3 - What makes a model?

Machine learning with tidymodels

Your turn

How do you fit a linear model in R?

How many different ways can you think of?

03:00

  • lm for linear model

  • glmnet for regularized regression

  • keras for regression using TensorFlow

  • stan for Bayesian regression

  • spark for large data sets

Artwork by @allison_horst

To specify a model

  • Choose a model
  • Specify an engine
  • Set the mode

To specify a model 

linear_reg()
#> Linear Regression Model Specification (regression)
#> 
#> Computational engine: lm

To specify a model

  • Choose a model
  • Specify an engine
  • Set the mode

To specify a model 

linear_reg() %>%
  set_engine("glmnet")
#> Linear Regression Model Specification (regression)
#> 
#> Computational engine: glmnet

To specify a model 

linear_reg() %>%
  set_engine("stan")
#> Linear Regression Model Specification (regression)
#> 
#> Computational engine: stan

To specify a model

  • Choose a model
  • Specify an engine
  • Set the mode

To specify a model 

decision_tree()
#> Decision Tree Model Specification (unknown)
#> 
#> Computational engine: rpart

To specify a model 

decision_tree() %>% 
  set_mode("regression")
#> Decision Tree Model Specification (regression)
#> 
#> Computational engine: rpart



All available models are listed at https://www.tidymodels.org/find/parsnip/

To specify a model

  • Choose a model
  • Specify an engine
  • Set the mode

Your turn

Run the tree_spec chunk in your .qmd.

Edit this code so it creates a different model, such as linear regression.

05:00



All available models are listed at https://www.tidymodels.org/find/parsnip/

A model workflow

Workflows bind preprocessors and models

What is wrong with this?

Why a workflow()?

  • You can use other preprocessors besides formulas (more on feature engineering later!)

  • They can help organize your work when working with multiple models

  • Most importantly, a workflow captures the entire modeling process: fit() and predict() apply to the preprocessing steps in addition to the actual model fit

A model workflow 

tree_spec <- decision_tree(mode = "regression")

tree_spec %>% 
  fit(rings ~ ., data = ring_train) 
#> parsnip model object
#> 
#> n= 3340 
#> 
#> node), split, n, deviance, yval
#>       * denotes terminal node
#> 
#>  1) root 3340 34681.9200  9.937425  
#>    2) shell_weight< 0.16775 1146  5102.2900  7.584642  
#>      4) shell_weight< 0.05325 242   484.3512  5.524793 *
#>      5) shell_weight>=0.05325 904  3316.2640  8.136062  
#>       10) sex=infant 557  1432.8580  7.565530 *
#>       11) sex=female,male 347  1411.0660  9.051873 *
#>    3) shell_weight>=0.16775 2194 19922.2800 11.166360  
#>      6) shell_weight< 0.35775 1588 11128.8300 10.587530  
#>       12) shell_weight< 0.24925 679  3807.1960  9.948454  
#>         24) shucked_weight>=0.24775 528  1773.1650  9.460227 *
#>         25) shucked_weight< 0.24775 151  1468.0930 11.655630 *
#>       13) shell_weight>=0.24925 909  6837.1710 11.064910  
#>         26) shucked_weight>=0.39975 620  2638.9340 10.372580 *
#>         27) shucked_weight< 0.39975 289  3263.5220 12.550170 *
#>      7) shell_weight>=0.35775 606  6867.1680 12.683170  
#>       14) shucked_weight>=0.55025 429  3609.9910 12.004660  
#>         28) shell_weight< 0.579 382  2243.0990 11.607330 *
#>         29) shell_weight>=0.579 47   816.4255 15.234040 *
#>       15) shucked_weight< 0.55025 177  2580.9940 14.327680 *

A model workflow 

tree_spec <- decision_tree(mode = "regression")

workflow(rings ~ ., tree_spec) %>% 
  fit(data = ring_train) 
#> ══ Workflow [trained] ════════════════════════════════════════════════
#> Preprocessor: Formula
#> Model: decision_tree()
#> 
#> ── Preprocessor ──────────────────────────────────────────────────────
#> rings ~ .
#> 
#> ── Model ─────────────────────────────────────────────────────────────
#> n= 3340 
#> 
#> node), split, n, deviance, yval
#>       * denotes terminal node
#> 
#>  1) root 3340 34681.9200  9.937425  
#>    2) shell_weight< 0.16775 1146  5102.2900  7.584642  
#>      4) shell_weight< 0.05325 242   484.3512  5.524793 *
#>      5) shell_weight>=0.05325 904  3316.2640  8.136062  
#>       10) sex=infant 557  1432.8580  7.565530 *
#>       11) sex=female,male 347  1411.0660  9.051873 *
#>    3) shell_weight>=0.16775 2194 19922.2800 11.166360  
#>      6) shell_weight< 0.35775 1588 11128.8300 10.587530  
#>       12) shell_weight< 0.24925 679  3807.1960  9.948454  
#>         24) shucked_weight>=0.24775 528  1773.1650  9.460227 *
#>         25) shucked_weight< 0.24775 151  1468.0930 11.655630 *
#>       13) shell_weight>=0.24925 909  6837.1710 11.064910  
#>         26) shucked_weight>=0.39975 620  2638.9340 10.372580 *
#>         27) shucked_weight< 0.39975 289  3263.5220 12.550170 *
#>      7) shell_weight>=0.35775 606  6867.1680 12.683170  
#>       14) shucked_weight>=0.55025 429  3609.9910 12.004660  
#>         28) shell_weight< 0.579 382  2243.0990 11.607330 *
#>         29) shell_weight>=0.579 47   816.4255 15.234040 *
#>       15) shucked_weight< 0.55025 177  2580.9940 14.327680 *

Your turn

Run the tree_wflow chunk in your .qmd.

Edit this code so it uses a linear model.

05:00

Predict with your model

How do you use your new tree_fit model?

tree_spec <- decision_tree(mode = "regression")

tree_fit <-
  workflow(rings ~ ., tree_spec) %>% 
  fit(data = ring_train)

Your turn

Run:

predict(tree_fit, new_data = ring_test)

What do you get?

03:00

Your turn

Run:

augment(tree_fit, new_data = ring_test)

What do you get?

03:00

The tidymodels prediction guarantee!

  • The predictions will always be inside a tibble
  • The column names and types are unsurprising and predictable
  • The number of rows in new_data and the output are the same

Understand your model

How do you understand your new tree_fit model?

You can use your fitted workflow for model and/or prediction explanations:

  • overall variable importance, such as with the vip package

  • flexible model explainers, such as with the DALEXtra package

Learn more at https://www.tmwr.org/explain.html

Understand your model

How do you understand your new tree_fit model?

Understand your model

How do you understand your new tree_fit model?

library(rpart.plot)
tree_fit %>%
  extract_fit_engine() %>%
  rpart.plot()

You can extract_*() several components of your fitted workflow: https://workflows.tidymodels.org/reference/extract-workflow.html

⚠️ Never predict() with any extracted components!

Deploy your model

How do you use your new tree_fit model in production?

library(vetiver)
v <- vetiver_model(tree_fit, "abalone-rings")
v
#> 
#> ── abalone-rings ─ <butchered_workflow> model for deployment 
#> A rpart regression modeling workflow using 8 features

Learn more at https://vetiver.rstudio.com

Deploy your model

How do you use your new model tree_fit in production?

library(plumber)
pr() %>%
  vetiver_api(v)
#> # Plumber router with 2 endpoints, 4 filters, and 1 sub-router.
#> # Use `pr_run()` on this object to start the API.
#> ├──[queryString]
#> ├──[body]
#> ├──[cookieParser]
#> ├──[sharedSecret]
#> ├──/logo
#> │  │ # Plumber static router serving from directory: /Library/Frameworks/R.framework/Versions/4.2-arm64/Resources/library/vetiver
#> ├──/ping (GET)
#> └──/predict (POST)

Learn more at https://vetiver.rstudio.com

Your turn

Run the vetiver chunk in your .qmd.

Check out the automated visual documentation.

05:00

https://bit.ly/learn-tidymodels