Do\u011fruluk<\/strong> : Pozitif tahminleri toplam pozitif tahminlere g\u00f6re d\u00fczeltin.<\/span><\/p>\n Bu \u015fu \u015fekilde hesaplan\u0131r:<\/span><\/p>\n Hat\u0131rlatma<\/strong> : Pozitif tahminlerin toplam ger\u00e7ek pozitiflere g\u00f6re d\u00fczeltilmesi<\/span><\/p>\n Bu \u015fu \u015fekilde hesaplan\u0131r:<\/span><\/p>\n Belirli bir modelin hassasiyetini ve geri \u00e7a\u011fr\u0131lmas\u0131n\u0131 g\u00f6rselle\u015ftirmek i\u00e7in bir hassasiyet-geri \u00e7a\u011f\u0131rma e\u011frisi<\/strong> olu\u015fturabiliriz.<\/span> Bu e\u011fri, farkl\u0131 e\u015fikler i\u00e7in kesinlik ve geri \u00e7a\u011f\u0131rma aras\u0131ndaki dengeyi g\u00f6sterir.<\/span> <\/p>\n A\u015fa\u011f\u0131daki ad\u0131m ad\u0131m \u00f6rnek, Python’da bir lojistik regresyon modeli i\u00e7in hassas geri \u00e7a\u011f\u0131rma e\u011frisinin nas\u0131l olu\u015fturulaca\u011f\u0131n\u0131 g\u00f6sterir.<\/span><\/p>\n \u0130lk \u00f6nce gerekli paketleri i\u00e7e aktaraca\u011f\u0131z:<\/span><\/p>\n Daha sonra bir veri k\u00fcmesi olu\u015fturaca\u011f\u0131z ve buna lojistik regresyon modelini yerle\u015ftirece\u011fiz:<\/span><\/p>\n Daha sonra modelin hassasiyetini ve geri \u00e7a\u011fr\u0131lmas\u0131n\u0131 hesaplay\u0131p bir hassasiyet-geri \u00e7a\u011f\u0131rma e\u011frisi olu\u015fturaca\u011f\u0131z:<\/span> <\/p>\n X ekseni geri \u00e7a\u011f\u0131rmay\u0131, y ekseni ise farkl\u0131 e\u015fikler i\u00e7in hassasiyeti g\u00f6sterir.<\/span><\/p>\n Hat\u0131rlama artt\u0131k\u00e7a kesinli\u011fin azald\u0131\u011f\u0131n\u0131 unutmay\u0131n.<\/span><\/p>\n Bu, iki \u00f6l\u00e7\u00fcm aras\u0131ndaki uzla\u015fmay\u0131 temsil eder. Modelimizin hat\u0131rlanabilirli\u011fini art\u0131rmak i\u00e7in hassasiyetin azalmas\u0131 veya bunun tersinin olmas\u0131 gerekir.<\/span><\/p>\n Python’da Lojistik Regresyon Nas\u0131l Ger\u00e7ekle\u015ftirilir<\/a> Makine \u00f6\u011freniminde s\u0131n\u0131fland\u0131rma modellerini kullan\u0131rken model kalitesini de\u011ferlendirmek i\u00e7in s\u0131kl\u0131kla kulland\u0131\u011f\u0131m\u0131z iki \u00f6l\u00e7\u00fcm hassasiyet ve hat\u0131rlamad\u0131r. Do\u011fruluk : Pozitif tahminleri toplam pozitif tahminlere g\u00f6re d\u00fczeltin. Bu \u015fu \u015fekilde hesaplan\u0131r: Do\u011fruluk = Ger\u00e7ek Pozitifler \/ (Ger\u00e7ek Pozitifler + Yanl\u0131\u015f Pozitifler) Hat\u0131rlatma : Pozitif tahminlerin toplam ger\u00e7ek pozitiflere g\u00f6re d\u00fczeltilmesi Bu \u015fu \u015fekilde hesaplan\u0131r: Hat\u0131rlatma = Do\u011fru […]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[11],"tags":[],"class_list":["post-2171","post","type-post","status-publish","format-standard","hentry","category-rehber"],"yoast_head":"\n\n
\n
![\"Python'da](\"https:\/\/statorials.org\/wp-content\/uploads\/2023\/08\/precisionrecall2.png\")
<\/p>\n 1. Ad\u0131m: Paketleri i\u00e7e aktar\u0131n
<\/strong><\/span><\/h3>\n from<\/span> sklearn import<\/span> datasets\nfrom<\/span> sklearn. model_selection<\/span> import<\/span> train_test_split\nfrom<\/span> sklearn. linear_model<\/span> import<\/span> LogisticRegression\nfrom<\/span> sklearn. metrics<\/span> import<\/span> precision_recall_curve\nimport<\/span> matplotlib. pyplot<\/span> as<\/span> plt\n<\/strong><\/pre>\n Ad\u0131m 2: Lojistik regresyon modelini yerle\u015ftirin<\/strong><\/span><\/h3>\n
#create dataset with 5 predictor variables\n<\/span>X, y = datasets. make_classification<\/span> (n_samples= 1000<\/span> ,\n n_features= 4<\/span> ,\n n_informative= 3<\/span> ,\n n_redundant= 1<\/span> ,\n random_state= 0<\/span> )\n\n#split dataset into training and testing set\n<\/span>X_train, X_test, y_train, y_test = train_test_split(X, y, test_size= .3<\/span> , random_state= 0<\/span> )\n\n#fit logistic regression model to dataset\n<\/span>classifier = LogisticRegression()\nclassify. fit<\/span> (X_train, y_train)\n\n#use logistic regression model to make predictions\n<\/span>y_score = classify. predict_proba<\/span> (X_test)[:, 1<\/span> ]<\/strong><\/span><\/pre>\n 3. Ad\u0131m: Hassasiyet geri \u00e7a\u011f\u0131rma e\u011frisini olu\u015fturun<\/strong><\/span><\/h3>\n
#calculate precision and recall\n<\/span>precision, recall, thresholds = precision_recall_curve(y_test, y_score)\n\n#create precision recall curve\n<\/span>fig, ax = plt. subplots<\/span> ()\nax. plot<\/span> (recall, precision, color=' purple<\/span> ')\n\n#add axis labels to plot\n<\/span>ax. set_title<\/span> (' Precision-Recall Curve<\/span> ')\nax. set_ylabel<\/span> (' Precision<\/span> ')\nax. set_xlabel<\/span> (' Recall<\/span> ')\n\n#displayplot<\/span>\nplt. show<\/span> ()<\/strong><\/span> <\/pre>\n<\/p>\n Ek kaynaklar<\/strong><\/span><\/h3>\n
Python’da Kar\u0131\u015f\u0131kl\u0131k Matrisi Nas\u0131l Olu\u015fturulur<\/a>
ROC E\u011frisi Nas\u0131l Yorumlan\u0131r (\u00d6rneklerle)<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"