Supervised Learning with tidylearn

library(tidylearn)
library(dplyr)
#> 
#> Attaching package: 'dplyr'
#> The following objects are masked from 'package:stats':
#> 
#>     filter, lag
#> The following objects are masked from 'package:base':
#> 
#>     intersect, setdiff, setequal, union

Introduction

This vignette demonstrates supervised learning capabilities in tidylearn. All methods shown here wrap established R packages - the algorithms are unchanged, tidylearn simply provides a consistent interface and tidy output.

Wrapped packages include:

• stats (lm(), glm()) for linear and logistic regression
• rpart for decision trees
• randomForest for random forests
• gbm and xgboost for gradient boosting
• glmnet for regularization (ridge, lasso, elastic net)
• e1071 for support vector machines
• nnet for neural networks

Access raw model objects via model$fit for package-specific functionality.

Classification

Binary Classification

Let’s create a binary classification problem from the iris dataset:

# Create binary classification dataset. setosa is linearly separable from
# the other two species, and logistic regression has no finite maximum
# likelihood estimate on separable data -- glm() fits, warns that the
# algorithm did not converge, and returns coefficients that diverged.
# versicolor and virginica overlap, so this is a real classification
# problem. Even here, a 70% split of 100 rows is separable at some seeds;
# this one is not.
iris_binary <- iris %>%
  filter(Species %in% c("versicolor", "virginica")) %>%
  mutate(Species = droplevels(Species))

# Split data
split <- tl_split(iris_binary, prop = 0.7, stratify = "Species", seed = 42)

Logistic Regression

# Train logistic regression
model_logistic <- tl_model(split$train, Species ~ ., method = "logistic")
print(model_logistic)
#> tidylearn Model
#> ===============
#> Paradigm: supervised 
#> Method: logistic 
#> Task: Classification 
#> Formula: Species ~ . 
#> 
#> Training observations: 70

The type argument controls what .pred holds. This matters: the default "response" returns probabilities for logistic regression but class labels for trees, forests and SVMs. Ask for type = "class" when you want labels regardless of method.

# Predicted labels
preds_logistic <- predict(model_logistic, new_data = split$test, type = "class")
head(preds_logistic)
#> # A tibble: 6 × 1
#>   .pred     
#>   <fct>     
#> 1 versicolor
#> 2 versicolor
#> 3 versicolor
#> 4 versicolor
#> 5 versicolor
#> 6 versicolor

# Class probabilities
head(predict(model_logistic, new_data = split$test, type = "prob"))
#> # A tibble: 6 × 2
#>   versicolor  virginica
#>        <dbl>      <dbl>
#> 1      1.000 0.0000690 
#> 2      1.000 0.0000917 
#> 3      0.999 0.00101   
#> 4      1.000 0.00000874
#> 5      1.000 0.000208  
#> 6      0.952 0.0480

tl_evaluate() resolves this for you — it always scores against class labels, and computes whichever metrics you ask for:

tl_evaluate(
  model_logistic,
  new_data = split$test,
  metrics = c("accuracy", "precision", "recall", "f1", "auc")
)
#> # A tibble: 5 × 2
#>   metric    value
#>   <chr>     <dbl>
#> 1 accuracy  0.867
#> 2 precision 1    
#> 3 recall    0.733
#> 4 f1        0.846
#> 5 auc       0.991

Decision Trees

# Train decision tree
model_tree <- tl_model(split$train, Species ~ ., method = "tree")
print(model_tree)
#> tidylearn Model
#> ===============
#> Paradigm: supervised 
#> Method: tree 
#> Task: Classification 
#> Formula: Species ~ . 
#> 
#> Training observations: 70

# Predictions
preds_tree <- predict(model_tree, new_data = split$test)

Multi-class Classification

# Split full iris dataset
split_multi <- tl_split(iris, prop = 0.7, stratify = "Species", seed = 123)

Random Forest

# Train random forest
model_forest <- tl_model(split_multi$train, Species ~ ., method = "forest")
print(model_forest)
#> tidylearn Model
#> ===============
#> Paradigm: supervised 
#> Method: forest 
#> Task: Classification 
#> Formula: Species ~ . 
#> 
#> Training observations: 105

# Predictions
preds_forest <- predict(model_forest, new_data = split_multi$test)
head(preds_forest)
#> # A tibble: 6 × 1
#>   .pred 
#>   <fct> 
#> 1 setosa
#> 2 setosa
#> 3 setosa
#> 4 setosa
#> 5 setosa
#> 6 setosa

# Accuracy on test set
mean(preds_forest$.pred == split_multi$test$Species)
#> [1] 0.9333333

Support Vector Machines

# Train SVM
model_svm <- tl_model(split_multi$train, Species ~ ., method = "svm")
print(model_svm)
#> tidylearn Model
#> ===============
#> Paradigm: supervised 
#> Method: svm 
#> Task: Classification 
#> Formula: Species ~ . 
#> 
#> Training observations: 105

# Predictions
preds_svm <- predict(model_svm, new_data = split_multi$test)

Regression

Linear Regression

# Split mtcars data
split_reg <- tl_split(mtcars, prop = 0.7, seed = 123)

# Train linear model
model_lm <- tl_model(split_reg$train, mpg ~ wt + hp + disp, method = "linear")
print(model_lm)
#> tidylearn Model
#> ===============
#> Paradigm: supervised 
#> Method: linear 
#> Task: Regression 
#> Formula: mpg ~ wt + hp + disp 
#> 
#> Training observations: 22

# Predictions
preds_lm <- predict(model_lm, new_data = split_reg$test)
head(preds_lm)
#> # A tibble: 6 × 1
#>   .pred
#>   <dbl>
#> 1  24.0
#> 2  23.1
#> 3  21.2
#> 4  20.7
#> 5  16.0
#> 6  17.2

# Calculate RMSE
rmse <- sqrt(mean((preds_lm$.pred - split_reg$test$mpg)^2))
cat("RMSE:", round(rmse, 2), "\n")
#> RMSE: 2.16

Polynomial Regression

# Polynomial regression for non-linear relationships
model_poly <- tl_model(
  split_reg$train, mpg ~ wt,
  method = "polynomial", degree = 2
)
print(model_poly)
#> tidylearn Model
#> ===============
#> Paradigm: supervised 
#> Method: polynomial 
#> Task: Regression 
#> Formula: mpg ~ wt 
#> 
#> Training observations: 22

# Predictions
preds_poly <- predict(model_poly, new_data = split_reg$test)

# RMSE
rmse_poly <- sqrt(mean((preds_poly$.pred - split_reg$test$mpg)^2))
cat("Polynomial RMSE:", round(rmse_poly, 2), "\n")
#> Polynomial RMSE: 2.09

Random Forest Regression

# Train random forest for regression
model_rf_reg <- tl_model(split_reg$train, mpg ~ ., method = "forest")
print(model_rf_reg)
#> tidylearn Model
#> ===============
#> Paradigm: supervised 
#> Method: forest 
#> Task: Regression 
#> Formula: mpg ~ . 
#> 
#> Training observations: 22

# Predictions
preds_rf <- predict(model_rf_reg, new_data = split_reg$test)

# RMSE
rmse_rf <- sqrt(mean((preds_rf$.pred - split_reg$test$mpg)^2))
cat("Random Forest RMSE:", round(rmse_rf, 2), "\n")
#> Random Forest RMSE: 1.97

Regularized Regression

Regularization helps prevent overfitting by adding penalties to model complexity.

Ridge Regression

# Ridge regression (L2 regularization)
model_ridge <- tl_model(split_reg$train, mpg ~ ., method = "ridge")
print(model_ridge)
#> tidylearn Model
#> ===============
#> Paradigm: supervised 
#> Method: ridge 
#> Task: Regression 
#> Formula: mpg ~ . 
#> 
#> Training observations: 22

# Predictions
preds_ridge <- predict(model_ridge, new_data = split_reg$test)

LASSO

# LASSO (L1 regularization) - performs feature selection
model_lasso <- tl_model(split_reg$train, mpg ~ ., method = "lasso")
print(model_lasso)
#> tidylearn Model
#> ===============
#> Paradigm: supervised 
#> Method: lasso 
#> Task: Regression 
#> Formula: mpg ~ . 
#> 
#> Training observations: 22

# Predictions
preds_lasso <- predict(model_lasso, new_data = split_reg$test)

Elastic Net

# Elastic Net - combines L1 and L2 regularization
model_enet <- tl_model(
  split_reg$train, mpg ~ .,
  method = "elastic_net", alpha = 0.5
)
print(model_enet)
#> tidylearn Model
#> ===============
#> Paradigm: supervised 
#> Method: elastic_net 
#> Task: Regression 
#> Formula: mpg ~ . 
#> 
#> Training observations: 22

# Predictions
preds_enet <- predict(model_enet, new_data = split_reg$test)

Model Comparison

# Compare multiple models
models <- list(
  linear = tl_model(split_reg$train, mpg ~ ., method = "linear"),
  tree = tl_model(split_reg$train, mpg ~ ., method = "tree"),
  forest = tl_model(split_reg$train, mpg ~ ., method = "forest")
)

# Calculate RMSE for each model
results <- data.frame(
  Model = character(),
  RMSE = numeric(),
  stringsAsFactors = FALSE
)

for (model_name in names(models)) {
  preds <- predict(models[[model_name]], new_data = split_reg$test)
  rmse <- sqrt(mean((preds$.pred - split_reg$test$mpg)^2))

  results <- rbind(results, data.frame(
    Model = model_name,
    RMSE = rmse
  ))
}

results <- results %>% arrange(RMSE)
print(results)
#>    Model     RMSE
#> 1 forest 2.021532
#> 2 linear 2.281450
#> 3   tree 4.095888

Advanced Features

Using Preprocessed Data

# Preprocess data
processed <- tl_prepare_data(
  split_reg$train,
  mpg ~ .,
  scale_method = "standardize",
  remove_correlated = TRUE,
  correlation_cutoff = 0.9
)
#> Removing 1 highly correlated features
#> Scaling numeric features using method: standardize

# Train on preprocessed data
model_processed <- tl_model(processed$data, mpg ~ ., method = "linear")
print(model_processed)
#> tidylearn Model
#> ===============
#> Paradigm: supervised 
#> Method: linear 
#> Task: Regression 
#> Formula: mpg ~ . 
#> 
#> Training observations: 22

Formula Variations

# Interaction terms
model_interact <- tl_model(split_reg$train, mpg ~ wt * hp, method = "linear")

# Polynomial terms using I()
model_poly_manual <- tl_model(
  split_reg$train, mpg ~ wt + I(wt^2), method = "linear"
)

# Subset of predictors
model_subset <- tl_model(
  split_reg$train, mpg ~ wt + hp + disp, method = "linear"
)

Handling Different Data Types

Categorical Predictors

# Create dataset with categorical variables
mtcars_cat <- mtcars %>%
  mutate(
    cyl = as.factor(cyl),
    gear = as.factor(gear),
    am = as.factor(am)
  )

split_cat <- tl_split(mtcars_cat, prop = 0.7, seed = 123)

# Model with categorical predictors
model_cat <- tl_model(split_cat$train, mpg ~ ., method = "forest")
print(model_cat)
#> tidylearn Model
#> ===============
#> Paradigm: supervised 
#> Method: forest 
#> Task: Regression 
#> Formula: mpg ~ . 
#> 
#> Training observations: 22

Missing Values

# Create data with missing values. Seeded so the vignette renders the
# same output on every build.
set.seed(123)
mtcars_missing <- mtcars
mtcars_missing[sample(seq_len(nrow(mtcars_missing)), 5), "hp"] <- NA
mtcars_missing[sample(seq_len(nrow(mtcars_missing)), 3), "wt"] <- NA

# Preprocess to handle missing values
processed_missing <- tl_prepare_data(
  mtcars_missing,
  mpg ~ .,
  impute_method = "mean",
  scale_method = "standardize"
)
#> Imputing missing values using method: mean
#> Scaling numeric features using method: standardize

# Train model
model_imputed <- tl_model(processed_missing$data, mpg ~ ., method = "linear")

Best Practices

1. Split before training. A metric computed on the rows the model was fitted to measures memorisation, not performance.
2. Stratify classification splits so both sets carry the same class proportions as the source data.
3. Scale inputs for methods that need it – regularised regression, SVM and neural networks – and replay the training transformation on the test set, as shown below. Trees and forests are scale-invariant.
4. Compare several models on the same split before committing to one.
5. Reach for regularisation when predictors outnumber observations, or when they are strongly correlated.
6. Match the metric to the task – accuracy, F1 or AUC for classification; RMSE or MAE for regression.

Summary

tidylearn provides a unified interface for supervised learning:

• Classification: Logistic regression, decision trees, random forests, SVM, etc.
• Regression: Linear, polynomial, random forests, regularized methods
• Preprocessing: Integrated data preparation tools
• Consistent API: Same function (tl_model()) for all methods
• Tidy Output: Easy-to-use predictions and model objects

# Complete workflow example
final_split <- tl_split(iris, prop = 0.7, stratify = "Species", seed = 42)

final_prep <- tl_prepare_data(
  final_split$train, Species ~ .,
  scale_method = "standardize"
)
#> Scaling numeric features using method: standardize
final_model <- tl_model(final_prep$data, Species ~ ., method = "forest")

A model trained on scaled features expects scaled inputs at prediction time. tl_prepare_data() keeps the parameters it learned, so the same transformation can be replayed on the test set — applying the training centre and scale, never recomputing them from the test data:

scaling <- final_prep$preprocessing_steps$scaling$scaling_params

final_test <- final_split$test
for (col in names(scaling)) {
  final_test[[col]] <-
    (final_test[[col]] - scaling[[col]]$mean) / scaling[[col]]$sd
}

final_preds <- predict(final_model, new_data = final_test)
accuracy <- mean(final_preds$.pred == final_split$test$Species)
cat("Test Accuracy:", round(accuracy * 100, 1), "%\n")
#> Test Accuracy: 95.6 %

Skipping that step and predicting on raw test data is a common and quiet mistake — the model still returns predictions, they are just wrong. For multi-step preprocessing, tl_pipeline() handles this bookkeeping for you: tl_predict_pipeline() replays the training preprocessing automatically.