Code

library(tidyverse)
library(tidymodels)

1 model type

2 specify model

3 Data splitting with rsample

Code

home_split <- initial_split(Sacramento,prop=0.75)
home_split

<Training/Testing/Total>
<699/233/932>

Code

home_train=training(home_split)
glimpse(home_train)

Rows: 699
Columns: 9
$ city      <fct> SACRAMENTO, LINCOLN, SACRAMENTO, ELK_GROVE, SACRAMENTO, SACR…
$ zip       <fct> z95842, z95648, z95835, z95758, z95831, z95864, z95624, z956…
$ beds      <int> 4, 5, 3, 3, 2, 3, 4, 2, 3, 4, 2, 3, 3, 4, 3, 2, 3, 3, 5, 3, …
$ baths     <dbl> 2.0, 3.0, 2.0, 2.0, 2.0, 1.0, 3.0, 2.0, 2.0, 4.0, 1.0, 2.0, …
$ sqft      <int> 1292, 3072, 1120, 1273, 1324, 1643, 3992, 1315, 1262, 2213, …
$ type      <fct> Residential, Residential, Residential, Residential, Resident…
$ price     <int> 105000, 315000, 209000, 190000, 234500, 99000, 460000, 23000…
$ latitude  <dbl> 38.67960, 38.86641, 38.68195, 38.43124, 38.48797, 38.58867, …
$ longitude <dbl> -121.3560, -121.3085, -121.5050, -121.4400, -121.5302, -121.…

Code

home_test=testing(home_split)
glimpse(home_test)

Rows: 233
Columns: 9
$ city      <fct> SACRAMENTO, SACRAMENTO, SACRAMENTO, SACRAMENTO, ELK_GROVE, S…
$ zip       <fct> z95823, z95841, z95823, z95827, z95758, z95828, z95660, z958…
$ beds      <int> 3, 3, 4, 3, 2, 4, 4, 4, 3, 3, 3, 3, 3, 4, 3, 3, 4, 3, 3, 4, …
$ baths     <dbl> 1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 4, 1, 2, 2, 2, 3, 2, …
$ sqft      <int> 1167, 1122, 1329, 1380, 1039, 1146, 1587, 1590, 1463, 1406, …
$ type      <fct> Residential, Condo, Residential, Residential, Condo, Residen…
$ price     <int> 68212, 89921, 122682, 136500, 141000, 149593, 161500, 173000…
$ latitude  <dbl> 38.47890, 38.66260, 38.46817, 38.56666, 38.42325, 38.49857, …
$ longitude <dbl> -121.4310, -121.3278, -121.4441, -121.3326, -121.4445, -121.…

3.1 closs validation

Code

set.seed(123)
vfold_cv(home_train,strata = price)

#  10-fold cross-validation using stratification 
# A tibble: 10 × 2
   splits           id    
   <list>           <chr> 
 1 <split [627/72]> Fold01
 2 <split [627/72]> Fold02
 3 <split [627/72]> Fold03
 4 <split [628/71]> Fold04
 5 <split [628/71]> Fold05
 6 <split [630/69]> Fold06
 7 <split [631/68]> Fold07
 8 <split [631/68]> Fold08
 9 <split [631/68]> Fold09
10 <split [631/68]> Fold10

3.2 Bootstrapping

sample with replacement

Code

set.seed(123)
bootstraps(home_train,strata = price)

# Bootstrap sampling using stratification 
# A tibble: 25 × 2
   splits            id         
   <list>            <chr>      
 1 <split [699/255]> Bootstrap01
 2 <split [699/263]> Bootstrap02
 3 <split [699/255]> Bootstrap03
 4 <split [699/258]> Bootstrap04
 5 <split [699/254]> Bootstrap05
 6 <split [699/249]> Bootstrap06
 7 <split [699/258]> Bootstrap07
 8 <split [699/259]> Bootstrap08
 9 <split [699/268]> Bootstrap09
10 <split [699/256]> Bootstrap10
# ℹ 15 more rows

4 recipe

feature engineering

5 workflow and model

6 Reference

[Youtuber]3 Reasons to Use Tidymodels with Julia Silge

https://www.youtube.com/watch?v=sv5r7CVAVwo