glmnet with custom tuning grid

Machine Learning with caret in R

Zach Mayer

Data Scientist at DataRobot and co-author of caret

Custom tuning glmnet models

2 tuning parameters: alpha and lambda
For single alpha, all values of lambda fit simultaneously
Many models for the "price" of one

Example: glmnet tuning

# Make a custom tuning grid
myGrid <- expand.grid(
  alpha = 0:1, 
  lambda = seq(0.0001, 0.1, length = 10)
)

# Fit a model
set.seed(42)
model <- train(
  y ~ ., 
  overfit, 
  method = "glmnet", 
  tuneGrid = myGrid, 
  trControl = myControl
)

# Plot results
plot(model)

Compare models visually

The image shows a line plot of ROC curve versus regularization parameter for two values of the mixing percentage: 0 and 1. The ROC is maximized when the mixing percentage is 1 and the regularization parameter is greater than 0.1.

Full regularization path

plot(model$finalModel)

The image shows a custom line plot for glmnet models. There are many lines of coefficients versus L1 norm. As L1 norm increases, the range of coefficients gets wider.

Let’s practice!

Machine Learning with caret in R