{"id":3729,"date":"2022-04-19T05:57:27","date_gmt":"2022-04-19T14:27:27","guid":{"rendered":"http:\/\/10ccc.cc\/?p=3729"},"modified":"2026-01-02T20:28:22","modified_gmt":"2026-01-02T12:28:22","slug":"%e3%80%90flag%e3%80%91%e6%9c%ba%e5%99%a8%e5%ad%a6%e4%b9%a0100%e5%a4%a9-%e7%ac%ac3%e5%a4%a9-%e5%a4%9a%e5%85%83%e7%ba%bf%e6%80%a7%e5%9b%9e%e5%bd%92","status":"publish","type":"post","link":"https:\/\/10ccc.cc\/?p=3729","title":{"rendered":"\u3010Flag\u3011\u673a\u5668\u5b66\u4e60100\u5929-[\u7b2c3\u5929]-\u591a\u5143\u7ebf\u6027\u56de\u5f52"},"content":{"rendered":"

\"\"<\/p>\n

 <\/p>\n

\u6709\u4e86\u4e00\u5143\u7ebf\u6027\u56de\u5f52\u8bd5\u9a8c(day_2<\/a>)\u57fa\u7840\u4e4b\u540e, \u6211\u4eec\u5f97\u4ee5\u5c1d\u8bd5\u8f83\u4e3a\u590d\u6742\u591a\u5143\u7ebf\u6027\u56de\u5f52\u8bd5\u9a8c\u3002\u5728day_3\u4e2d\uff0c\u6211\u4eec\u5c06\u6839\u636e\u7ed9\u5b9a\u7684\u6570\u636e\u96c650_Startups.csv<\/a>\uff0c\u901a\u8fc7\u5176\u4e2d\u7684\u81ea\u53d8\u91cf R&D Spend\u3001 Administration\u3001Marketing Spend \u3001 State\u5b57\u6bb5\u9884\u6d4b\u56e0\u53d8\u91cfProfits\u7684\u3002<\/p><\/blockquote>\n

 <\/p>\n

\u672c\u6587\u5c06\u6db5\u76d6\u4ee5\u4e0b\u5185\u5bb9<\/strong>\uff1a<\/p>\n

    \n
  1. \u00a0 \u00a0 \u00a0 \u00a0 \u00a0day_3\u6e90\u7801\u5b66\u4e60\u53ca\u9010\u6b65\u5206\u6790\uff1b<\/li>\n
  2. \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\u4e3a\u4f55\u53ca\u5982\u4f55\u907f\u514dDummy Variable Trap\uff1b<\/li>\n
  3. \u00a0 \u00a0 \u00a0 \u00a0 \u00a0one-hot\u53ca\u5176\u4ed6\u5e38\u7528\u7684\u7f16\u7801\u65b9\u5f0f \uff1b<\/li>\n
  4. \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\u591a\u5143\u7ebf\u6027\u56de\u5f52\u7684\u76f8\u5173\u77e5\u8bc6\u8865\u5145;<\/li>\n
  5. \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\u4f7f\u7528excel\u5b8c\u6210day_3\u7684\u56de\u5f52\u5206\u6790\u4efb\u52a1\uff1b<\/li>\n<\/ol>\n

     <\/p>\n

    day_3\u5185\u5bb9\u5b66\u4e60\u53ca\u9010\u6b65\u5206\u6790<\/h2>\n
    \"\u6570\u636e\u96c6\"
    \u6570\u636e\u96c6<\/figcaption><\/figure>\n

    \u6b64\u90e8\u5206\u5185\u5bb9\u4e0e\u3010Flag\u3011\u673a\u5668\u5b66\u4e60100\u5929-[\u7b2c2\u5929]-\u7ebf\u6027\u56de\u5f52<\/a> \u4e2d\u4ee3\u7801\u7684\u90e8\u5206\u522b\u65e0\u4e8c\u81f4\uff0c\u56e0\u6b64\uff0c\u5982\u975e\u5fc5\u8981\uff0c\u5c06\u4ec5\u505a\u7b80\u7565\u7684\u6b65\u9aa4\u8bf4\u660e\u3002<\/p>\n

    \u9996\u5148\uff0c\u6211\u4eec\u901a\u8fc7\u4e0b\u65b9\u547d\u4ee4\uff0c\u5c06\u8bfb\u53d6\u6570\u636e\u96c6\u5e76\u5c06\u5176\u62c6\u5206\u6210\u8bad\u7ec3\u96c6<\/strong>\u4ee5\u53ca\u6d4b\u8bd5\u96c6<\/strong>\u3002\u5e76\u67e5\u770b\u62c6\u5206\u540e\u7684\u7ed3\u679c\u3002<\/p>\n

    \n\t\t\t
    \n
    \n
    \n
    <\/div>\n
    <\/div>\n
    <\/div>\n <\/div>\n
    \n \n Copy Code<\/span>\n Copied<\/span>\n Use a different Browser<\/span><\/a>\n <\/div>\n <\/div>\n 
    <\/p>\n
    import pandas as pd\r\nimport numpy as np\r\n\r\n\r\ndataset = pd.read_csv('50_Startups.csv')\r\nX = dataset.iloc[ : , :-1].values\r\nY = dataset.iloc[ : , 4 ].values\r\n# print(\"X:\",X, \"\\n\")\r\n# print(\"Y:\",Y, \"\\n\")<\/pre>\n
    <\/code><\/pre>\n\t\t\t<\/div>\n        <\/div>\n
    \u89e3\u6790\u5206\u7c7b\u6570\u636e\uff0c\u5e76\u5bf9\u7b2c\u56db\u5217\u8fdb\u884conehot\u7f16\u7801<\/strong><\/p>\n
    \n\t\t\t
    \n                
    \n                    
    \n                        
    <\/div>\n                        
    <\/div>\n                        
    <\/div>\n                    <\/div>\n                    
    \n                        \n                        Copy Code<\/span>\n                        Copied<\/span>\n                        Use a different Browser<\/span><\/a>\n                    <\/div>\n                <\/div>\n                
    <\/p>\n
    from sklearn.preprocessing import LabelEncoder, OneHotEncoder\r\n\r\n\r\nlabelencoder = LabelEncoder()\r\nX[: , 3] = labelencoder.fit_transform(X[ : , 3])\r\nonehotencoder = OneHotEncoder(categorical_features = [3])\r\nX = onehotencoder.fit_transform(X).toarray()\r\nprint(X)<\/pre>\n
    <\/code><\/pre>\n\t\t\t<\/div>\n        <\/div>\n
    \"\"<\/p>\n
     <\/p>\n
    Avoid Dummy Variable Trap\uff1a\u672c\u6587\u4e2d\u4e0e\u524d\u7bc7\u4e2d\u4ee3\u7801\u6700\u5927\u7684\u4e0d\u540c\u5c31\u5728\u4e8e\u6b64\u3002\u901a\u8fc7 X = X[: , 1:] \u5c06\u7ecf\u8fc7one-hot encode\u7684 state\u53d8\u91cf \u622a\u53bb\u9996\u4f4d\u6765\u907f\u514d\u865a\u62df\u53d8\u91cf\u9677\u9631\u3002\u5728\u672c\u6587\u63a5\u4e0b\u6765\u7684 \u4e3a\u4f55\u53ca\u5982\u4f55\u907f\u514dDummy Variable Trap \u90e8\u5206\uff0c\u6211\u4eec\u5c06\u5c1d\u8bd5\u9610\u8ff0\u865a\u62df\u53d8\u91cf\u4ea7\u751f\u7684\u539f\u56e0\u53ca\u5982\u4f55\u907f\u514d\u865a\u62df\u53d8\u91cf\u9677\u9631\u3002<\/p>\n
    \n\t\t\t
    \n                
    \n                    
    \n                        
    <\/div>\n                        
    <\/div>\n                        
    <\/div>\n                    <\/div>\n                    
    \n                        \n                        Copy Code<\/span>\n                        Copied<\/span>\n                        Use a different Browser<\/span><\/a>\n                    <\/div>\n                <\/div>\n                
    <\/p>\n
    X = X[: , 1:]\r\nprint(X)<\/pre>\n
    <\/code><\/pre>\n\t\t\t<\/div>\n        <\/div>\n
     <\/p>\n
    \u62c6\u5206\u8bad\u7ec3\u96c6<\/strong>\u4ee5\u53ca\u6d4b\u8bd5\u96c6<\/strong>\u3002\u5e76\u67e5\u770b\u62c6\u5206\u540e\u7684\u7ed3\u679c\u3002<\/p>\n
    \n\t\t\t
    \n                
    \n                    
    \n                        
    <\/div>\n                        
    <\/div>\n                        
    <\/div>\n                    <\/div>\n                    
    \n                        \n                        Copy Code<\/span>\n                        Copied<\/span>\n                        Use a different Browser<\/span><\/a>\n                    <\/div>\n                <\/div>\n                
    <\/p>\n
    from sklearn.model_selection import train_test_split\r\nX_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size = 0.2, random_state = 0)<\/pre>\n
    <\/code><\/pre>\n\t\t\t<\/div>\n        <\/div>\n
     <\/p>\n
    \u4f7f\u7528\u8bad\u7ec3\u96c6X_train, Y_train \u8fdb\u884c\u7ebf\u6027\u56de\u5f52<\/p>\n
    \n\t\t\t
    \n                
    \n                    
    \n                        
    <\/div>\n                        
    <\/div>\n                        
    <\/div>\n                    <\/div>\n                    
    \n                        \n                        Copy Code<\/span>\n                        Copied<\/span>\n                        Use a different Browser<\/span><\/a>\n                    <\/div>\n                <\/div>\n                
    <\/p>\n
    from sklearn.linear_model import LinearRegression\r\n\r\nregressor = LinearRegression()\r\nregressor.fit(X_train, Y_train)<\/pre>\n
    <\/code><\/pre>\n\t\t\t<\/div>\n        <\/div>\n
     <\/p>\n
    \u4f7f\u7528\u6d4b\u8bd5\u96c6Y_test, X_test \u6570\u636e\u68c0\u9a8c\u56de\u5f52\u65b9\u7a0bregressor<\/p>\n
    \n\t\t\t
    \n                
    \n                    
    \n                        
    <\/div>\n                        
    <\/div>\n                        
    <\/div>\n                    <\/div>\n                    
    \n                        \n                        Copy Code<\/span>\n                        Copied<\/span>\n                        Use a different Browser<\/span><\/a>\n                    <\/div>\n                <\/div>\n                
    <\/p>\n
    from sklearn.linear_model import LinearRegression\r\n\r\ny_pred = regressor.predict(X_test)\r\nprint(y_pred)\r\nprint(Y_test)<\/pre>\n
    <\/code><\/pre>\n\t\t\t<\/div>\n        <\/div>\n
    \n
    \"\u6d4b\u8bd5\u96c6\u6570\u636e\u9a8c\u8bc1\"
    \u6d4b\u8bd5\u96c6\u6570\u636e\u9a8c\u8bc1<\/figcaption><\/figure><\/blockquote>\n
     <\/p>\n
     <\/p>\n
    \u00a0\u4e3a\u4f55\u53ca\u5982\u4f55\u907f\u514dDummy Variable Trap(\u865a\u62df\u53d8\u91cf\u9677\u9631)<\/h2>\n
    \u4ea7\u751f\u539f\u56e0<\/h3>\n
    \u5f53\u6211\u4eec\u5c06\u4e00\u4e2a\u53d8\u91cf\u8fdb\u884cone-hot\u7f16\u7801\u4e4b\u540e\uff0c\u4f1a\u4ea7\u751f\u591a\u4e2a\u865a\u62df\u53d8\u91cf\u3002\u800c\u5f53\u4ea7\u751f\u7684\u865a\u62df\u53d8\u91cf\u4e4b\u95f4\u5b58\u5728\u7ebf\u6027\u5173\u7cfb\u65f6(multi-colinear)\uff0c\u5373\u53ef\u4ee5\u901a\u8fc7\u4e00\u4e2a\u53d8\u91cf\u53ef\u4ee5\u9884\u6d4b\u5230\u53e6\u4e00\u4e2a\u53d8\u91cf\u3002multi-colinear\u4f1a\u8ba9\u6211\u4eec\u5bf9\u56de\u5f52\u7cfb\u6570(coefficient)\u8ba1\u7b97\u7ed3\u679c\u53d8\u5f97\u4e0d\u7a33\u5b9a\u3002<\/p>\n
    \u6211\u4eec\u7528\u56fe\u5f62\u6765\u89e3\u91ca\u3002<\/p>\n
    \"multi-colinear\u7684\u56fe\u5f62\u89e3\u91ca\"
    multi-colinear\u7684\u56fe\u5f62\u89e3\u91ca(Axure\u7ed8\u5236)<\/figcaption><\/figure>\n
     <\/p>\n
    \u6211\u4eec\u5047\u8bbe\u53ea\u6709\u4e24\u4e2a\u81ea\u53d8\u91cf\u5411\u91cf(\u72ec\u70ed\u7f16\u7801\u540e\u7684\u53d8\u91cf) X1\u3001 X2<\/strong>\u3002 Y<\/strong>\u4e3a\u56e0\u53d8\u91cf\uff0c\u90a3\u4e48 Y1<\/strong> \u5373\u4e3aPredictions\u3002 \u5f53X1\u3001 X2<\/strong>\u65e0\u7ebf\u6027\u5173\u7cfb\u65f6\uff0c\u5219\u4e24\u4e2a\u5411\u91cf\u4e4b\u95f4\u5b58\u5728\u5939\u89d2\uff0c\u53ef\u786e\u5b9a\u552f\u4e00\u7684\u4e00\u4e2a\u6295\u5f71\uff0c\u8fdb\u800c\u53ef\u786e\u5b9a\u552f\u4e00\u5e73\u9762(non-linear - \u7ea2\u8272);\u800c\u5f53 X1\u3001 X2 <\/strong>\u5b58\u5728\u7ebf\u6027\u5173\u7cfb\u65f6\uff0c\u7531\u4e8e\u5411\u91cf\u5171\u7ebf\u3002\u56e0\u800c\u65e0\u6cd5\u786e\u5b9a\u4e00\u4e2a\u5e73\u9762(multi-colinear - \u7ea2\u8272\/\u7eff\u8272) \uff0c\u8fd9\u5c06\u5bfc\u81f4 Y1<\/strong>\u00a0\u4e0d\u786e\u5b9a\uff0c\u5373coefficient \u03b2<\/strong>\u4e0d\u7a33\u5b9a\u3002<\/p>\n
     <\/p>\n
     <\/p>\n
    \u5982\u679c\u4ece\u4ee3\u6570\u65b9\u9762\u770b\uff0c\u57fa\u4e8e\u6700\u5c0f\u4e8c\u4e58\u6cd5\uff0ccoefficient \u03b2\u53ef\u4ee5\u89c6\u4e3a\u4e0b\u8ff0\u516c\u5f0f\u8ba1\u7b97<\/p>\n
    \"\"<\/p>\n
    \"\"\u53ef\u4ee5\u770b\u4f5c\u81ea\u53d8\u91cf\"\"\u7684\u534f\u65b9\u5dee\u77e9\u9635<\/a>\u3002\u5982\u679c\u6709\u4e24\u4e2a\u81ea\u53d8\u91cf\u5b58\u5728\u5b8c\u5168\u7ebf\u6027\u5173\u7cfb\uff0c\u534f\u65b9\u5dee\u77e9\u9635\u4e2d\u5c31\u4f1a\u6709\u4e24\u5217\u53d8\u91cf\u5411\u91cf\u8868\u73b0\u7ebf\u6027\u5173\u7cfb(\u53ef\u4e92\u76f8\u9884\u6d4b)\uff0c\u8fd9\u5c06\u5bfc\u81f4\u534f\u65b9\u5dee\u77e9\u9635\u4e0d\u53ef\u9006(\u65e0\u9006\u77e9\u9635)\u3002\u800c\u5728\u4e0a\u8ff0\u516c\u5f0f\u4e2d\uff0c\u5b58\u5728\u6c42\u9006\u77e9\u9635\u7684\u8ba1\u7b97\u8fc7\u7a0b(\"\")\uff0c\u4e24\u8005\u4e92\u76f8\u77db\u76fe\uff0c\u6240\u4ee5\u6211\u4eec\u65e0\u6cd5\u627e\u5230\u89e3\u00a0 \u03b2\u3002<\/p>\n
     <\/p>\n
    \u6ce8\u610f\uff1a<\/p>\n
      \n
    1. \u77e9\u9635\u4e58\u6cd5\u3001\u77e9\u9635\u7684\u9006\u3001\u4ee5\u53catranspose\u7684\u8ba1\u7b97\u65b9\u6cd5\u53ef\u53c2\u8003 \u7ebf\u6027\u4ee3\u6570\u57fa\u7840<\/a>\u3002<\/li>\n
    2. \u4ee3\u7801\u89e3\u91ca\u53ef\u53c2\u8003Dummy Variable Trap<\/a><\/li>\n
    3. A\u77e9\u9635\u4e0d\u53ef\u9006
      \n<=> |A| = 0
      \n<=> A\u7684\u5217(\u884c)\u5411\u91cf\u7ec4\u7ebf\u6027\u76f8\u5173
      \n<=> R(A)<n
      \n<=> AX=0 \u6709\u975e\u96f6\u89e3
      \n<=> A\u6709\u7279\u5f81\u503c0.
      \n<=> A\u4e0d\u80fd\u8868\u793a\u6210\u521d\u7b49\u77e9\u9635\u7684\u4e58\u79ef
      \n<=> A\u7684\u7b49\u4ef7\u6807\u51c6\u5f62\u4e0d\u662f\u5355\u4f4d\u77e9\u9635<\/li>\n<\/ol>\n<\/blockquote>\n
       <\/p>\n
      \u5904\u7406\u65b9\u5f0f<\/h3>\n
      \u5982\u679c\u4e00\u4e2a\u53d8\u91cf\u7ecf\u8fc7one-hot encode\u5f97\u5230\u7684\u865a\u62df\u53d8\u91cf\u7c7b\u522b\u6709\u00a0 n\u00a0 \u4f4d\uff0c\u5219\u53ef\u4ee5\u4ec5\u4f7f\u7528\u00a0 n-1\u00a0 \u4e2a\u865a\u62df\u53d8\u91cf\uff0c\u90a3\u4e48\u5c06\u7ecf\u8fc7one-hot encode\u7684 state\u53d8\u91cf \u622a\u53bb\u9996\u4f4d\uff0c\u5373\u53ef\u907f\u514dmulti-colinearity\u3002<\/p>\n
       <\/p>\n
      One-hot\u53ca\u5176\u4ed6\u5e38\u7528\u7684\u7f16\u7801\u65b9\u5f0f<\/h2>\n
      Day 3\u7684\u5b9e\u8df5\uff0c\u5728\u4f7f\u7528one-hot encoding\u65f6\uff0c\u53ef\u4ee5\u53d1\u73b0\u5176\u5b58\u5728\u7684\u4e00\u4e9b\u95ee\u9898\uff1a<\/p>\n
      \n
        \n
      1. \u5f53\u9700\u8981\u7f16\u7801\u7684\u53d8\u91cf\u8f83\u591a\u65f6\uff0c\u5373\u5bb9\u6613\u5927\u5e45\u6570\u636e\u7684\u7ef4\u5ea6\uff0c\u589e\u52a0\u8fd0\u7b97\u91cf\uff1b<\/li>\n
      2. \u5bb9\u6613\u4ea7\u751f\u56e0multi-colinearity\u5bfc\u81f4\u7684coefficient\u4e0d\u7a33\u5b9a\uff1b<\/li>\n
      3. \u90e8\u5206\u7b97\u6cd5\u9700\u8981\u5bf9\u53d8\u91cf\u6570\u503c\u654f\u611f\uff0c\u6b64\u65f6\u4e0d\u9002\u7528one-hot encoding;<\/li>\n<\/ol>\n<\/blockquote>\n
        \u5982\u679c\u6211\u4eec\u8fdb\u884c\u601d\u8003\uff0c\u6211\u4eec\u5927\u6982\u7387\u4f1a\u63d0\u51fa\u95ee\u9898\uff1a\u5e38\u7528\u7684\u7f16\u7801\u65b9\u5f0f\u6709\u54ea\u4e9b\uff1f\u9002\u7528\u573a\u666f\u662f\u54ea\u4e9b\uff1f\u6bcf\u79cd\u7f16\u7801\u65b9\u5f0f\u9002\u7528\u7684\u81ea\u53d8\u91cf\u7c7b\u578b\uff1f\u6211\u4eec\u4eceday 1 \u81f3\u4eca \u6211\u4eec\u5728\u5b66\u4e60\u5185\u5bb9\u4e2d\u9047\u5230\u8fc7\u7684\u90e8\u5206\u7f16\u7801\u65b9\u5f0f\u8fdb\u884c\u89e3\u8bfb\u3002\u540e\u7eed\u5c06\u6301\u7eed\u66f4\u65b0\u3002<\/p>\n
         <\/p>\n
        One-hot Encoding<\/h3>\n
        \u901a\u5e38\u7528\u4e8e\u4e0d\u5177\u5907\u5927\u5c0f\u3001\u987a\u5e8f\u7684\u533a\u5206\u7684 \u5b9a\u7c7b\u53d8\u91cf<\/strong> \u7684\u7f16\u7801\u3002\u5176\u7279\u70b9\u662f\u662f\u8bb2\u5b9a\u7c7b\u53d8\u91cf\u8f6c\u6362\u6210\u75310 \u548c 1 \u7ec4\u6210\u7684\u5411\u91cf\u3002\u4f46\u7f3a\u70b9\u4e5f\u663e\u800c\u6613\u89c1\uff0c\u5373\u4e0a\u6587\u6240\u8ff0\u7684\u865a\u62df\u53d8\u91cf\u9677\u9631<\/p>\n
        \u5982\uff1a<\/p>\n
        \n\t\t\t
        \n                
        \n                    
        \n                        
        <\/div>\n                        
        <\/div>\n                        
        <\/div>\n                    <\/div>\n                    
        \n                        \n                        Copy Code<\/span>\n                        Copied<\/span>\n                        Use a different Browser<\/span><\/a>\n                    <\/div>\n                <\/div>\n                
        <\/p>\n
        import pandas as pd
        \nfrom sklearn.preprocessing import OneHotEncoder<\/p>\n
        categorical_df = pd.DataFrame({'my_id': ['101', '102', '103', '104'],
        \n'name': ['allen', 'bob', 'chartten', 'dory'],
        \n'place': ['third', 'second', 'first', 'second']})
        \nprint(categorical_df)
        \nprint('--'*20)<\/p>\n
        encoder = OneHotEncoder(categories=[['dory', 'bob', 'chartten', 'allen']])
        \nencoder.fit_transform(categorical_df[['name']]).toarray()<\/p>\n
        <\/code><\/pre>\n\t\t\t<\/div>\n        <\/div>\n
        \u6211\u4eec\u81ea\u5b9a\u4e49\u7684\u7f16\u7801\u987a\u5e8f\u662f['dory', 'bob', 'chartten', 'allen']\uff0cdory=[1,0,0,0], bob=[0,1,0,0], chartten=[0,0,1,0], allen=[0,0,0,1],\u00a0 \u5728\u8f93\u51fa\u7ed3\u679c\u7684\u7b2c\u4e00\u5217\u4e2d\u8868\u793aallen\uff0c\u5373\u4e3a[0,0,0,1]\uff0c\u7b2c\u4e8c\u5217\u8868\u793abob\uff0c\u5373\u4e3a[0,1,0,0]\uff0c\u4ee5\u6b64\u7c7b\u63a8\u3002<\/p>\n
         <\/p>\n
        Label Encoding<\/h3>\n
        Label Encoding\u662f\u7ed9\u67d0\u4e00\u5217\u6570\u636e\u7f16\u7801\uff0c\u901a\u5e38\u7528\u4e8e \u5b9a\u7eed\u53d8\u91cf <\/strong>\u7684\u7f16\u7801\u3002<\/p>\n
        \u503c\u5f97\u6ce8\u610f\u7684\u662f\uff0c\u5b9a\u5e8f\u53d8\u91cf \u548c \u5b9a\u7c7b\u53d8\u91cf\u4e00\u6837\uff0c\u4e5f\u662f\u5206\u7c7b\uff0c\u4f46\u6709\u6392\u5e8f\u903b\u8f91\u5173\u7cfb\uff0c\u7b49\u7ea7\u4e0a\u9ad8\u4e8e\u5b9a\u7c7b\u3002\u6bd4\u5982\uff0c\u5b66\u5386\u5206\u5c0f\u5b66\uff0c\u521d\u4e2d\uff0c\u9ad8\u4e2d\uff0c\u672c\u79d1\uff0c\u7814\u7a76\u751f\uff0c\u5404\u4e2a\u7c7b\u522b\u4e4b\u95f4\u5b58\u5728\u4e00\u5b9a\u7684\u903b\u8f91\uff0c\u663e\u7136\u7814\u7a76\u751f\u5b66\u5386\u662f\u6700\u9ad8\u7684\uff0c\u5c0f\u5b66\u6700\u4f4e\u3002\u8fd9\u65f6\u5019\u4f7f\u7528Label encoding\u4f1a\u663e\u5f97\u66f4\u5408\u9002\uff0c\u56e0\u4e3a\u81ea\u5b9a\u4e49\u7684\u6570\u5b57\u987a\u5e8f\u53ef\u4ee5\u4e0d\u7834\u574f\u539f\u6709\u903b\u8f91\uff0c\u5e76\u4e0e\u8fd9\u4e2a\u903b\u8f91\u76f8\u5bf9\u5e94\u3002 \u4f46\u4e00\u4e2a\u53d8\u91cf\u662f \u5b9a\u7c7b \u8fd8\u662f \u5b9a\u7eed\uff0c\u8fd9\u5f88\u5927\u7a0b\u5ea6\u4e0a\u53d6\u51b3\u4e8e\u4f60\u81ea\u8eab\u5bf9\u7814\u7a76\u7684\u7406\u89e3\uff0c\u56e0\u6b64\u6211\u4eec\u5e76\u4e0d\u80fd\u89c4\u5b9a\u54ea\u4e9b\u60c5\u51b5\u4e00\u5b9a\u8981\u5c06\u5176\u89e3\u91ca\u4e3a\u5b9a\u7eed\uff0c\u54ea\u4e9b\u60c5\u51b5\u53c8\u8981\u5c06\u5176\u89e3\u91ca\u4e3a\u5b9a\u7c7b\u3002<\/p>\n
         <\/p>\n
        \u53e6\u5916\uff0c\u7f16\u7801\u65b9\u5f0f\u8fd8\u4f1a\u5728\u4e00\u5b9a\u7a0b\u5ea6\u4e0a\u5f71\u54cd\u6570\u636e\u5728\u201c\u6570\u503c\u201d\u4e0a\u7684\u5267\u70c8\u7a0b\u5ea6\u3002<\/p>\n
        \u5982\uff0c\u6211\u4eec\u5bf9\u6570\u636e\u8fdb\u884c\u5982\u4e0b\u5904\u7406<\/p>\n
        \n\t\t\t
        \n                
        \n                    
        \n                        
        <\/div>\n                        
        <\/div>\n                        
        <\/div>\n                    <\/div>\n                    
        \n                        \n                        Copy Code<\/span>\n                        Copied<\/span>\n                        Use a different Browser<\/span><\/a>\n                    <\/div>\n                <\/div>\n                
        <\/p>\n
        import pandas as pd\r\nfrom sklearn.preprocessing import LabelEncoder\r\n\r\ncategorical_df = pd.DataFrame({'my_id': ['101', '102', '103', '104'],\r\n                                                             'name': ['allen', 'bob', 'chartten', 'dory'],\r\n                                                             'place': ['third', 'second', 'first', 'second'],\r\n                                                             })\r\nprint(categorical_df)\r\nprint('--' * 20)\r\n\r\nencoder = LabelEncoder()\r\ncategorical_df['place']=encoder.fit_transform(categorical_df['place'])\r\nprint(categorical_df)<\/pre>\n
        <\/code><\/pre>\n\t\t\t<\/div>\n        <\/div>\n
         <\/p>\n
        \u8f93\u51fa\u7ed3\u679c\uff1a<\/p>\n
        \"\"<\/p>\n
        \u6211\u4eec\u901a\u8fc7label encode \u5c06place \u7f16\u7801\u6210\u4e86 [0, 2]; \u800cone hot encode\u5219\u5c06\u5176\u7f16\u7801\u6210\u4e86 0 \u548c 1 \uff08\u6ce8\u610f\u5230\uff1a\u6211\u5728\u56fe\u4e2d\u5148\u5c06\u5176\u8fdb\u884clabel encode \u518d one hot encode\u662f\u7531\u4e8esklearn\u7248\u672c\u8f83\u4f4e\u539f\u56e0\uff1a\u8be5\u5217\u662fstring\u800c\u4f4e\u7248\u672csklearn \u8fdb\u884cone hot encode\u7684\u8981\u6c42\u8f93\u5165\u76842D array(-like)\uff09<\/p>\n
        \"\" \"\"<\/p>\n
        \u4f46\u4e0d\u53ef\u601d\u8bae\u7684\u662f\uff0c\u6211\u4eec\u4f7f\u7528\u7684\u7248\u672c\u786e\u5b9e\u662f 0.20.0 (day_0:\u5b89\u88c5sklearn<\/a>)\u3002\u800c\u4e14\u4e4b\u524d\uff0c\u4e5f\u786e\u5b9e\u53ef\u4ee5\u76f4\u63a5\u8fdb\u884c\u72ec\u70ed\u7f16\u7801\uff1b\u767e\u601d\u4e0d\u5f97\u5176\u89e3\u3002<\/p>\n
         <\/p>\n
         <\/p>\n
        \u4f7f\u7528\u6ce8\u610f\u4e8b\u9879\uff1a<\/strong><\/p>\n
        \u6240\u4f7f\u7528\u7684\u6a21\u578b\u5bf9\u6570\u503c\u5927\u5c0f\u654f\u611f\u7684\u6a21\u578b<\/strong>\uff1a\u5fc5\u987b\u4f7f\u7528one-hot encoding\u3002\u5178\u578b\u7684\u4f8b\u5b50\u5c31\u662fLR\u548cSVM\u3002\u4e8c\u8005\u7684\u635f\u5931\u51fd\u6570\u5bf9\u6570\u503c\u5927\u5c0f\u662f\u654f\u611f\u7684\uff0c\u5e76\u4e14\u53d8\u91cf\u95f4\u7684\u6570\u503c\u5927\u5c0f\u662f\u6709\u6bd4\u8f83\u610f\u4e49\u7684\u3002\u800cLabel encoding\u7684\u6570\u5b57\u7f16\u7801\u6ca1\u6709\u6570\u503c\u5927\u5c0f\u7684\u542b\u4e49\uff0c\u53ea\u662f\u4e00\u79cd\u6392\u5e8f\uff0c\u56e0\u6b64\u5bf9\u4e8e\u8fd9\u4e9b\u6a21\u578b\u90fd\u4f7f\u7528one-hot encoding\u3002<\/p>\n
        \u5bf9\u6570\u503c\u5927\u5c0f\u4e0d\u654f\u611f\u7684\u6a21\u578b\uff08\u5982\u6811\u6a21\u578b<\/strong>\uff09\uff1a\u4e0d\u5efa\u8bae\u4f7f\u7528one-hot encoding\u3002\u4e00\u822c\u8fd9\u7c7b\u6a21\u578b\u4e3a\u6811\u6a21\u578b\u3002\u5982\u679c\u5206\u7c7b\u7c7b\u522b\u7279\u522b\u591a\uff0c\u90a3\u4e48one-hot encoding\u4f1a\u5206\u88c2\u51fa\u5f88\u591a\u7279\u5f81\u53d8\u91cf\u3002\u8fd9\u65f6\u5019\uff0c\u5982\u679c\u6211\u4eec\u9650\u5236\u4e86\u6811\u6a21\u578b\u7684\u6df1\u5ea6\u800c\u4e0d\u80fd\u5411\u4e0b\u5206\u88c2\u7684\u8bdd\uff0c\u4e00\u4e9b\u7279\u5f81\u53d8\u91cf\u53ef\u80fd\u5c31\u56e0\u4e3a\u6a21\u578b\u65e0\u6cd5\u7ee7\u7eed\u5206\u88c2\u800c\u88ab\u820d\u5f03\u635f\u5931\u6389\u4e86\u3002\u56e0\u6b64\uff0c\u6b64\u79cd\u60c5\u51b5\u4e0b\u53ef\u4ee5\u8003\u8651\u4f7f\u7528Label encoding<\/p>\n
         <\/p>\n
        \u5f85\u540e\u7eed\u66f4\u65b0<\/h3>\n
         <\/p>\n
        \u591a\u5143\u7ebf\u6027\u56de\u5f52\u7684\u76f8\u5173\u77e5\u8bc6\u8865\u5145<\/h2>\n
        Pearson \u7cfb\u6570<\/h3>\n
        \u7edf\u8ba1\u5b66\u7684\u76f8\u5173\u7cfb\u6570\u7ecf\u5e38\u4f7f\u7528\u7684\u6709\u4e09\u79cd:\u76ae\u5c14\u68ee\uff08pearson\uff09\u76f8\u5173\u7cfb\u6570\u548c\u65af\u76ae\u5c14\u66fc\uff08spearman\uff09\u76f8\u5173\u7cfb\u6570\u548c\u80af\u5fb7\u5c14\uff08kendall\uff09\u76f8\u5173\u7cfb\u6570.\u76ae\u5c14\u68ee\u76f8\u5173\u7cfb\u6570\u662f\u8861\u91cf\u7ebf\u6027\u5173\u8054\u6027\u7684\u7a0b\u5ea6\uff0cp\u7684\u4e00\u4e2a\u51e0\u4f55\u89e3\u91ca\u662f\u5176\u4ee3\u8868\u4e24\u4e2a\u53d8\u91cf\u7684\u53d6\u503c\u6839\u636e\u5747\u503c\u96c6\u4e2d\u540e\u6784\u6210\u7684\u5411\u91cf\u4e4b\u95f4\u5939\u89d2\u7684\u4f59\u5f26.<\/p>\n
        \u76f8\u5173\u7cfb\u6570\uff1a\u8003\u5bdf\u4e24\u4e2a\u4e8b\u7269\uff08\u5728\u6570\u636e\u91cc\u6211\u4eec\u79f0\u4e4b\u4e3a\u53d8\u91cf\uff09\u4e4b\u95f4\u7684\u76f8\u5173\u7a0b\u5ea6\u3002<\/p>\n
        \u5982\u679c\u6709\u4e24\u4e2a\u53d8\u91cf\uff1aX\u3001Y\uff0c\u6700\u7ec8\u8ba1\u7b97\u51fa\u7684\u76f8\u5173\u7cfb\u6570\u7684\u542b\u4e49\u53ef\u4ee5\u6709\u5982\u4e0b\u7406\u89e3\uff1a<\/p>\n
        (1)\u3001\u5f53\u76f8\u5173\u7cfb\u6570\u4e3a0\u65f6\uff0cX\u548cY\u4e24\u53d8\u91cf\u65e0\u5173\u7cfb\u3002<\/p>\n
        (2)\u3001\u5f53X\u7684\u503c\u589e\u5927\uff08\u51cf\u5c0f\uff09\uff0cY\u503c\u589e\u5927\uff08\u51cf\u5c0f\uff09\uff0c\u4e24\u4e2a\u53d8\u91cf\u4e3a\u6b63\u76f8\u5173\uff0c\u76f8\u5173\u7cfb\u6570\u57280.00\u4e0e1.00\u4e4b\u95f4\u3002<\/p>\n
        (3)\u3001\u5f53X\u7684\u503c\u589e\u5927\uff08\u51cf\u5c0f\uff09\uff0cY\u503c\u51cf\u5c0f\uff08\u589e\u5927\uff09\uff0c\u4e24\u4e2a\u53d8\u91cf\u4e3a\u8d1f\u76f8\u5173\uff0c\u76f8\u5173\u7cfb\u6570\u5728-1.00\u4e0e0.00\u4e4b\u95f4\u3002<\/p>\n
         <\/p>\n
        \u76f8\u5173\u7cfb\u6570\u7684\u7edd\u5bf9\u503c\u8d8a\u5927\uff0c\u76f8\u5173\u6027\u8d8a\u5f3a\uff0c\u76f8\u5173\u7cfb\u6570\u8d8a\u63a5\u8fd1\u4e8e1\u6216-1\uff0c\u76f8\u5173\u5ea6\u8d8a\u5f3a\uff0c\u76f8\u5173\u7cfb\u6570\u8d8a\u63a5\u8fd1\u4e8e0\uff0c\u76f8\u5173\u5ea6\u8d8a\u5f31\u3002<\/p>\n
        \u901a\u5e38\u60c5\u51b5\u4e0b\u901a\u8fc7\u4ee5\u4e0b\u53d6\u503c\u8303\u56f4\u5224\u65ad\u53d8\u91cf\u7684\u76f8\u5173\u5f3a\u5ea6\uff1a
        \n\u76f8\u5173\u7cfb\u6570 0.8-1.0 \u6781\u5f3a\u76f8\u5173
        \n0.6-0.8 \u5f3a\u76f8\u5173
        \n0.4-0.6 \u4e2d\u7b49\u7a0b\u5ea6\u76f8\u5173
        \n0.2-0.4 \u5f31\u76f8\u5173
        \n0.0-0.2 \u6781\u5f31\u76f8\u5173\u6216\u65e0\u76f8\u5173<\/p><\/blockquote>\n
        \u76ae\u5c14\u68ee(pearson)\u76f8\u5173\u7cfb\u6570
        \n\u5728\u8fd9\u4e09\u5927\u76f8\u5173\u7cfb\u6570\u4e2d\uff0cspearman\u548ckendall\u5c5e\u4e8e\u7b49\u7ea7\u76f8\u5173\u7cfb\u6570\u4ea6\u79f0\u4e3a\u201c\u79e9\u76f8\u5173\u7cfb\u6570\u201d\uff0c\u662f\u53cd\u6620\u7b49\u7ea7\u76f8\u5173\u7a0b\u5ea6\u7684\u7edf\u8ba1\u5206\u6790\u6307\u6807\u3002\u4eca\u5929\u6682\u65f6\u7528\u4e0d\u5230\uff0c\u6240\u4ee5\u73b0\u5728\u53ea\u505apearson\u7684\u76f8\u5173\u7814\u7a76\u3002<\/p>\n
        \u5176\u8ba1\u7b97\u516c\u5f0f\u4e3a\uff1a<\/p>\n
        \"pearson
        pearson coefficient<\/figcaption><\/figure>\n
        \u4e3a\u65b9\u4fbf\u8ba1\u7b97\uff0c\u6211\u4eec\u4e00\u822c\u91c7\u7528\u7b2c\u4e8c\u4e2a\u516c\u5f0f\u3002\u53c2\u8003\u6211\u5927\u5b66\u8bfe\u672c<\/a>\u3002<\/p>\n
        \u56de\u5f52\u7cfb\u6570<\/h3>\n
        \u53c2\u8003\u4ee5\u4e0b\u94fe\u63a5\uff1a<\/p>\n
          \n
        1. What is a regression coefficient?<\/a><\/li>\n
        2. \u56de\u5f52\u7cfb\u6570<\/a><\/li>\n<\/ol>\n
           <\/p>\n
          \u4f7f\u7528excel\u5b8c\u6210day_3\u7684\u56de\u5f52\u5206\u6790\u4efb\u52a1<\/h2>\n
          \"\u4f7f\u7528excel\u8fdb\u884c\u56de\u5f52\u5206\u6790\"
          \u4f7f\u7528excel\u8fdb\u884c\u56de\u5f52\u5206\u6790<\/figcaption><\/figure>\n
          \u4e0b\u8f7d\u672c\u6b21\u5206\u6790\u6587\u4ef6\uff1a<\/p>\n50_Startups<\/a>\" rel=\"external\"\r\ntarget=\"_blank\" title=\"\u4e0b\u8f7d\u5730\u5740\">\r\n<\/i>Download<\/span><\/a>\n
           <\/p>\n
          \u6211\u4eec\u4f7f\u7528excel\u8fdb\u884c\u56de\u5f52\u5206\u6790\uff0c\u67e5\u770b\u5404\u4e2a\u53d8\u91cf\u5bf9\u6700\u7ec8profits\u7684\u5f71\u54cd\uff1a<\/p>\n