Menangani prediktor berinformasi rendah

Machine Learning dengan caret di R

Zach Mayer

Data Scientist at DataRobot and co-author of caret

Variabel tanpa (atau rendah) varians

Beberapa variabel minim informasi
- Konstan (tanpa varians)
- Hampir konstan (varians rendah)
Mudah bagi satu fold CV berakhir dengan kolom konstan
- Dapat menimbulkan masalah pada model
Biasanya hapus variabel dengan varians sangat rendah

Contoh: kolom konstan di mtcars

# Reproduce dataset from last video
data(mtcars)
set.seed(42)
mtcars[sample(1:nrow(mtcars), 10), "hp"] <- NA
Y <- mtcars$mpg
X <- mtcars[, 2:4]

# Add constant-valued column to mtcars
X$bad <- 1

Contoh: kolom konstan di mtcars

# Try to fit a model with PCA + glm
model <- train(
  X, Y, method = "glm", 
  preProcess = c("center", "scale", "medianImpute", "pca"))

Warning in preProcess.default(thresh = 0.95, k = 5, method = c("medianImpute",  :
  These variables have zero variances: bad
Something is wrong; all the RMSE metric values are missing:
      RMSE        Rsquared  
 Min.   : NA   Min.   : NA  
 1st Qu.: NA   1st Qu.: NA  
 Median : NA   Median : NA  
 Mean   :NaN   Mean   :NaN  
 3rd Qu.: NA   3rd Qu.: NA  
 Max.   : NA   Max.   : NA  
 NA's   :1     NA's   :1

caret menyelamatkan (lagi)

"zv" menghapus kolom konstan
"nzv" menghapus kolom hampir konstan

# Minta caret menghapus kolom tersebut saat pemodelan
set.seed(42)
model <- train(
  X, Y, method = "glm", 
  preProcess = c("zv", "center", "scale", "medianImpute", "pca")
)
min(model$results$RMSE)

3.402557

Ayo berlatih!

Machine Learning dengan caret di R