Introduction to TidyModel in R

This post will introduce you to how to use tidymodel pacakge in R

Basic Introduction to Tidymodels using R

In this blog post, we will explore Tidymodels, a collection of packages for modeling and machine learning using R. This is part of what I learnt in the R for Data Science online Learning Community.

What is Tidymodels?

Tidymodels is a suite of packages that provides a consistent and flexible approach to modeling in R. It is part of the tidyverse, an ecosystem of R packages designed for data science.

Installing Tidymodels

To install Tidymodels, you can use the install.packages() function in R.

Basic Usage of Tidymodels

Let’s go through a simple example of using Tidymodels for linear regression.

Loading the necessary libraries

library(tidymodels)
tidymodels_prefer()

Preparing the data

For this example, we’ll use the mtcars dataset that comes with R. Let’s split this data into a training set and a testing set.

data(mtcars)
set.seed(123) # For reproducibility

car_split <- initial_split(mtcars, prop = 0.75)

car_train <- training(car_split)

car_test <- testing(car_split)

Building the model

We’ll try to predict miles per gallon (mpg) based on the other variables in the dataset. First, let’s specify our model:

lm_spec <- linear_reg() %>%
set_engine("lm") %>%
set_mode("regression")

Next, let’s fit our model to the training data:

lm_fit <- lm_spec %>%
fit(mpg ~ ., data = car_train)

We can now use this model to make predictions on the test data:

predictions <- lm_fit %>%
predict(new_data = car_test)


predictions <- car_test %>% 
  select(mpg) %>% 
  bind_cols(predict(lm_fit, car_test))

predictions

                     mpg    .pred
Mazda RX4 Wag       21.0 21.71246
Valiant             18.1 20.64933
Merc 450SE          16.4 12.94019
Merc 450SL          17.3 14.67981
Lincoln Continental 10.4 10.79525
Toyota Corona       21.5 25.01139
Camaro Z28          13.3 13.08460
Pontiac Firebird    19.2 16.27870

Conclusion

In this post, we have introduced Tidymodels, a powerful tool for modeling in R. We have seen how to install and use Tidymodels, and how it integrates with the tidyverse ecosystem. With Tidymodels, you can streamline your modeling workflow and make it more consistent and reproducible. In my next blog post I will explain how to tune hyper-parameters and also how to perform cross validation.