Build better training data

# Build better training data

## Tidymodels, virtually &mdash; Session 03

### Alison Hill

---
class: inverse, middle, center

# Build a better training set

# with recipes

---
class: middle, center, frame

# Recipes

---
background-image: url(images/cranes.jpg)
background-position: left
background-size: contain
class: middle

# Preprocessing options

+ Encode categorical predictors

+ Center and scale variables

+ Handle class imbalance

+ Impute missing data

+ Perform dimensionality reduction

+ *A lot more!*
]

---
background-image: url(images/workflows/workflows.013.jpeg)
background-size: contain
background-position: center

---
class: middle, center, frame

# To build a recipe

1\. Start the `recipe()`

2\. Define the .display[variables] involved

3\. Describe **prep**rocessing .display[step-by-step]

---
class: middle, center

# `recipe()`

Creates a recipe for a set of variables

```r
recipe(Class ~ ., data = alz)
```

---
class: middle

# .center[`recipe()`]

```r
rec <- 
  recipe(Class ~ ., data = alz)
```

---
class: middle

# .center[`step_*()`]

.center[Adds a single transformation to a recipe. 
Transformations are replayed in order when the recipe is run on data.]

```r
rec <- 
  recipe(Class ~ ., data = alz) %>%
  step_other(Genotype, threshold = .03)
```

---

---
class: middle

```
# # A tibble: 6 × 2
#   Genotype     n
#   <fct>    <int>
# 1 E3E3       153
# 2 E3E4        93
# 3 E2E3        31
# 4 E4E4        13
# 5 E2E4         8
# 6 E2E2         2
```

]

## After recipe

```
# # A tibble: 5 × 2
#   Genotype     n
#   <fct>    <int>
# 1 E3E3       153
# 2 E3E4        93
# 3 E2E3        31
# 4 E4E4        13
# 5 other       10
```

]
---
class: middle, center

# .center[`step_*()`]

Complete list at:
<https://recipes.tidymodels.org/reference/index.html>

---
class: middle, center, frame

# K Nearest Neighbors (KNN)

To predict the outcome of a new data point:

Find the K most similar old data points

Take the average/mode/etc. outcome

---
class: middle, frame

# .center[To specify a model with parsnip]

1\. Pick a .display[model] + .display[engine]

2\. Set the .display[mode] (if needed)

]

---
class: middle, frame

# .center[To specify a KNN model with parsnip]

```r
knn_mod <-
  nearest_neighbor() %>%              
  set_engine("kknn") %>%             
  set_mode("classification")        
```

---
class: middle, center

# Fact

KNN requires all numeric predictors, and all need to be **centered** and **scaled**.

What does that mean?

---
class: middle, center

# Quiz

Why do you need to "train" a recipe?

Imagine "scaling" a new data point. What do you subtract from it? 
What do you divide it by?

---
background-image: url(images/pca.002.jpeg)
background-size: contain

---
background-image: url(images/pca.003.jpeg)
background-size: contain

---
background-image: url(images/pca.004.jpeg)
background-size: contain

---

# Guess
]

```
# A tibble: 6 × 1
  Genotype
  <fct>   
1 E3E3    
2 E3E4    
3 E4E4    
4 E2E3    
5 E2E4    
6 E2E2    
```
]

```
# A tibble: 333 × 6
    E3E3  E3E4  E4E4  E2E3  E2E4  E2E2
   <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
 1     1     0     0     0     0     0
 2     0     1     0     0     0     0
 3     0     1     0     0     0     0
 4     0     1     0     0     0     0
 5     1     0     0     0     0     0
 6     0     0     1     0     0     0
 7     0     0     0     1     0     0
 8     0     0     0     1     0     0
 9     1     0     0     0     0     0
10     0     0     0     1     0     0
# … with 323 more rows
```

]

---
class: middle, center

# Dummy Variables

```r
glm(Class ~ Genotype, family = "binomial", data = alz)
```

```
# # A tibble: 6 × 5
#   term         estimate std.error statistic p.value
#   <chr>           <dbl>     <dbl>     <dbl>   <dbl>
# 1 (Intercept)      14.6      624.    0.0233   0.981
# 2 GenotypeE2E3    -13.1      624.   -0.0210   0.983
# 3 GenotypeE2E4    -12.6      624.   -0.0202   0.984
# 4 GenotypeE3E3    -13.2      624.   -0.0212   0.983
# 5 GenotypeE3E4    -14.1      624.   -0.0226   0.982
# 6 GenotypeE4E4    -15.0      624.   -0.0241   0.981
```

---
class: middle

Converts nominal data into numeric dummy variables,
needed as predictors for models like KNN.

]

```r
rec <- 
  recipe(Class ~ ., data = alz) %>%
  step_other(Genotype, threshold = .03) %>% 
* step_dummy(all_nominal(), -all_outcomes())
```

---
class: middle

# .center[Combining selectors]

Use commas to separate

```r
rec <- 
  recipe(Class ~ ., data = alz) %>%
  step_other(Genotype, threshold = .03) %>% 
* step_dummy(all_nominal(), -all_outcomes())
```

---
class: middle

How does recipes know which variables are **numeric** and what is **nominal**?
]

```r
rec <- 
  recipe(Class ~ ., 
*        data = alz)
```

---
class: middle

How does recipes know what is a **predictor** and what is an **outcome**?
]
--

```r
rec <-
* recipe(Sale_Price ~ .,
         data = ames)
```

---
class: middle

# .center[selectors]

Helper functions for selecting sets of variables

```r
rec %>% 
  step_novel(all_nominal()) %>%
  step_zv(all_predictors())
```

---
class: middle

<div id="mespryzeab" style="overflow-x:auto;overflow-y:auto;width:auto;height:auto;">
<style>html {
  font-family: -apple-system, BlinkMacSystemFont, 'Segoe UI', Roboto, Oxygen, Ubuntu, Cantarell, 'Helvetica Neue', 'Fira Sans', 'Droid Sans', Arial, sans-serif;
}

#mespryzeab .gt_table {
  display: table;
  border-collapse: collapse;
  margin-left: auto;
  margin-right: auto;
  color: #333333;
  font-size: 100px;
  font-weight: normal;
  font-style: normal;
  background-color: #FFFFFF;
  width: 10%;
  border-top-style: solid;
  border-top-width: 2px;
  border-top-color: #A8A8A8;
  border-right-style: none;
  border-right-width: 2px;
  border-right-color: #D3D3D3;
  border-bottom-style: solid;
  border-bottom-width: 2px;
  border-bottom-color: #A8A8A8;
  border-left-style: none;
  border-left-width: 2px;
  border-left-color: #D3D3D3;
}

#mespryzeab .gt_heading {
  background-color: #FFFFFF;
  text-align: center;
  border-bottom-color: #FFFFFF;
  border-left-style: none;
  border-left-width: 1px;
  border-left-color: #D3D3D3;
  border-right-style: none;
  border-right-width: 1px;
  border-right-color: #D3D3D3;
}

#mespryzeab .gt_title {
  color: #333333;
  font-size: 125%;
  font-weight: initial;
  padding-top: 4px;
  padding-bottom: 4px;
  border-bottom-color: #FFFFFF;
  border-bottom-width: 0;
}

#mespryzeab .gt_subtitle {
  color: #333333;
  font-size: 85%;
  font-weight: initial;
  padding-top: 0;
  padding-bottom: 4px;
  border-top-color: #FFFFFF;
  border-top-width: 0;
}

#mespryzeab .gt_bottom_border {
  border-bottom-style: solid;
  border-bottom-width: 2px;
  border-bottom-color: #D3D3D3;
}

#mespryzeab .gt_col_headings {
  border-top-style: solid;
  border-top-width: 2px;
  border-top-color: #D3D3D3;
  border-bottom-style: solid;
  border-bottom-width: 2px;
  border-bottom-color: #D3D3D3;
  border-left-style: none;
  border-left-width: 1px;
  border-left-color: #D3D3D3;
  border-right-style: none;
  border-right-width: 1px;
  border-right-color: #D3D3D3;
}

#mespryzeab .gt_col_heading {
  color: #333333;
  background-color: #FFFFFF;
  font-size: 100%;
  font-weight: normal;
  text-transform: inherit;
  border-left-style: none;
  border-left-width: 1px;
  border-left-color: #D3D3D3;
  border-right-style: none;
  border-right-width: 1px;
  border-right-color: #D3D3D3;
  vertical-align: bottom;
  padding-top: 5px;
  padding-bottom: 6px;
  padding-left: 5px;
  padding-right: 5px;
  overflow-x: hidden;
}

#mespryzeab .gt_column_spanner_outer {
  color: #333333;
  background-color: #FFFFFF;
  font-size: 100%;
  font-weight: normal;
  text-transform: inherit;
  padding-top: 0;
  padding-bottom: 0;
  padding-left: 4px;
  padding-right: 4px;
}

#mespryzeab .gt_column_spanner_outer:first-child {
  padding-left: 0;
}

#mespryzeab .gt_column_spanner_outer:last-child {
  padding-right: 0;
}

#mespryzeab .gt_column_spanner {
  border-bottom-style: solid;
  border-bottom-width: 2px;
  border-bottom-color: #D3D3D3;
  vertical-align: bottom;
  padding-top: 5px;
  padding-bottom: 6px;
  overflow-x: hidden;
  display: inline-block;
  width: 100%;
}

#mespryzeab .gt_group_heading {
  padding: 8px;
  color: #333333;
  background-color: #FFFFFF;
  font-size: 100%;
  font-weight: initial;
  text-transform: inherit;
  border-top-style: solid;
  border-top-width: 2px;
  border-top-color: #D3D3D3;
  border-bottom-style: solid;
  border-bottom-width: 2px;
  border-bottom-color: #D3D3D3;
  border-left-style: none;
  border-left-width: 1px;
  border-left-color: #D3D3D3;
  border-right-style: none;
  border-right-width: 1px;
  border-right-color: #D3D3D3;
  vertical-align: middle;
}

#mespryzeab .gt_empty_group_heading {
  padding: 0.5px;
  color: #333333;
  background-color: #FFFFFF;
  font-size: 100%;
  font-weight: initial;
  border-top-style: solid;
  border-top-width: 2px;
  border-top-color: #D3D3D3;
  border-bottom-style: solid;
  border-bottom-width: 2px;
  border-bottom-color: #D3D3D3;
  vertical-align: middle;
}

#mespryzeab .gt_from_md > :first-child {
  margin-top: 0;
}

#mespryzeab .gt_from_md > :last-child {
  margin-bottom: 0;
}

#mespryzeab .gt_row {
  padding-top: 8px;
  padding-bottom: 8px;
  padding-left: 5px;
  padding-right: 5px;
  margin: 10px;
  border-top-style: solid;
  border-top-width: 1px;
  border-top-color: #D3D3D3;
  border-left-style: none;
  border-left-width: 1px;
  border-left-color: #D3D3D3;
  border-right-style: none;
  border-right-width: 1px;
  border-right-color: #D3D3D3;
  vertical-align: middle;
  overflow-x: hidden;
}

#mespryzeab .gt_stub {
  color: #333333;
  background-color: #FFFFFF;
  font-size: 100%;
  font-weight: initial;
  text-transform: inherit;
  border-right-style: solid;
  border-right-width: 2px;
  border-right-color: #D3D3D3;
  padding-left: 12px;
}

#mespryzeab .gt_summary_row {
  color: #333333;
  background-color: #FFFFFF;
  text-transform: inherit;
  padding-top: 8px;
  padding-bottom: 8px;
  padding-left: 5px;
  padding-right: 5px;
}

#mespryzeab .gt_first_summary_row {
  padding-top: 8px;
  padding-bottom: 8px;
  padding-left: 5px;
  padding-right: 5px;
  border-top-style: solid;
  border-top-width: 2px;
  border-top-color: #D3D3D3;
}

#mespryzeab .gt_grand_summary_row {
  color: #333333;
  background-color: #FFFFFF;
  text-transform: inherit;
  padding-top: 8px;
  padding-bottom: 8px;
  padding-left: 5px;
  padding-right: 5px;
}

#mespryzeab .gt_first_grand_summary_row {
  padding-top: 8px;
  padding-bottom: 8px;
  padding-left: 5px;
  padding-right: 5px;
  border-top-style: double;
  border-top-width: 6px;
  border-top-color: #D3D3D3;
}

#mespryzeab .gt_striped {
  background-color: rgba(128, 128, 128, 0.05);
}

#mespryzeab .gt_table_body {
  border-top-style: solid;
  border-top-width: 2px;
  border-top-color: #D3D3D3;
  border-bottom-style: solid;
  border-bottom-width: 2px;
  border-bottom-color: #D3D3D3;
}

#mespryzeab .gt_footnotes {
  color: #333333;
  background-color: #FFFFFF;
  border-bottom-style: none;
  border-bottom-width: 2px;
  border-bottom-color: #D3D3D3;
  border-left-style: none;
  border-left-width: 2px;
  border-left-color: #D3D3D3;
  border-right-style: none;
  border-right-width: 2px;
  border-right-color: #D3D3D3;
}

#mespryzeab .gt_footnote {
  margin: 0px;
  font-size: 90%;
  padding: 4px;
}

#mespryzeab .gt_sourcenotes {
  color: #333333;
  background-color: #FFFFFF;
  border-bottom-style: none;
  border-bottom-width: 2px;
  border-bottom-color: #D3D3D3;
  border-left-style: none;
  border-left-width: 2px;
  border-left-color: #D3D3D3;
  border-right-style: none;
  border-right-width: 2px;
  border-right-color: #D3D3D3;
}

#mespryzeab .gt_sourcenote {
  font-size: 90%;
  padding: 4px;
}

#mespryzeab .gt_left {
  text-align: left;
}

#mespryzeab .gt_center {
  text-align: center;
}

#mespryzeab .gt_right {
  text-align: right;
  font-variant-numeric: tabular-nums;
}

#mespryzeab .gt_font_normal {
  font-weight: normal;
}

#mespryzeab .gt_font_bold {
  font-weight: bold;
}

#mespryzeab .gt_font_italic {
  font-style: italic;
}

#mespryzeab .gt_super {
  font-size: 65%;
}

#mespryzeab .gt_footnote_marks {
  font-style: italic;
  font-weight: normal;
  font-size: 65%;
}
</style>
<table class="gt_table">
  
  <thead class="gt_col_headings">
    <tr>
      <th class="gt_col_heading gt_columns_bottom_border gt_left" rowspan="1" colspan="1">selector</th>
      <th class="gt_col_heading gt_columns_bottom_border gt_left" rowspan="1" colspan="1">description</th>
    </tr>
  </thead>
  <tbody class="gt_table_body">
    <tr><td class="gt_row gt_left"><div class='gt_from_md'><p><code>all_predictors()</code></p>
</div></td>
<td class="gt_row gt_left"><div class='gt_from_md'><p>Each x variable  (right side of ~)</p>
</div></td></tr>
    <tr><td class="gt_row gt_left"><div class='gt_from_md'><p><code>all_outcomes()</code></p>
</div></td>
<td class="gt_row gt_left"><div class='gt_from_md'><p>Each y variable  (left side of ~)</p>
</div></td></tr>
    <tr><td class="gt_row gt_left"><div class='gt_from_md'><p><code>all_numeric()</code></p>
</div></td>
<td class="gt_row gt_left"><div class='gt_from_md'><p>Each numeric variable</p>
</div></td></tr>
    <tr><td class="gt_row gt_left"><div class='gt_from_md'><p><code>all_nominal()</code></p>
</div></td>
<td class="gt_row gt_left"><div class='gt_from_md'><p>Each categorical variable (e.g. factor, string)</p>
</div></td></tr>
    <tr><td class="gt_row gt_left"><div class='gt_from_md'><p><code>dplyr::select()</code> helpers</p>
</div></td>
<td class="gt_row gt_left"><div class='gt_from_md'><p><code>starts_with('IL_')</code>, etc.</p>
</div></td></tr>
  </tbody>
  
  
</table>
</div>

---
class: middle, center

# Let's think about the modeling.

What if there were no individuals with `Genotype` E2E2 in the training data?

Will the model have a coefficient for `Genotype` E2E2?

What will happen if the test data includes a person with `Genotype` E2E2?

---
class: middle

Adds a catch-all level to a factor for any new values not encountered in model training, 
which lets R intelligently predict new levels in the test set.

]

```r
rec <- 
  recipe(Class ~ ., data = alz) %>%
  step_other(Genotype, threshold = .03) %>% 
* step_novel(all_nominal(), -all_outcomes()) %>%
  step_dummy(all_nominal(), -all_outcomes()) 
```

---
class: middle, center

# Guess

What would happen if you try to normalize a variable that doesn't vary?

Error! You'd be dividing by zero!

---
class: middle

Intelligently handles zero variance variables 
(variables that contain only a single value)

]

```r
rec <- 
  recipe(Class ~ ., data = alz) %>%
  step_other(Genotype, threshold = .03) %>% 
  step_novel(all_nominal(), -all_outcomes()) %>% 
  step_dummy(all_nominal(), -all_outcomes()) %>% 
* step_zv(all_predictors())
```

---
class: middle

Centers then scales numeric variable (mean = 0, sd = 1)

]

# Your Turn 1

Unscramble! You have all the steps from our `knn_rec`- your challenge is to *unscramble* them into the right order!

Save the result as `knn_rec`

---

```r
knn_rec <- 
  recipe(formula = Class ~ ., data = alz) %>% 
  step_novel(all_nominal(), -all_outcomes()) %>% 
  step_dummy(all_nominal(), -all_outcomes()) %>% 
  step_zv(all_predictors()) %>% 
  step_normalize(all_predictors(), -all_nominal()) 
knn_rec
# Data Recipe
# 
# Inputs:
# 
#       role #variables
#    outcome          1
#  predictor        130
# 
# Operations:
# 
# Novel factor level assignment for all_nominal(), -all_outcomes()
# Dummy variables from all_nominal(), -all_outcomes()
# Zero variance filter on all_predictors()
# Centering and scaling for all_predictors(), -all_nominal()
```

---
class: middle, center, frame

# usemodels

<https://tidymodels.github.io/usemodels/>

---

```r
library(usemodels)
use_kknn(Class ~ ., data = alz, verbose = TRUE, tune = FALSE)
# kknn_recipe <- 
#   recipe(formula = Class ~ ., data = alz) %>% 
#   step_novel(all_nominal(), -all_outcomes()) %>% 
#   ## This model requires the predictors to be numeric. The most common 
#   ## method to convert qualitative predictors to numeric is to create 
#   ## binary indicator variables (aka dummy variables) from these 
#   ## predictors. 
#   step_dummy(all_nominal(), -all_outcomes()) %>% 
#   ## Since distance calculations are used, the predictor variables should 
#   ## be on the same scale. Before centering and scaling the numeric 
#   ## predictors, any predictors with a single unique value are filtered 
#   ## out. 
#   step_zv(all_predictors()) %>% 
#   step_normalize(all_predictors(), -all_nominal()) 
# 
# kknn_spec <- 
#   nearest_neighbor() %>% 
#   set_mode("classification") %>% 
#   set_engine("kknn") 
# 
# kknn_workflow <- 
#   workflow() %>% 
#   add_recipe(kknn_recipe) %>% 
#   add_model(kknn_spec)
```

---

```r
use_glmnet(Class ~ ., data = alz, verbose = TRUE, tune = FALSE)
# glmnet_recipe <- 
#   recipe(formula = Class ~ ., data = alz) %>% 
#   step_novel(all_nominal(), -all_outcomes()) %>% 
#   ## This model requires the predictors to be numeric. The most common 
#   ## method to convert qualitative predictors to numeric is to create 
#   ## binary indicator variables (aka dummy variables) from these 
#   ## predictors. 
#   step_dummy(all_nominal(), -all_outcomes()) %>% 
#   ## Regularization methods sum up functions of the model slope 
#   ## coefficients. Because of this, the predictor variables should be on 
#   ## the same scale. Before centering and scaling the numeric predictors, 
#   ## any predictors with a single unique value are filtered out. 
#   step_zv(all_predictors()) %>% 
#   step_normalize(all_predictors(), -all_nominal()) 
# 
# glmnet_spec <- 
#   logistic_reg() %>% 
#   set_mode("classification") %>% 
#   set_engine("glmnet") 
# 
# glmnet_workflow <- 
#   workflow() %>% 
#   add_recipe(glmnet_recipe) %>% 
#   add_model(glmnet_spec)
```

---
class: inverse, middle, center

## Now we've built a recipe.

## But, how do we *use* a recipe?

---
class: center, middle, frame

# Axiom

Feature engineering and modeling are two halves of a single predictive workflow.

---
background-image: url(images/workflows/workflows.001.jpeg)
background-size: contain

---
background-image: url(images/workflows/workflows.002.jpeg)
background-size: contain

---
background-image: url(images/workflows/workflows.003.jpeg)
background-size: contain

---
background-image: url(images/workflows/workflows.004.jpeg)
background-size: contain

---
background-image: url(images/workflows/workflows.005.jpeg)
background-size: contain

---
background-image: url(images/workflows/workflows.006.jpeg)
background-size: contain

---
background-image: url(images/workflows/workflows.007.jpeg)
background-size: contain

---
background-image: url(images/workflows/workflows.008.jpeg)
background-size: contain

---
background-image: url(images/workflows/workflows.009.jpeg)
background-size: contain

---
background-image: url(images/workflows/workflows.010.jpeg)
background-size: contain

---
background-image: url(images/workflows/workflows.011.jpeg)
background-size: contain

---
background-image: url(images/workflows/workflows.012.jpeg)
background-size: contain

---
background-image: url(images/workflows/workflows.013.jpeg)
background-size: contain

---
class: center, middle, inverse

# Workflows

---
class: middle, center

# `workflow()`

Creates a workflow to add a model and more to

```r
workflow()
```

---
class: middle, center

# `add_formula()`

Adds a formula to a workflow `*`

```r
workflow() %>% add_formula(Class ~ tau)
```

---
class: middle, center

# `add_model()`

Adds a parsnip model spec to a workflow

```r
workflow() %>% add_model(knn_mod)
```

---
class: middle, center

# Guess

If we use `add_model()` to add a model to a workflow, what would we use to add a recipe?

Let's see!

---
class: your-turn

# Your Turn 2

Fill in the blanks to make a workflow that combines `knn_rec` and with `knn_mod`.

---

```r
knn_wf <-
  workflow() %>% 
  add_recipe(knn_rec) %>% 
  add_model(knn_mod)
knn_wf
# ══ Workflow ════════════════════════════════════════════════════════════════════
# Preprocessor: Recipe
# Model: nearest_neighbor()
# 
# ── Preprocessor ────────────────────────────────────────────────────────────────
# 4 Recipe Steps
# 
# • step_novel()
# • step_dummy()
# • step_zv()
# • step_normalize()
# 
# ── Model ───────────────────────────────────────────────────────────────────────
# K-Nearest Neighbor Model Specification (classification)
# 
# Computational engine: kknn
```

---
class: middle

Adds a recipe to a workflow.

]

```r
knn_wf <- 
  workflow() %>%
* add_recipe(knn_rec) %>%
  add_model(knn_mod)
```

---
class: middle

Do you need to add a formula if you have a recipe?
]
--
.center[
Nope!
]

```r
rec <- 
* recipe(Class ~ .,
         data = alz)
```

---
class: middle

Fit a workflow that bundles a recipe`*` and a model.

]

```r
_wf %>% 
  fit(data = alz_train) %>% 
  predict(alz_test)...
```

---
class: middle

]

```r
set.seed(100)
alz_folds <- 
    vfold_cv(alz_train, v = 10, strata = Class)
```

---
class: middle

Fit a workflow that bundles a recipe`*` and a model with resampling.

]

```r
_wf %>% 
  fit_resamples(resamples = alz_folds)
```

---
class: your-turn

# Your Turn 3

Run the first chunk. Then try our KNN workflow on `alz_folds`. What is the ROC AUC?

---

```r
set.seed(100)
alz_folds <- 
  vfold_cv(alz_train, v = 10, strata = Class)

knn_wf %>% 
  fit_resamples(resamples = alz_folds) %>% 
  collect_metrics()
# # A tibble: 2 × 6
#   .metric  .estimator  mean     n std_err .config           
#   <chr>    <chr>      <dbl> <int>   <dbl> <chr>             
# 1 accuracy binary     0.713    10  0.0177 Preprocessor1_Mod…
# 2 roc_auc  binary     0.761    10  0.0285 Preprocessor1_Mod…
```

---
class: middle, center

# Feature Engineering

![](https://media.giphy.com/media/Wn74RUT0vjnoU98Hnt/giphy.gif)
]

![](https://media.giphy.com/media/108GZES8iG0myc/giphy.gif)
]

---

Replace the recipe in a workflow.

]

```r
_wf %>%
* update_recipe(glmnet_rec)
```

---

Replace the model in a workflow.

]

```r
_wf %>%
* update_model(tree_mod)
```

---
class: your-turn

# Your Turn 4

Turns out, the same `knn_rec` recipe can also be used to fit a penalized logistic regression model. Let's try it out!

```r
plr_mod <- 
  logistic_reg(penalty = .01, mixture = 1) %>% 
  set_engine("glmnet") %>% 
  set_mode("classification")

plr_mod %>% 
  translate()
# Logistic Regression Model Specification (classification)
# 
# Main Arguments:
#   penalty = 0.01
#   mixture = 1
# 
# Computational engine: glmnet 
# 
# Model fit template:
# glmnet::glmnet(x = missing_arg(), y = missing_arg(), weights = missing_arg(), 
#     alpha = 1, family = "binomial")
```

---

```r
glmnet_wf <-
  knn_wf %>% 
  update_model(plr_mod)

glmnet_wf %>% 
  fit_resamples(resamples = alz_folds) %>% 
  collect_metrics() 
# # A tibble: 2 × 6
#   .metric  .estimator  mean     n std_err .config           
#   <chr>    <chr>      <dbl> <int>   <dbl> <chr>             
# 1 accuracy binary     0.897    10  0.0128 Preprocessor1_Mod…
# 2 roc_auc  binary     0.926    10  0.0160 Preprocessor1_Mod…
```