Comparing Knime and R: Simple Random Forest

Sebastian Sauer / 8 mins read
2020-11-28

Knime Workflow

Consider this Knime workflow:

knime-rf
de picture

That’s a simple Random Forest workflow, you can download the Knime worfklow here.

Translate it to R!

Load Packages

library(tidyverse)  # data wrangling
library(tidymodels)  # modelling
library(conflicted)  # name clashes
library(caret)  # confusion matrix
library(vip)  # variance importance plots
library(gt)  # nice table formatting

Load Data

data(iris)
glimpse(iris)
## Rows: 150
## Columns: 5
## $ Sepal.Length <dbl> 5.1, 4.9, 4.7, 4.6, 5.0, 5.4, 4.6, 5.0, 4.4, 4.9, 5.4, 4…
## $ Sepal.Width  <dbl> 3.5, 3.0, 3.2, 3.1, 3.6, 3.9, 3.4, 3.4, 2.9, 3.1, 3.7, 3…
## $ Petal.Length <dbl> 1.4, 1.4, 1.3, 1.5, 1.4, 1.7, 1.4, 1.5, 1.4, 1.5, 1.5, 1…
## $ Petal.Width  <dbl> 0.2, 0.2, 0.2, 0.2, 0.2, 0.4, 0.3, 0.2, 0.2, 0.1, 0.2, 0…
## $ Species      <fct> setosa, setosa, setosa, setosa, setosa, setosa, setosa, …

Stratified sampling

split_vector <- initial_split(iris,
                              strata = Species)
traindata <- training(split_vector)

testdata <- testing(split_vector)

Random Forest classification model in R

Define and run Random Forest classification model

Define learner (model)

First define the model:

rf_mod <- 
  rand_forest(trees = 100) %>% 
  set_engine("ranger") %>% 
  set_mode("classification")

Define recipe

rf_recipe <-
  recipe(Species ~ ., data = traindata) %>% 
  step_zv(all_predictors())  # filter zero variance variables

A recipe only contains

  • names of variables
  • their roles
  • planned prepossing steps (here only filter all zero variance variables)
rf_recipe
## Data Recipe
## 
## Inputs:
## 
##       role #variables
##    outcome          1
##  predictor          4
## 
## Operations:
## 
## Zero variance filter on all_predictors()

Put workflow together

workflow() %>% 
  add_recipe(rf_recipe) %>% 
  add_model(rf_mod) -> rf_wf

rf_wf
## ══ Workflow ════════════════════════════════════════════════════════════════════
## Preprocessor: Recipe
## Model: rand_forest()
## 
## ── Preprocessor ────────────────────────────────────────────────────────────────
## 1 Recipe Step
## 
## ● step_zv()
## 
## ── Model ───────────────────────────────────────────────────────────────────────
## Random Forest Model Specification (classification)
## 
## Main Arguments:
##   trees = 100
## 
## Computational engine: ranger

Fit the model to the train data

Then fit it:

set.seed(123)

rf_fit <-
  rf_wf %>% 
  fit(data = traindata)

OOB results

rf_fit
## ══ Workflow [trained] ══════════════════════════════════════════════════════════
## Preprocessor: Recipe
## Model: rand_forest()
## 
## ── Preprocessor ────────────────────────────────────────────────────────────────
## 1 Recipe Step
## 
## ● step_zv()
## 
## ── Model ───────────────────────────────────────────────────────────────────────
## Ranger result
## 
## Call:
##  ranger::ranger(x = maybe_data_frame(x), y = y, num.trees = ~100,      num.threads = 1, verbose = FALSE, seed = sample.int(10^5,          1), probability = TRUE) 
## 
## Type:                             Probability estimation 
## Number of trees:                  100 
## Sample size:                      114 
## Number of independent variables:  4 
## Mtry:                             2 
## Target node size:                 10 
## Variable importance mode:         none 
## Splitrule:                        gini 
## OOB prediction error (Brier s.):  0.04269909

Model results in test data

testdata %>% 
  bind_cols(predict(rf_fit, testdata, type = "prob")) %>% 
  bind_cols(predict(rf_fit, testdata)) -> testdata_augmented

Multiclass accuracy

testdata_augmented %>% 
  roc_auc(truth = Species, 
          .pred_setosa:.pred_virginica)
## # A tibble: 1 x 3
##   .metric .estimator .estimate
##   <chr>   <chr>          <dbl>
## 1 roc_auc hand_till          1
testdata_augmented %>% 
  metrics(truth = Species, 
          estimate = .pred_class)
## # A tibble: 2 x 3
##   .metric  .estimator .estimate
##   <chr>    <chr>          <dbl>
## 1 accuracy multiclass         1
## 2 kap      multiclass         1

ROC

testdata_augmented %>% 
  roc_curve(truth = Species, 
            estimate = .pred_setosa:.pred_virginica) %>% 
  autoplot()

Confusion Matrix

confusionMatrix(data = testdata_augmented$.pred_class,
                reference = testdata_augmented$Species)
## Confusion Matrix and Statistics
## 
##             Reference
## Prediction   setosa versicolor virginica
##   setosa         12          0         0
##   versicolor      0         12         0
##   virginica       0          0        12
## 
## Overall Statistics
##                                      
##                Accuracy : 1          
##                  95% CI : (0.9026, 1)
##     No Information Rate : 0.3333     
##     P-Value [Acc > NIR] : < 2.2e-16  
##                                      
##                   Kappa : 1          
##                                      
##  Mcnemar's Test P-Value : NA         
## 
## Statistics by Class:
## 
##                      Class: setosa Class: versicolor Class: virginica
## Sensitivity                 1.0000            1.0000           1.0000
## Specificity                 1.0000            1.0000           1.0000
## Pos Pred Value              1.0000            1.0000           1.0000
## Neg Pred Value              1.0000            1.0000           1.0000
## Prevalence                  0.3333            0.3333           0.3333
## Detection Rate              0.3333            0.3333           0.3333
## Detection Prevalence        0.3333            0.3333           0.3333
## Balanced Accuracy           1.0000            1.0000           1.0000

Random Forest regression model in R

Define and run the model

Update model to regression

Update the classification model to regression

rf_mod  %>%  
  set_mode("regression") %>% 
  set_args(mtry = 3, importance = "impurity") -> rf_mod_regr

Define recipe

rf_recipe_regr <-
  recipe(Petal.Width ~ ., data = traindata)

Put workflow together

workflow() %>% 
  add_recipe(rf_recipe_regr) %>% 
  add_model(rf_mod_regr)  -> rf_wf_regr

rf_wf_regr
## ══ Workflow ════════════════════════════════════════════════════════════════════
## Preprocessor: Recipe
## Model: rand_forest()
## 
## ── Preprocessor ────────────────────────────────────────────────────────────────
## 0 Recipe Steps
## 
## ── Model ───────────────────────────────────────────────────────────────────────
## Random Forest Model Specification (regression)
## 
## Main Arguments:
##   mtry = 3
##   trees = 100
## 
## Engine-Specific Arguments:
##   importance = impurity
## 
## Computational engine: ranger

OOB results

Then fit the model:

set.seed(123)

rf_fit_regr <-
  rf_wf_regr %>% 
  fit(data = traindata)
rf_fit_regr
## ══ Workflow [trained] ══════════════════════════════════════════════════════════
## Preprocessor: Recipe
## Model: rand_forest()
## 
## ── Preprocessor ────────────────────────────────────────────────────────────────
## 0 Recipe Steps
## 
## ── Model ───────────────────────────────────────────────────────────────────────
## Ranger result
## 
## Call:
##  ranger::ranger(x = maybe_data_frame(x), y = y, mtry = min_cols(~3,      x), num.trees = ~100, importance = ~"impurity", num.threads = 1,      verbose = FALSE, seed = sample.int(10^5, 1)) 
## 
## Type:                             Regression 
## Number of trees:                  100 
## Sample size:                      114 
## Number of independent variables:  4 
## Mtry:                             3 
## Target node size:                 5 
## Variable importance mode:         impurity 
## Splitrule:                        variance 
## OOB prediction error (MSE):       0.03710244 
## R squared (OOB):                  0.9355139

Model results in test data

testdata %>% 
  bind_cols(predict(rf_fit_regr, testdata)) -> testdata_augmented

Variabble importance

pull_workflow_fit(rf_fit_regr) %>% 
  vip()

Collect performance metrics

testdata %>% 
  bind_cols(rf_fit_regr %>% predict(testdata)) %>% 
  metrics(truth = Petal.Width,
          estimate = .pred) %>% 
  gt()
.metric .estimator .estimate
rmse standard 0.1543815
rsq standard 0.9727964
mae standard 0.1024894