<\/p>\n
\u0130statistiklerde ortalama mutlak hata<\/strong> (MAE), belirli bir modelin do\u011frulu\u011funu \u00f6l\u00e7menin bir yoludur. A\u015fa\u011f\u0131daki \u015fekilde hesaplan\u0131r:<\/span><\/p>\n MAE = (1\/n) * \u03a3|y i<\/sub> \u2013 x i<\/sub> |<\/span><\/p>\n Alt\u0131n:<\/span><\/p>\n Metrics<\/strong> paketindeki mae (ger\u00e7ek, tahmin edilen)<\/strong> fonksiyonunu kullanarak R’deki ortalama mutlak hatay\u0131 hesaplayabiliriz.<\/span><\/p>\n Bu e\u011fitimde bu \u00f6zelli\u011fin pratikte nas\u0131l kullan\u0131laca\u011f\u0131na dair iki \u00f6rnek verilmektedir.<\/span><\/p>\n A\u015fa\u011f\u0131daki kod, g\u00f6zlemlenen de\u011ferlerin bir vekt\u00f6r\u00fc ile tahmin edilen de\u011ferlerin bir vekt\u00f6r\u00fc aras\u0131ndaki ortalama mutlak hatan\u0131n nas\u0131l hesaplanaca\u011f\u0131n\u0131 g\u00f6sterir:<\/span><\/p>\n Ortalama mutlak hatan\u0131n (MAE) 1,909<\/strong> oldu\u011fu ortaya \u00e7\u0131k\u0131yor.<\/span><\/p>\n Bu bize g\u00f6zlemlenen de\u011ferler ile tahmin edilen de\u011ferler aras\u0131ndaki ortalama mutlak fark\u0131n 1,909 oldu\u011funu s\u00f6yl\u00fcyor.<\/span><\/p>\n A\u015fa\u011f\u0131daki kod, R’ye bir regresyon modelinin<\/a> nas\u0131l s\u0131\u011fd\u0131r\u0131laca\u011f\u0131n\u0131 ve ard\u0131ndan model taraf\u0131ndan yap\u0131lan tahminler ile g\u00f6zlemlenen ger\u00e7ek yan\u0131t de\u011ferleri aras\u0131ndaki ortalama mutlak hatan\u0131n nas\u0131l hesaplanaca\u011f\u0131n\u0131 g\u00f6sterir:<\/span><\/p>\n Ortalama mutlak hatan\u0131n (MAE) 1,238<\/strong> oldu\u011fu ortaya \u00e7\u0131k\u0131yor.<\/span><\/p>\n Bu bize g\u00f6zlemlenen de\u011ferler ile tahmin edilen de\u011ferler aras\u0131ndaki ortalama mutlak fark\u0131n 1,238 oldu\u011funu s\u00f6yl\u00fcyor.<\/span><\/p>\n Genel olarak MAE de\u011feri ne kadar d\u00fc\u015f\u00fck olursa, model bir veri setine o kadar iyi uyum sa\u011flayabilir. \u0130ki farkl\u0131 modeli kar\u015f\u0131la\u015ft\u0131rd\u0131\u011f\u0131m\u0131zda, hangisinin veri setine en iyi uyumu sa\u011flad\u0131\u011f\u0131n\u0131 bulmak i\u00e7in her modelin MAE’sini kar\u015f\u0131la\u015ft\u0131rabiliriz.<\/span><\/p>\n Ortalama Mutlak Hata Hesaplay\u0131c\u0131 \u0130statistiklerde ortalama mutlak hata (MAE), belirli bir modelin do\u011frulu\u011funu \u00f6l\u00e7menin bir yoludur. A\u015fa\u011f\u0131daki \u015fekilde hesaplan\u0131r: MAE = (1\/n) * \u03a3|y i \u2013 x i | Alt\u0131n: \u03a3: \u201ctoplam\u201d anlam\u0131na gelen Yunanca bir sembol y i : i’inci g\u00f6zlem i\u00e7in g\u00f6zlemlenen de\u011fer x i : i’inci g\u00f6zlem i\u00e7in tahmin edilen de\u011fer n: Toplam g\u00f6zlem say\u0131s\u0131 Metrics […]<\/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-1623","post","type-post","status-publish","format-standard","hentry","category-rehber"],"yoast_head":"\n\n
\u00d6rnek 1: \u0130ki vekt\u00f6r aras\u0131ndaki ortalama mutlak hatay\u0131 hesaplay\u0131n<\/strong><\/span><\/h3>\n
library<\/span> (Metrics)\n\n#define observed and predicted values\n<\/span>observed <- c(12, 13, 14, 15, 15, 22, 27, 29, 29, 30, 32)\npredicted <- c(11, 13, 14, 14, 16, 19, 24, 30, 32, 36, 30)\n\n#calculate mean absolute error between vectors\n<\/span>mae(observed, predicted)\n\n[1] 1.909091\n<\/strong><\/pre>\n \u00d6rnek 2: Bir Regresyon Modeli i\u00e7in Ortalama Mutlak Hatan\u0131n Hesaplanmas\u0131<\/strong><\/span><\/h3>\n
library<\/span> (Metrics)\n\n#create data\n<\/span>df <- data. frame<\/span> (x1=c(1, 3, 3, 4, 4, 6, 6, 8, 9, 3),\n x2=c(7, 7, 4, 10, 13, 12, 17, 19, 20, 34),\n y=c(17, 18, 19, 20, 24, 28, 25, 29, 30, 32))\n\n#view first six rows of data<\/span>\nhead(df)\n\n x1 x2 y\n1 1 7 17\n2 3 7 18\n3 3 4 19\n4 4 10 20\n5 4 13 24\n6 6 12 28\n\n#fit regression model\n<\/span>model <- lm(y~x1+x2, data=df)\n\n#calculate MAE between predicted values and observed values<\/span>\nmae(df$y, predict(model))\n\n[1] 1.238241\n<\/strong><\/pre>\n Ek kaynaklar<\/strong><\/span><\/h3>\n
Excel’de Ortalama Mutlak Hata Nas\u0131l Hesaplan\u0131r<\/a>
Python’da Ortalama Mutlak Hata Nas\u0131l Hesaplan\u0131r?<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"