dtf = read.csv(“Salary_Data.csv”)
library(caTools)
set.seed(123)
split = sample.split(dtf$Salary, SplitRatio = 0,7)
set_training = subset(dtf, split == TRUE)
set_tes = subset(dtf, split == FALSE)
regressor = lm(formula = dtf$Salary ~ dtf$YearsExperience,
data = set_tes)
y = predict(regressor, newdata = set_tes)
# jika run
# Warning message:
# ‘newdata’ had 10 rows but variables found have 30 rows
result = cbind(set_tes, y)
# hasilnya tetep 30 row, kalo cuma ingin predict 10 row yang ada di set_tes aja gimana kak codingannya?
# apakah ada perlakuan khusus?