2d&#xff0c;\u6a21\u578b\u53ea\u6210\u529f\u9884\u6d4b\u51fa\u4e86\u00a040%&#xff0c;\u800c\u6f0f\u6389\u4e86\u5269\u4e0b\u7684\u00a060%\u3002\u8fd9\u4e2a\u6a21\u578b\u5728\u5b9e\u9645\u4e1a\u52a1\u4e2d\u5e2e\u52a9\u4e0d\u5927&#xff0c;\u56e0\u4e3a\u5927\u90e8\u5206\u60f3\u8d70\u7684\u5ba2\u6237\u5b83\u90fd\u6ca1\u53d1\u73b0\u3002<\/p>\n<\/li>\n<li>\n<p>\u6837\u672c\u4e0d\u5e73\u8861&#xff1a;\u4ece\u00a0support\u00a0\u5217&#xff08;1012 vs 397&#xff09;\u53ef\u4ee5\u6e05\u6670\u770b\u5230&#xff0c;\u6d4b\u8bd5\u96c6\u4e2d\u201c\u4e0d\u6d41\u5931\u201d\u7684\u6837\u672c\u8fdc\u591a\u4e8e\u201c\u6d41\u5931\u201d\u7684\u6837\u672c\u3002\u6a21\u578b\u4e3a\u4e86\u6574\u4f53\u51c6\u786e\u7387&#xff0c;\u4f1a\u66f4\u503e\u5411\u4e8e\u5b66\u4e60\u201c\u4e0d\u6d41\u5931\u201d\u7684\u7279\u5f81&#xff0c;\u5bfc\u81f4\u5bf9\u201c\u6d41\u5931\u201d\u7684\u5224\u65ad\u80fd\u529b\u5f31\u3002<\/p>\n<\/li>\n<\/ul>\n<h5>3. \u4ee3\u7801\u903b\u8f91\u5370\u8bc1<\/h5>\n<ul>\n<li>\n<p>\u7279\u5f81\u9009\u62e9&#xff1a;\u4ee3\u7801\u4e2d\u53ea\u7528\u4e86\u4e09\u4e2a\u7279\u5f81&#xff1a;[&#039;Contract_Month&#039;&#xff0c; &#039;PaymentElectronic&#039;, &#039;internet_other&#039;]\u3002\u8fd9\u53ef\u80fd\u8fc7\u4e8e\u7b80\u5316&#xff0c;\u9057\u6f0f\u4e86\u5982\u201c\u7528\u6237\u6708\u6d88\u8d39\u91d1\u989d\u201d\u3001\u201c\u5ba2\u670d\u901a\u8bdd\u6b21\u6570\u201d\u7b49\u91cd\u8981\u7279\u5f81&#xff0c;\u5bfc\u81f4\u6a21\u578b\u5b66\u4e60\u80fd\u529b\u6709\u9650\u3002<\/p>\n<\/li>\n<li>\n<p>\u6a21\u578b\u9009\u62e9&#xff1a;\u4f7f\u7528\u4e86\u9ed8\u8ba4\u53c2\u6570\u7684\u00a0LogisticRegression&#xff0c;\u6ca1\u6709\u9488\u5bf9\u6837\u672c\u4e0d\u5e73\u8861\u95ee\u9898\u8fdb\u884c\u8c03\u6574&#xff08;\u5982\u8bbe\u7f6e\u00a0class_weight&#061;&#039;balanced&#039;&#xff09;\u3002<\/p>\n<\/li>\n<\/ul>\n<hr \/>\n<h4>\u56db\u3001\u603b\u7ed3\u4e0e\u4f18\u5316\u5efa\u8bae<\/h4>\n<p>\u4e00\u53e5\u8bdd\u603b\u7ed3\u62a5\u544a&#xff1a;\u8fd9\u4e2a\u903b\u8f91\u56de\u5f52\u6a21\u578b\u5bf9\u4e0d\u6d41\u5931\u7684\u5ba2\u6237\u9884\u6d4b\u5f97\u5f88\u597d&#xff08;\u53ec\u56de\u7387 90%&#xff09;&#xff0c;\u4f46\u5bf9\u4e1a\u52a1\u771f\u6b63\u5173\u5fc3\u7684\u6d41\u5931\u5ba2\u6237\u9884\u6d4b\u80fd\u529b\u5f88\u5dee&#xff08;\u53ec\u56de\u7387 40%&#xff09;&#xff0c;\u6a21\u578b\u53ef\u7528\u6027\u8f83\u4f4e\u3002<\/p>\n<p>\u5982\u679c\u8981\u5411\u9762\u8bd5\u5b98\u6216\u9886\u5bfc\u6c47\u62a5&#xff0c;\u53ef\u4ee5\u8fd9\u6837\u8bf4&#xff1a;<\/p>\n<p>\u201c\u4ece\u5206\u7c7b\u8bc4\u4f30\u62a5\u544a\u770b&#xff0c;\u6a21\u578b\u6574\u4f53\u51c6\u786e\u7387\u6709 76%&#xff0c;\u4f46\u8fd9\u4e2a\u6307\u6807\u88ab\u591a\u6570\u7c7b&#xff08;\u4e0d\u6d41\u5931&#xff09;\u62c9\u9ad8\u4e86\u3002\u771f\u6b63\u7684\u4e1a\u52a1\u77ed\u677f\u5728\u4e8e\u5bf9\u6d41\u5931\u5ba2\u6237\u7684\u53ec\u56de\u7387\u53ea\u6709 40%&#xff0c;\u8fd9\u610f\u5473\u7740 60% \u5373\u5c06\u6d41\u5931\u7684\u5ba2\u6237\u6ca1\u6709\u88ab\u8bc6\u522b\u51fa\u6765\u3002\u4e3b\u8981\u539f\u56e0\u662f\u6837\u672c\u4e0d\u5e73\u8861&#xff0c;\u4e14\u76ee\u524d\u4f7f\u7528\u7684\u7279\u5f81\u8f83\u5c11\u3002\u4e0b\u4e00\u6b65\u9700\u8981\u9488\u5bf9\u8fd9\u4e9b\u95ee\u9898\u8fdb\u884c\u4f18\u5316\u3002\u201d\u6211\u89c9\u5f97\u5206\u6790\u7684\u5f88\u5230\u4f4d!!!<\/p>\n<p>\u540e\u7eed\u4f18\u5316\u65b9\u5411&#xff1a;<\/p>\n<li>\n<p>\u589e\u52a0\u7279\u5f81&#xff1a;\u5f15\u5165\u66f4\u591a\u4e0e\u6d41\u5931\u76f8\u5173\u7684\u7279\u5f81&#xff0c;\u5982\u6d88\u8d39\u91d1\u989d\u3001\u4f7f\u7528\u65f6\u957f\u3001\u6295\u8bc9\u6b21\u6570\u7b49\u3002<\/p>\n<\/li>\n<li>\n<p>\u5904\u7406\u6837\u672c\u4e0d\u5e73\u8861&#xff1a;<\/p>\n<ul>\n<li>\n<p>\u8c03\u6574\u6a21\u578b\u6743\u91cd&#xff1a;\u5728\u00a0LogisticRegression\u00a0\u4e2d\u8bbe\u7f6e\u00a0class_weight&#061;&#039;balanced&#039;\u3002<\/p>\n<\/li>\n<li>\n<p>\u91cd\u91c7\u6837&#xff1a;\u5bf9\u5c11\u6570\u7c7b&#xff08;\u6d41\u5931&#xff09;\u8fdb\u884c\u8fc7\u91c7\u6837&#xff08;\u5982SMOTE&#xff09;&#xff0c;\u6216\u5bf9\u591a\u6570\u7c7b\u8fdb\u884c\u6b20\u91c7\u6837\u3002<\/p>\n<\/li>\n<\/ul>\n<\/li>\n<li>\n<p>\u8c03\u6574\u5206\u7c7b\u9608\u503c&#xff1a;\u6a21\u578b\u9ed8\u8ba4\u7528 0.5 \u4f5c\u4e3a\u5224\u65ad\u6d41\u5931\/\u4e0d\u6d41\u5931\u7684\u9608\u503c\u3002\u4e3a\u4e86\u63d0\u9ad8\u53ec\u56de\u7387&#xff0c;\u53ef\u4ee5\u9002\u5f53\u964d\u4f4e\u9608\u503c&#xff08;\u5982 0.3&#xff09;&#xff0c;\u8ba9\u6a21\u578b\u5bf9\u6d41\u5931\u66f4\u654f\u611f&#xff08;\u4f46\u8fd9\u4f1a\u589e\u52a0\u8bef\u62a5&#xff0c;\u9700\u8981\u6743\u8861&#xff09;\u3002<\/p>\n<\/li>\n<p><span style=\"color:#ff9900\">\u4ee5\u4e0a\u5c31\u662f\u8be5\u5206\u7c7b\u8bc4\u4f30\u62a5\u544a\u7684\u6574\u4f53\u60c5\u51b5\u6c47\u603b\u4e86!!!<\/span><\/p>\n","protected":false},"excerpt":{"rendered":"<p>\u6982\u8981\u4ecb\u7ecd\u5bf9\u4e8e\u903b\u8f91\u56de\u5f52\u6211\u4eec\u53ef\u4ee5\u4ece\u4e00\u4e2a\u95ee\u9898\u5165\u624b,\u5230\u5e95\u4ec0\u4e48\u662f\u903b\u8f91\u56de\u5f52\u7684\u7b97\u6cd5\u601d\u60f3?\u9762\u8bd5\u5b98\u95ee\u201c\u8bf7\u63cf\u8ff0\u903b\u8f91\u56de\u5f52\u7684\u7b97\u6cd5\u601d\u60f3\u201d\u8fd9\u4e2a\u95ee\u9898\u65f6&#xff0c;\u5176\u5b9e\u662f\u5728\u8003\u5bdf\u4f60\u5bf9\u8fd9\u4e2a\u57fa\u7840\u6a21\u578b\u7684<br \/>\n\u672c\u8d28\u7406\u89e3\u2014\u2014\u662f\u53ea\u4f1a\u8c03\u7528sklearn.learn.LogisticRegression&#xff0c;\u8fd8\u662f\u771f\u6b63\u660e\u767d\u5b83\u4e3a\u4ec0\u4e48\u53eb\u201c\u56de\u5f52\u201d\u5374\u505a\u5206\u7c7b&#xff0c;\u5b83\u7684\u6838\u5fc3\u5728\u7b97\u4ec0\u4e48\u3002\u201c\u903b\u8f91\u56de\u5f52\u867d\u7136\u540d\u5b57\u91cc\u6709\u2018\u56de\u5f52\u2019&#xff0c;\u4f46\u5b83\u5b9e\u9645\u4e0a\u662f\u4e00\u79cd\u7528\u4e8e\u89e3\u51b3\u4e8c\u5206\u7c7b\u95ee\u9898\u7684\u7ebf\u6027\u6a21\u578b\u3002\u5b83\u7684\u6838\u5fc3\u601d\u60f3\u662f&#xff1a;\u5148\u62df\u5408\u51b3\u7b56\u8fb9\u754c&amp;<\/p>\n","protected":false},"author":2,"featured_media":78638,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[50,801,62,207,3407],"topic":[],"class_list":["post-78647","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-server","tag-50","tag-801","tag-62","tag-207","tag-3407"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v20.3 - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>\u673a\u5668\u5b66\u4e60\u4e4b\u903b\u8f91\u56de\u5f52 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3<\/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.wsisp.com\/helps\/78647.html\" \/>\n<meta property=\"og:locale\" content=\"zh_CN\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"\u673a\u5668\u5b66\u4e60\u4e4b\u903b\u8f91\u56de\u5f52 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3\" \/>\n<meta property=\"og:description\" content=\"\u6982\u8981\u4ecb\u7ecd\u5bf9\u4e8e\u903b\u8f91\u56de\u5f52\u6211\u4eec\u53ef\u4ee5\u4ece\u4e00\u4e2a\u95ee\u9898\u5165\u624b,\u5230\u5e95\u4ec0\u4e48\u662f\u903b\u8f91\u56de\u5f52\u7684\u7b97\u6cd5\u601d\u60f3?\u9762\u8bd5\u5b98\u95ee\u201c\u8bf7\u63cf\u8ff0\u903b\u8f91\u56de\u5f52\u7684\u7b97\u6cd5\u601d\u60f3\u201d\u8fd9\u4e2a\u95ee\u9898\u65f6&#xff0c;\u5176\u5b9e\u662f\u5728\u8003\u5bdf\u4f60\u5bf9\u8fd9\u4e2a\u57fa\u7840\u6a21\u578b\u7684 \u672c\u8d28\u7406\u89e3\u2014\u2014\u662f\u53ea\u4f1a\u8c03\u7528sklearn.learn.LogisticRegression&#xff0c;\u8fd8\u662f\u771f\u6b63\u660e\u767d\u5b83\u4e3a\u4ec0\u4e48\u53eb\u201c\u56de\u5f52\u201d\u5374\u505a\u5206\u7c7b&#xff0c;\u5b83\u7684\u6838\u5fc3\u5728\u7b97\u4ec0\u4e48\u3002\u201c\u903b\u8f91\u56de\u5f52\u867d\u7136\u540d\u5b57\u91cc\u6709\u2018\u56de\u5f52\u2019&#xff0c;\u4f46\u5b83\u5b9e\u9645\u4e0a\u662f\u4e00\u79cd\u7528\u4e8e\u89e3\u51b3\u4e8c\u5206\u7c7b\u95ee\u9898\u7684\u7ebf\u6027\u6a21\u578b\u3002\u5b83\u7684\u6838\u5fc3\u601d\u60f3\u662f&#xff1a;\u5148\u62df\u5408\u51b3\u7b56\u8fb9\u754c&amp;\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.wsisp.com\/helps\/78647.html\" \/>\n<meta property=\"og:site_name\" content=\"\u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3\" \/>\n<meta property=\"article:published_time\" content=\"2026-02-27T14:32:13+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/www.wsisp.com\/helps\/wp-content\/uploads\/2026\/02\/20260227143209-69a1aae98600a.png\" \/>\n<meta name=\"author\" content=\"admin\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"\u4f5c\u8005\" \/>\n\t<meta name=\"twitter:data1\" content=\"admin\" \/>\n\t<meta name=\"twitter:label2\" content=\"\u9884\u8ba1\u9605\u8bfb\u65f6\u95f4\" \/>\n\t<meta name=\"twitter:data2\" content=\"5 \u5206\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\/\/www.wsisp.com\/helps\/78647.html\",\"url\":\"https:\/\/www.wsisp.com\/helps\/78647.html\",\"name\":\"\u673a\u5668\u5b66\u4e60\u4e4b\u903b\u8f91\u56de\u5f52 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3\",\"isPartOf\":{\"@id\":\"https:\/\/www.wsisp.com\/helps\/#website\"},\"datePublished\":\"2026-02-27T14:32:13+00:00\",\"dateModified\":\"2026-02-27T14:32:13+00:00\",\"author\":{\"@id\":\"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/358e386c577a3ab51c4493330a20ad41\"},\"breadcrumb\":{\"@id\":\"https:\/\/www.wsisp.com\/helps\/78647.html#breadcrumb\"},\"inLanguage\":\"zh-Hans\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/www.wsisp.com\/helps\/78647.html\"]}]},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/www.wsisp.com\/helps\/78647.html#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"\u9996\u9875\",\"item\":\"https:\/\/www.wsisp.com\/helps\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"\u673a\u5668\u5b66\u4e60\u4e4b\u903b\u8f91\u56de\u5f52\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/www.wsisp.com\/helps\/#website\",\"url\":\"https:\/\/www.wsisp.com\/helps\/\",\"name\":\"\u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3\",\"description\":\"\u9999\u6e2f\u670d\u52a1\u5668_\u9999\u6e2f\u4e91\u670d\u52a1\u5668\u8d44\u8baf_\u670d\u52a1\u5668\u5e2e\u52a9\u6587\u6863_\u670d\u52a1\u5668\u6559\u7a0b\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/www.wsisp.com\/helps\/?s={search_term_string}\"},\"query-input\":\"required name=search_term_string\"}],\"inLanguage\":\"zh-Hans\"},{\"@type\":\"Person\",\"@id\":\"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/358e386c577a3ab51c4493330a20ad41\",\"name\":\"admin\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"zh-Hans\",\"@id\":\"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/gravatar.wp-china-yes.net\/avatar\/?s=96&d=mystery\",\"contentUrl\":\"https:\/\/gravatar.wp-china-yes.net\/avatar\/?s=96&d=mystery\",\"caption\":\"admin\"},\"sameAs\":[\"http:\/\/wp.wsisp.com\"],\"url\":\"https:\/\/www.wsisp.com\/helps\/author\/admin\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"\u673a\u5668\u5b66\u4e60\u4e4b\u903b\u8f91\u56de\u5f52 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3","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.wsisp.com\/helps\/78647.html","og_locale":"zh_CN","og_type":"article","og_title":"\u673a\u5668\u5b66\u4e60\u4e4b\u903b\u8f91\u56de\u5f52 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3","og_description":"\u6982\u8981\u4ecb\u7ecd\u5bf9\u4e8e\u903b\u8f91\u56de\u5f52\u6211\u4eec\u53ef\u4ee5\u4ece\u4e00\u4e2a\u95ee\u9898\u5165\u624b,\u5230\u5e95\u4ec0\u4e48\u662f\u903b\u8f91\u56de\u5f52\u7684\u7b97\u6cd5\u601d\u60f3?\u9762\u8bd5\u5b98\u95ee\u201c\u8bf7\u63cf\u8ff0\u903b\u8f91\u56de\u5f52\u7684\u7b97\u6cd5\u601d\u60f3\u201d\u8fd9\u4e2a\u95ee\u9898\u65f6&#xff0c;\u5176\u5b9e\u662f\u5728\u8003\u5bdf\u4f60\u5bf9\u8fd9\u4e2a\u57fa\u7840\u6a21\u578b\u7684 \u672c\u8d28\u7406\u89e3\u2014\u2014\u662f\u53ea\u4f1a\u8c03\u7528sklearn.learn.LogisticRegression&#xff0c;\u8fd8\u662f\u771f\u6b63\u660e\u767d\u5b83\u4e3a\u4ec0\u4e48\u53eb\u201c\u56de\u5f52\u201d\u5374\u505a\u5206\u7c7b&#xff0c;\u5b83\u7684\u6838\u5fc3\u5728\u7b97\u4ec0\u4e48\u3002\u201c\u903b\u8f91\u56de\u5f52\u867d\u7136\u540d\u5b57\u91cc\u6709\u2018\u56de\u5f52\u2019&#xff0c;\u4f46\u5b83\u5b9e\u9645\u4e0a\u662f\u4e00\u79cd\u7528\u4e8e\u89e3\u51b3\u4e8c\u5206\u7c7b\u95ee\u9898\u7684\u7ebf\u6027\u6a21\u578b\u3002\u5b83\u7684\u6838\u5fc3\u601d\u60f3\u662f&#xff1a;\u5148\u62df\u5408\u51b3\u7b56\u8fb9\u754c&amp;","og_url":"https:\/\/www.wsisp.com\/helps\/78647.html","og_site_name":"\u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3","article_published_time":"2026-02-27T14:32:13+00:00","og_image":[{"url":"https:\/\/www.wsisp.com\/helps\/wp-content\/uploads\/2026\/02\/20260227143209-69a1aae98600a.png"}],"author":"admin","twitter_card":"summary_large_image","twitter_misc":{"\u4f5c\u8005":"admin","\u9884\u8ba1\u9605\u8bfb\u65f6\u95f4":"5 \u5206"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/www.wsisp.com\/helps\/78647.html","url":"https:\/\/www.wsisp.com\/helps\/78647.html","name":"\u673a\u5668\u5b66\u4e60\u4e4b\u903b\u8f91\u56de\u5f52 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3","isPartOf":{"@id":"https:\/\/www.wsisp.com\/helps\/#website"},"datePublished":"2026-02-27T14:32:13+00:00","dateModified":"2026-02-27T14:32:13+00:00","author":{"@id":"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/358e386c577a3ab51c4493330a20ad41"},"breadcrumb":{"@id":"https:\/\/www.wsisp.com\/helps\/78647.html#breadcrumb"},"inLanguage":"zh-Hans","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.wsisp.com\/helps\/78647.html"]}]},{"@type":"BreadcrumbList","@id":"https:\/\/www.wsisp.com\/helps\/78647.html#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"\u9996\u9875","item":"https:\/\/www.wsisp.com\/helps"},{"@type":"ListItem","position":2,"name":"\u673a\u5668\u5b66\u4e60\u4e4b\u903b\u8f91\u56de\u5f52"}]},{"@type":"WebSite","@id":"https:\/\/www.wsisp.com\/helps\/#website","url":"https:\/\/www.wsisp.com\/helps\/","name":"\u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3","description":"\u9999\u6e2f\u670d\u52a1\u5668_\u9999\u6e2f\u4e91\u670d\u52a1\u5668\u8d44\u8baf_\u670d\u52a1\u5668\u5e2e\u52a9\u6587\u6863_\u670d\u52a1\u5668\u6559\u7a0b","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.wsisp.com\/helps\/?s={search_term_string}"},"query-input":"required name=search_term_string"}],"inLanguage":"zh-Hans"},{"@type":"Person","@id":"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/358e386c577a3ab51c4493330a20ad41","name":"admin","image":{"@type":"ImageObject","inLanguage":"zh-Hans","@id":"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/image\/","url":"https:\/\/gravatar.wp-china-yes.net\/avatar\/?s=96&d=mystery","contentUrl":"https:\/\/gravatar.wp-china-yes.net\/avatar\/?s=96&d=mystery","caption":"admin"},"sameAs":["http:\/\/wp.wsisp.com"],"url":"https:\/\/www.wsisp.com\/helps\/author\/admin"}]}},"_links":{"self":[{"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/posts\/78647","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/comments?post=78647"}],"version-history":[{"count":0,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/posts\/78647\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/media\/78638"}],"wp:attachment":[{"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/media?parent=78647"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/categories?post=78647"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/tags?post=78647"},{"taxonomy":"topic","embeddable":true,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/topic?post=78647"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}