Goal: To predict the rental prices in the SF rental market Click here for the data.

Import Data

rent <- readr::read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2022/2022-07-05/rent.csv')
## Rows: 200796 Columns: 17
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (8): post_id, nhood, city, county, address, title, descr, details
## dbl (9): date, year, price, beds, baths, sqft, room_in_apt, lat, lon
## 
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
skimr::skim(rent)
Data summary
Name rent
Number of rows 200796
Number of columns 17
_______________________
Column type frequency:
character 8
numeric 9
________________________
Group variables None

Variable type: character

skim_variable n_missing complete_rate min max empty n_unique whitespace
post_id 0 1.00 9 14 0 200796 0
nhood 0 1.00 4 43 0 167 0
city 0 1.00 5 19 0 104 0
county 1394 0.99 4 13 0 10 0
address 196888 0.02 1 38 0 2869 0
title 2517 0.99 2 298 0 184961 0
descr 197542 0.02 13 16975 0 3025 0
details 192780 0.04 4 595 0 7667 0

Variable type: numeric

skim_variable n_missing complete_rate mean sd p0 p25 p50 p75 p100 hist
date 0 1.00 20095718.38 44694.07 20000902.00 20050227.00 20110924.00 20120805.0 20180717.00 ▁▇▁▆▃
year 0 1.00 2009.51 4.48 2000.00 2005.00 2011.00 2012.0 2018.00 ▁▇▁▆▃
price 0 1.00 2135.36 1427.75 220.00 1295.00 1800.00 2505.0 40000.00 ▇▁▁▁▁
beds 6608 0.97 1.89 1.08 0.00 1.00 2.00 3.0 12.00 ▇▂▁▁▁
baths 158121 0.21 1.68 0.69 1.00 1.00 2.00 2.0 8.00 ▇▁▁▁▁
sqft 136117 0.32 1201.83 5000.22 80.00 750.00 1000.00 1360.0 900000.00 ▇▁▁▁▁
room_in_apt 0 1.00 0.00 0.04 0.00 0.00 0.00 0.0 1.00 ▇▁▁▁▁
lat 193145 0.04 37.67 0.35 33.57 37.40 37.76 37.8 40.43 ▁▁▅▇▁
lon 196484 0.02 -122.21 0.78 -123.20 -122.42 -122.26 -122.0 -74.20 ▇▁▁▁▁ 
data <- rent %>%
  
  # Treat missing values
  select(-address, -descr, -details, -lat, -lon, -date, -year, -room_in_apt) %>%
  na.omit() %>%
  
  # log transform variable with pos-skewed distributions
  mutate(price = log(price))

Explore Data

Identify good predictors

sqft

data %>%
  ggplot(aes(price, sqft)) +
  scale_y_log10() +
  geom_point()

beds

data %>%
  ggplot(aes(price, as.factor(beds))) +
  geom_boxplot()

title

data %>%
  
  
  #tokenize title
  unnest_tokens(output = word, input = title) %>%
  
  # calculate avg rent per word
  group_by(word) %>%
  summarise(price = mean(price),
            n     = n()) %>%
  ungroup() %>%
  
  filter(n > 10, !str_detect(word, "\\d")) %>%
  slice_max(order_by = price, n = 20) %>%
  
  # Plot
  ggplot(aes(price, fct_reorder(word, price))) +
  geom_point() +
  
  labs(y = "Words in Title")

EDA shortcut

# Step 1: Prepare Data
data_binarized_tbl <- data %>%
  select(-post_id, -title) %>%
  binarize

data_binarized_tbl %>% glimpse()
## Rows: 14,394
## Columns: 85
## $ nhood__campbell                             <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ `nhood__concord_/_pleasant_hill_/_martinez` <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ nhood__cupertino                            <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ nhood__daly_city                            <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ `nhood__danville_/_san_ramon`               <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ `nhood__dublin_/_pleasanton`                <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ `nhood__fairfield_/_vacaville`              <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ nhood__foster_city                          <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ `nhood__hayward_/_castro_valley`            <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ nhood__milpitas                             <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ nhood__mountain_view                        <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ nhood__napa_county                          <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ nhood__palo_alto                            <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ nhood__petaluma                             <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ `nhood__pittsburg_/_antioch`                <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ `nhood__rohnert_pk_/_cotati`                <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ nhood__san_francisco                        <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ nhood__san_jose_central                     <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ nhood__san_jose_east                        <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ nhood__san_jose_north                       <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ nhood__san_jose_south                       <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ nhood__san_jose_west                        <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ nhood__san_mateo                            <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ nhood__san_rafael                           <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ nhood__santa_clara                          <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ nhood__santa_cruz                           <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ nhood__santa_rosa                           <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ `nhood__SOMA_/_south_beach`                 <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ nhood__sunnyvale                            <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ nhood__union_city                           <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ `nhood__vallejo_/_benicia`                  <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ `nhood__willow_glen_/_cambrian`             <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ `nhood__-OTHER`                             <dbl> 1, 1, 1, 1, 1, 1, 1, 1, 1,…
## $ city__cambrian                              <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ city__campbell                              <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ city__concord                               <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ city__cupertino                             <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ city__daly_city                             <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ city__dublin                                <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ city__fairfield                             <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ city__foster_city                           <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ city__hayward                               <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ city__milpitas                              <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ city__mountain_view                         <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ city__napa_county                           <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ city__oakland                               <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ city__palo_alto                             <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ city__petaluma                              <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ city__pittsburg                             <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ city__rohnert_park                          <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ city__san_francisco                         <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ city__san_jose                              <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ city__san_mateo                             <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ city__san_rafael                            <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ city__san_ramon                             <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ city__santa_clara                           <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ city__santa_cruz                            <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ city__santa_rosa                            <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ city__sunnyvale                             <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ city__union_city                            <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ city__vallejo                               <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ `city__-OTHER`                              <dbl> 1, 1, 1, 1, 1, 1, 1, 1, 1,…
## $ county__alameda                             <dbl> 1, 1, 1, 1, 1, 1, 1, 1, 1,…
## $ county__contra_costa                        <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ county__marin                               <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ county__napa                                <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ county__san_francisco                       <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ county__san_mateo                           <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ county__santa_clara                         <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ county__santa_cruz                          <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ county__solano                              <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ county__sonoma                              <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ `price__-Inf_7.52294091807237`              <dbl> 0, 1, 0, 1, 0, 1, 1, 0, 0,…
## $ price__7.52294091807237_7.80384330353877    <dbl> 0, 0, 1, 0, 0, 0, 0, 1, 1,…
## $ price__7.80384330353877_8.07868822922987    <dbl> 1, 0, 0, 0, 1, 0, 0, 0, 0,…
## $ price__8.07868822922987_Inf                 <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ `beds__-Inf_2`                              <dbl> 0, 1, 0, 1, 1, 1, 0, 0, 1,…
## $ beds__2_3                                   <dbl> 0, 0, 1, 0, 0, 0, 1, 1, 0,…
## $ beds__3_Inf                                 <dbl> 1, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ `baths__-Inf_2`                             <dbl> 0, 1, 1, 1, 1, 1, 0, 0, 1,…
## $ baths__2_Inf                                <dbl> 1, 0, 0, 0, 0, 0, 1, 1, 0,…
## $ `sqft__-Inf_887`                            <dbl> 0, 1, 0, 0, 0, 0, 0, 0, 1,…
## $ sqft__887_1100                              <dbl> 0, 0, 0, 1, 0, 0, 0, 1, 0,…
## $ sqft__1100_1500                             <dbl> 0, 0, 1, 0, 1, 1, 0, 0, 0,…
## $ sqft__1500_Inf                              <dbl> 1, 0, 0, 0, 0, 0, 1, 0, 0,…
# Step 2:Correlate
data_corr_tbl <- data_binarized_tbl %>%
  correlate(price__8.07868822922987_Inf)

data_corr_tbl
## # A tibble: 85 × 3
##    feature bin                               correlation
##    <fct>   <chr>                                   <dbl>
##  1 price   8.07868822922987_Inf                    1    
##  2 city    san_francisco                           0.389
##  3 county  san_francisco                           0.389
##  4 price   -Inf_7.52294091807237                  -0.342
##  5 price   7.80384330353877_8.07868822922987      -0.330
##  6 price   7.52294091807237_7.80384330353877      -0.328
##  7 sqft    1500_Inf                                0.324
##  8 beds    -Inf_2                                 -0.254
##  9 beds    3_Inf                                   0.241
## 10 sqft    -Inf_887                               -0.240
## # ℹ 75 more rows
# Step 3: Plot
data_corr_tbl %>%
  plot_correlation_funnel()
## Warning: ggrepel: 69 unlabeled data points (too many overlaps). Consider
## increasing max.overlaps

Build Models

Split data

# data <- sample_n(data, 100)

# Split into training and testing data set
set.seed(1234)
data_split <- rsample::initial_split(data)
data_train <- training(data_split)
data_test <- testing(data_split)

# Further split training data set for cross-validation 
# Use rsample::bootstraps(data_train) instead of vfold_cv(data_train) for data sets under 1000
set.seed(2345)
data_cv <- rsample::vfold_cv(data_train)
data_cv
## #  10-fold cross-validation 
## # A tibble: 10 × 2
##    splits              id    
##    <list>              <chr> 
##  1 <split [9715/1080]> Fold01
##  2 <split [9715/1080]> Fold02
##  3 <split [9715/1080]> Fold03
##  4 <split [9715/1080]> Fold04
##  5 <split [9715/1080]> Fold05
##  6 <split [9716/1079]> Fold06
##  7 <split [9716/1079]> Fold07
##  8 <split [9716/1079]> Fold08
##  9 <split [9716/1079]> Fold09
## 10 <split [9716/1079]> Fold10
library(usemodels)
## Warning: package 'usemodels' was built under R version 4.4.1
usemodels::use_xgboost(price ~ ., data = data_train)
## xgboost_recipe <- 
##   recipe(formula = price ~ ., data = data_train) %>% 
##   step_zv(all_predictors()) 
## 
## xgboost_spec <- 
##   boost_tree(trees = tune(), min_n = tune(), tree_depth = tune(), learn_rate = tune(), 
##     loss_reduction = tune(), sample_size = tune()) %>% 
##   set_mode("classification") %>% 
##   set_engine("xgboost") 
## 
## xgboost_workflow <- 
##   workflow() %>% 
##   add_recipe(xgboost_recipe) %>% 
##   add_model(xgboost_spec) 
## 
## set.seed(30220)
## xgboost_tune <-
##   tune_grid(xgboost_workflow, resamples = stop("add your rsample object"), grid = stop("add number of candidate points"))
# Specify recipe
 xgboost_recipe <- 
  recipe(formula = price ~ ., data = data_train) %>%
  recipes::update_role(post_id, new_role = "id variable") %>%
  step_tokenize(title) %>%
  step_tokenfilter(title, max_tokens = 100) %>%
  step_tfidf(title) %>%
  step_other(nhood, city, county) %>%
  step_dummy(nhood, city, county, one_hot = TRUE) %>%
  step_YeoJohnson(sqft, beds, baths)
  


xgboost_recipe %>% prep() %>% juice() %>% glimpse()
## Rows: 10,795
## Columns: 117
## $ post_id               <fct> 4811120754, 4924595622, 6359733516, pre2013_1696…
## $ beds                  <dbl> 1.4109310, 1.4109310, 1.9069741, 1.9069741, 0.80…
## $ baths                 <dbl> 0.5801628, 0.8329395, 0.8329395, 0.9820716, 0.58…
## $ sqft                  <dbl> 4.770187, 4.655974, 4.807771, 4.911460, 4.451772…
## $ price                 <dbl> 7.374629, 7.945910, 8.341649, 7.495542, 7.130899…
## $ tfidf_title_1         <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.48…
## $ tfidf_title_1.5       <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_1ba       <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_1bath     <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_1br       <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.32…
## $ tfidf_title_2         <dbl> 0.26170848, 0.00000000, 0.16356780, 0.00000000, …
## $ tfidf_title_2.5       <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_2ba       <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_2bath     <dbl> 0.000000, 0.000000, 0.000000, 0.000000, 0.000000…
## $ tfidf_title_2bd       <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_2br       <dbl> 0.00000000, 0.72145594, 0.00000000, 0.00000000, …
## $ tfidf_title_3         <dbl> 0.0000000, 0.0000000, 0.2490204, 0.0000000, 0.00…
## $ tfidf_title_3br       <dbl> 0.0000000, 0.0000000, 0.0000000, 0.3722824, 0.00…
## $ tfidf_title_4         <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_4br       <dbl> 0.0000000, 0.0000000, 0.0000000, 0.5368510, 0.00…
## $ tfidf_title_5         <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_6         <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_7         <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_8         <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_9         <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_a         <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_amp       <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_and       <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_apartment <dbl> 0.536851, 0.000000, 0.000000, 0.000000, 0.000000…
## $ tfidf_title_apt       <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_at        <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_available <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_ba        <dbl> 0.0000000, 0.0000000, 0.3011678, 0.0000000, 0.00…
## $ tfidf_title_bath      <dbl> 0.00000000, 0.00000000, 0.00000000, 0.00000000, …
## $ tfidf_title_bathroom  <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_baths     <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_bd        <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_beautiful <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_bed       <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.29…
## $ tfidf_title_bedroom   <dbl> 0.3570456, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_bedrooms  <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_br        <dbl> 0.0000000, 0.0000000, 0.3225057, 0.0000000, 0.00…
## $ tfidf_title_car       <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_condo     <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_downtown  <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_family    <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_floor     <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_for       <dbl> 0.0000000, 0.0000000, 0.0000000, 0.5048902, 0.36…
## $ tfidf_title_ft        <dbl> 0.0000000, 0.0000000, 0.4319118, 0.0000000, 0.00…
## $ tfidf_title_full      <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_garage    <dbl> 0.0000000, 0.0000000, 0.3921030, 0.0000000, 0.00…
## $ tfidf_title_great     <dbl> 0.6102327, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_har       <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_hide      <dbl> 0.000000, 0.000000, 0.000000, 0.000000, 0.000000…
## $ tfidf_title_hill      <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_home      <dbl> 0.0000000, 0.0000000, 0.0000000, 0.4025397, 0.00…
## $ tfidf_title_house     <dbl> 0.0000000, 0.0000000, 0.3059722, 0.0000000, 0.00…
## $ tfidf_title_in        <dbl> 0.00000000, 0.83344405, 0.00000000, 0.00000000, …
## $ tfidf_title_jose      <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_large     <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_location  <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_luxury    <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_map       <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_move      <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_near      <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_new       <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_nice      <dbl> 0.0000000, 0.0000000, 0.0000000, 0.7484949, 0.00…
## $ tfidf_title_now       <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_of        <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_one       <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_open      <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_parking   <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_pic       <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_posting   <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_remodeled <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_rent      <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_resim     <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_restore   <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_san       <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_santa     <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_single    <dbl> 0.000000, 0.000000, 0.000000, 0.000000, 0.000000…
## $ tfidf_title_spacious  <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_sq        <dbl> 0.0000000, 0.0000000, 0.4393652, 0.0000000, 0.00…
## $ tfidf_title_şub       <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_the       <dbl> 0.000000, 0.000000, 0.000000, 0.000000, 0.464113…
## $ tfidf_title_this      <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_to        <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_top       <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_townhome  <dbl> 0.000000, 0.000000, 0.000000, 0.000000, 0.000000…
## $ tfidf_title_townhouse <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_two       <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_unit      <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_valley    <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_view      <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_views     <dbl> 0.6856356, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_w         <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_with      <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_게시물    <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_사진      <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_월        <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_이        <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_지도      <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_图片      <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_地图      <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_月        <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ nhood_sunnyvale       <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ nhood_other           <dbl> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, …
## $ city_san.francisco    <dbl> 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, …
## $ city_san.jose         <dbl> 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ city_other            <dbl> 1, 1, 0, 1, 1, 0, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, …
## $ county_alameda        <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, …
## $ county_contra.costa   <dbl> 1, 1, 0, 1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, …
## $ county_san.francisco  <dbl> 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, …
## $ county_san.mateo      <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, …
## $ county_santa.clara    <dbl> 0, 0, 0, 0, 0, 1, 0, 1, 1, 0, 0, 0, 0, 0, 1, 0, …
## $ county_sonoma         <dbl> 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, …
## $ county_other          <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, …
# Specify model
xgboost_spec <- 
  boost_tree(trees = tune(), min_n = tune(), mtry = tune(), learn_rate = tune()) %>% 
  set_mode("regression") %>% 
  set_engine("xgboost") 

# Combine recipe and model using workflow
xgboost_workflow <- 
  workflow() %>% 
  add_recipe(xgboost_recipe) %>% 
  add_model(xgboost_spec) 

# Tune hyperparameters
set.seed(344)
xgboost_tune <-
  tune_grid(xgboost_workflow, 
            resamples = data_cv, 
            grid = 5)
## i Creating pre-processing data to finalize unknown parameter: mtry
## Warning: package 'xgboost' was built under R version 4.4.1

Evaluate Models

tune::show_best(xgboost_tune, metric = "rmse")
## # A tibble: 5 × 10
##    mtry trees min_n learn_rate .metric .estimator  mean     n std_err .config   
##   <int> <int> <int>      <dbl> <chr>   <chr>      <dbl> <int>   <dbl> <chr>     
## 1    69   768    12    0.112   rmse    standard   0.260    10 0.00261 Preproces…
## 2    91  1524    23    0.0836  rmse    standard   0.261    10 0.00302 Preproces…
## 3     7  1613    36    0.0290  rmse    standard   0.262    10 0.00338 Preproces…
## 4    37  1104    28    0.00484 rmse    standard   0.279    10 0.00318 Preproces…
## 5   103   162     7    0.00108 rmse    standard   6.16     10 0.00400 Preproces…
# Update the model by selecting the best hyperparameters.
xgboost_fw <- tune::finalize_workflow(xgboost_workflow,
                        tune::select_best(xgboost_tune, metric = "rmse"))


# Fit the model on the entire training data and test it on the test data.
data_fit <- tune::last_fit(xgboost_fw, data_split)
tune::collect_metrics(data_fit)
## # A tibble: 2 × 4
##   .metric .estimator .estimate .config             
##   <chr>   <chr>          <dbl> <chr>               
## 1 rmse    standard       0.254 Preprocessor1_Model1
## 2 rsq     standard       0.692 Preprocessor1_Model1
tune::collect_predictions(data_fit) %>%
  ggplot(aes(price, .pred)) +
  geom_point(alpha = 0.3, fill = "midnightblue") +
  geom_abline(lty = 2, color = "gray50") +
  coord_fixed()

Make Predictions