Best and worst fitting models
In this exercise you will answer the following questions:
- Overall, how well do your models fit your data?
- Which are the best fitting models?
- Which models do not fit the data well?
This exercise is part of the course
Machine Learning in the Tidyverse
Exercise instructions
- Plot a histogram of the \(R^2\) values of the 77 models
- Extract the 4 best fitting models (based on \(R^2\)) and store this data frame as
best_fit - Extract the 4 worst fitting models (based on \(R^2\)) and store this data frame as
worst_fit
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Plot a histogram of rsquared for the 77 models
model_perf %>%
ggplot(aes(x = ___)) +
___()
# Extract the 4 best fitting models
best_fit <- model_perf %>%
slice_max(___, n = ___)
# Extract the 4 models with the worst fit
worst_fit <- model_perf %>%
slice_min(___, n = ___)