{"id":3010,"date":"2023-07-19T15:56:18","date_gmt":"2023-07-19T15:56:18","guid":{"rendered":"https:\/\/statorials.org\/tr\/sklearn-dogrusal-regresyon-ozeti\/"},"modified":"2023-07-19T15:56:18","modified_gmt":"2023-07-19T15:56:18","slug":"sklearn-dogrusal-regresyon-ozeti","status":"publish","type":"post","link":"https:\/\/statorials.org\/tr\/sklearn-dogrusal-regresyon-ozeti\/","title":{"rendered":"Scikit-learn'den regresyon modeli \u00f6zeti nas\u0131l al\u0131n\u0131r?"},"content":{"rendered":"

<\/p>\n

\n

\u00c7o\u011funlukla Python’da scikit-learn<\/a> kullan\u0131larak olu\u015fturulan bir regresyon modelinin \u00f6zetini \u00e7\u0131karmak isteyebilirsiniz.<\/span><\/p>\n

Ne yaz\u0131k ki, scikit-learn genellikle yaln\u0131zca tahmin ama\u00e7l\u0131<\/a> kullan\u0131ld\u0131\u011f\u0131ndan, bir regresyon modelinin \u00f6zetini analiz etmek i\u00e7in pek \u00e7ok yerle\u015fik i\u015flev sunmaz.<\/span><\/p>\n

Dolay\u0131s\u0131yla Python’da bir regresyon modelinin \u00f6zetini almak istiyorsan\u0131z iki se\u00e7ene\u011finiz vard\u0131r:<\/span><\/p>\n

1.<\/strong> Scikit-learn’in s\u0131n\u0131rl\u0131 i\u015flevlerini kullan\u0131n.<\/span><\/p>\n

2.<\/strong> Bunun yerine istatistiksel modelleri<\/a> kullan\u0131n.<\/span><\/p>\n

A\u015fa\u011f\u0131daki \u00f6rnekler, her y\u00f6ntemin pratikte a\u015fa\u011f\u0131daki pandalar DataFrame ile nas\u0131l kullan\u0131laca\u011f\u0131n\u0131 g\u00f6sterir:<\/span><\/p>\n

 import<\/span> pandas as<\/span> pd\n\n#createDataFrame\ndf = pd. DataFrame<\/span> ({' x1<\/span> ': [1, 2, 2, 4, 2, 1, 5, 4, 2, 4, 4],\n                   ' x2<\/span> ': [1, 3, 3, 5, 2, 2, 1, 1, 0, 3, 4],\n                   ' y<\/span> ': [76, 78, 85, 88, 72, 69, 94, 94, 88, 92, 90]})\n\n#view first five rows of DataFrame\n<\/span>df. head<\/span> ()\n\n       x1 x2 y\n0 1 1 76\n1 2 3 78\n2 2 3 85\n3 4 5 88\n4 2 2 72\n<\/span><\/span><\/strong><\/pre>\n
 Y\u00f6ntem 1: Scikit-Learn’den Regresyon Modeli \u00d6zetini Al\u0131n<\/strong><\/span><\/h3>\n
 Scikit-learn’i kullanarak \u00e7oklu do\u011frusal regresyon<\/a> modeline uyum sa\u011flamak i\u00e7in a\u015fa\u011f\u0131daki kodu kullanabiliriz:<\/span><\/p>\n
 from<\/span> sklearn. linear_model<\/span> import<\/span> LinearRegression\n\n#initiate linear regression model\n<\/span>model = LinearRegression()\n\n#define predictor and response variables\n<\/span>x, y = df[[' x1<\/span> ', ' x2<\/span> ']], df. y<\/span>\n\n#fit regression model\n<\/span>model. fit<\/span> (x,y)\n<\/span><\/span><\/strong><\/pre>\n
 #display regression coefficients and R-squared value of model<\/span>\nprint<\/span> (model. intercept_<\/span> , model. coef_<\/span> , model. score<\/span> (X, y))\n\n70.4828205704 [5.7945 -1.1576] 0.766742556527\n<\/span><\/span><\/strong><\/pre>\n
 Bu \u00e7\u0131kt\u0131y\u0131 kullanarak uygun regresyon modelinin denklemini yazabiliriz:<\/span><\/p>\n
 y = 70,48 + 5,79×1<\/sub> \u2013 1,16×2<\/sub><\/span><\/p>\n
 Ayr\u0131ca modelin R 2<\/sup> de\u011ferinin de 76,67 oldu\u011fu g\u00f6r\u00fclmektedir.<\/span><\/p>\n
 Bu, yan\u0131t de\u011fi\u015fkenindeki varyasyonun %76,67’sinin<\/strong> modeldeki iki yorday\u0131c\u0131 de\u011fi\u015fken taraf\u0131ndan a\u00e7\u0131klanabilece\u011fi anlam\u0131na gelmektedir.<\/span><\/p>\n
 Y\u00f6ntem 2: Statsmodels’ten regresyon modeli \u00f6zetini al\u0131n<\/strong><\/span><\/h3>\n
 Python’da bir regresyon modelinin \u00f6zetini \u00e7\u0131karmak istiyorsan\u0131z statsmodels<\/strong> paketini kullanmak en iyisidir.<\/span><\/p>\n
 A\u015fa\u011f\u0131daki kod, \u00f6nceki \u00f6rnekle ayn\u0131 \u00e7oklu do\u011frusal regresyon modeline uymak ve model \u00f6zetini \u00e7\u0131karmak i\u00e7in bu paketin nas\u0131l kullan\u0131laca\u011f\u0131n\u0131 g\u00f6sterir:<\/span><\/p>\n
 import<\/span> statsmodels. api<\/span> as<\/span> sm\n\n#define response variable\n<\/span>y = df[' y<\/span> ']\n\n#define predictor variables\n<\/span>x = df[[' x1<\/span> ', ' x2<\/span> ']]\n\n#add constant to predictor variables\n<\/span>x = sm. add_constant<\/span> (x)\n\n#fit linear regression model\n<\/span>model = sm. OLS<\/span> (y,x). fit<\/span> ()\n\n#view model summary\n<\/span>print<\/span> ( model.summary<\/span> ())\n\n                            OLS Regression Results                            \n==================================================== ============================\nDept. Variable: y R-squared: 0.767\nModel: OLS Adj. R-squared: 0.708\nMethod: Least Squares F-statistic: 13.15\nDate: Fri, 01 Apr 2022 Prob (F-statistic): 0.00296\nTime: 11:10:16 Log-Likelihood: -31.191\nNo. Comments: 11 AIC: 68.38\nDf Residuals: 8 BIC: 69.57\nDf Model: 2                                         \nCovariance Type: non-robust                                         \n==================================================== ============================\n                 coef std err t P>|t| [0.025 0.975]\n-------------------------------------------------- ----------------------------\nconst 70.4828 3.749 18.803 0.000 61.839 79.127\nx1 5.7945 1.132 5.120 0.001 3.185 8.404\nx2 -1.1576 1.065 -1.087 0.309 -3.613 1.298\n==================================================== ============================\nOmnibus: 0.198 Durbin-Watson: 1.240\nProb(Omnibus): 0.906 Jarque-Bera (JB): 0.296\nSkew: -0.242 Prob(JB): 0.862\nKurtosis: 2.359 Cond. No. 10.7\n==================================================== ============================\n<\/strong><\/span><\/pre>\n
 Regresyon katsay\u0131lar\u0131 ve R-kare de\u011ferinin scikit-learn taraf\u0131ndan hesaplananlarla e\u015fle\u015fti\u011fini unutmay\u0131n, ancak regresyon modeli i\u00e7in ayr\u0131ca bir s\u00fcr\u00fc ba\u015fka faydal\u0131 \u00f6l\u00e7\u00fcm\u00fcm\u00fcz de var.<\/span><\/p>\n
 \u00d6rne\u011fin, her bir tahmin de\u011fi\u015fkeni i\u00e7in p de\u011ferlerini g\u00f6rebiliriz:<\/span><\/p>\n
 Bu sonu\u00e7 yararl\u0131 olmas\u0131na ra\u011fmen, modelin genel F istatisti\u011fini<\/a> , bireysel regresyon katsay\u0131lar\u0131n\u0131n<\/a> p de\u011ferlerini ve modelin modele ne kadar iyi uydu\u011funu anlamam\u0131za yard\u0131mc\u0131 olabilecek di\u011fer yararl\u0131 \u00f6nlemleri hala bilmiyoruz. veri k\u00fcmesi. veri k\u00fcmesi.<\/span><\/p>\n
 Daha sonra modelden regresyon katsay\u0131lar\u0131n\u0131 ve modelin R-kare de\u011ferini<\/a> \u00e7\u0131karmak i\u00e7in a\u015fa\u011f\u0131daki kodu kullanabiliriz:<\/span><\/span><\/p>\n