Training and Evaluating Random Forests

Random Forests: Training and Evaluating Random Forests

library(randomForest)

## Warning: package 'randomForest' was built under R version 4.2.2

## randomForest 4.7-1.1

## Type rfNews() to see new features/changes/bug fixes.

library(caret)

## Warning: package 'caret' was built under R version 4.2.2

## Loading required package: ggplot2

## 
## Attaching package: 'ggplot2'

## The following object is masked from 'package:randomForest':
## 
##     margin

## Loading required package: lattice

library(rpart)

## Warning: package 'rpart' was built under R version 4.2.1

Lila <- read.csv("Ghemri.csv")

set.seed(300)

rf <- randomForest(Left ~., data = Lila)
rf

## 
## Call:
##  randomForest(formula = Left ~ ., data = Lila) 
##                Type of random forest: regression
##                      Number of trees: 500
## No. of variables tried at each split: 1
## 
##           Mean of squared residuals: 594.5238
##                     % Var explained: -64.5

Evaluating Random Forest performance

ctrl <- trainControl(method = "cv", number=3)

grid_rf <- expand.grid(mtry=3)

m_rf <- train(Left ~., data = Lila, method = "rf", metric = "RMSE",
              trControl = ctrl, tuneGrid = grid_rf)

m_rf

## Random Forest 
## 
## 10 samples
##  3 predictor
## 
## No pre-processing
## Resampling: Cross-Validated (3 fold) 
## Summary of sample sizes: 8, 6, 6 
## Resampling results:
## 
##   RMSE      Rsquared  MAE     
##   21.23442  0.615621  18.62709
## 
## Tuning parameter 'mtry' was held constant at a value of 3

Making a prediction (B. Lantz, P.363)

pred <- predict(m_rf, Lila)
table(pred, Lila$Left)

##                   
## pred               25 30 32 42 45 50 55 68 76 84
##   37.2257           0  0  1  0  0  0  0  0  0  0
##   40.1264333333333  0  1  0  0  0  0  0  0  0  0
##   41.4908666666667  0  0  0  0  0  1  0  0  0  0
##   41.5614666666667  1  0  0  0  0  0  0  0  0  0
##   49.6420333333334  0  0  0  1  0  0  0  0  0  0
##   54.4244           0  0  0  0  1  0  0  0  0  0
##   55.5200666666667  0  0  0  0  0  0  1  0  0  0
##   56.6163333333334  0  0  0  0  0  0  0  0  1  0
##   61.7091           0  0  0  0  0  0  0  1  0  0
##   68.6468666666667  0  0  0  0  0  0  0  0  0  1