Code Along 3: SF Rental Market
Michael Toronto
2025-02-06
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/main/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)| 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) %>%
na.omit() %>%
# log transform variables with pos-skewed distribution
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 average 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: 95
## $ `date__-Inf_20120214` <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ date__20120214_20140817 <dbl> 1, 1, 1, 1, 0, 1, 1, 1, 0,…
## $ date__20140817_20160126 <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ date__20160126_Inf <dbl> 0, 0, 0, 0, 1, 0, 0, 0, 1,…
## $ `year__-Inf_2012` <dbl> 0, 1, 1, 1, 0, 1, 1, 0, 0,…
## $ year__2012_2014 <dbl> 1, 0, 0, 0, 0, 0, 0, 1, 0,…
## $ year__2014_2016 <dbl> 0, 0, 0, 0, 1, 0, 0, 0, 1,…
## $ year__2016_Inf <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ 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,…
## $ room_in_apt__0 <dbl> 1, 1, 1, 1, 1, 1, 1, 1, 1,…
## $ `room_in_apt__-OTHER` <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0,…# Step 2 : Correlate
data_corr_tbl <- data_binarized_tbl %>%
correlate(price__8.07868822922987_Inf)
data_corr_tbl## # A tibble: 95 × 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 year -Inf_2012 -0.246
## 10 beds 3_Inf 0.241
## # ℹ 85 more rows#Step 3: Plot
data_corr_tbl %>%
plot_correlation_funnel()## Warning: ggrepel: 70 unlabeled data points (too many overlaps). Consider
## increasing max.overlaps
Build Models
Split Data
data <- sample_n(data, 100)
#Split into train and test dataset
set.seed(1234)
data_split <- rsample::initial_split(data)
data_train <- training(data_split)
data_test <- testing(data_split)
#Further split training dataset for cross-validation
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 [67/8]> Fold01
## 2 <split [67/8]> Fold02
## 3 <split [67/8]> Fold03
## 4 <split [67/8]> Fold04
## 5 <split [67/8]> Fold05
## 6 <split [68/7]> Fold06
## 7 <split [68/7]> Fold07
## 8 <split [68/7]> Fold08
## 9 <split [68/7]> Fold09
## 10 <split [68/7]> Fold10usemodels::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(6804)
## 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) %>%
step_dummy(nhood, city, county, one_hot = TRUE) %>%
step_YeoJohnson(sqft, beds, baths)
xgboost_recipe %>% prep() %>% juice() %>% glimpse()## Rows: 75
## Columns: 125
## $ post_id <chr> "3812150630", "pre2013_49593", "4614967425", "pr…
## $ date <dbl> 20130518, 20120115, 20140817, 20120112, 20150106…
## $ year <dbl> 2013, 2012, 2014, 2012, 2015, 2012, 2012, 2012, …
## $ beds <dbl> 1.5987282, 2.2408127, 2.2408127, 0.8740267, 2.24…
## $ baths <dbl> 1.3387891, 1.3387891, 1.3387891, 0.7841239, 1.57…
## $ sqft <dbl> 3.718957, 3.835847, 3.808913, 3.652452, 3.876690…
## $ room_in_apt <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ price <dbl> 7.261225, 7.718685, 7.935587, 7.003065, 8.160518…
## $ tfidf_title_1 <dbl> 0.0000000, 0.0000000, 0.0000000, 0.4105569, 0.00…
## $ tfidf_title_12 <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_1550 <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_16 <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_17 <dbl> 0.000000, 0.000000, 0.608443, 0.000000, 0.000000…
## $ tfidf_title_1795 <dbl> 0.000000, 0.000000, 0.000000, 0.000000, 0.000000…
## $ tfidf_title_1850 <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_1ba <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, 0.2791990, 0.00…
## $ tfidf_title_2 <dbl> 0.0000000, 0.1899782, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_2.5 <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.69…
## $ tfidf_title_2.5ba <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_2500ft2 <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_2600 <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_2800 <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_2980 <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_2ba <dbl> 0.0000000, 0.0000000, 0.3566777, 0.0000000, 0.00…
## $ tfidf_title_2bath <dbl> 0.5430161, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_2bd <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_2bed <dbl> 0.608443, 0.000000, 0.000000, 0.000000, 0.000000…
## $ tfidf_title_2br <dbl> 0.2737046, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_3 <dbl> 0.0000000, 0.2172838, 0.0000000, 0.0000000, 0.43…
## $ tfidf_title_3200 <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_3500 <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_3ba <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_3bd <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_3br <dbl> 0.0000000, 0.1854586, 0.2472781, 0.0000000, 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, 0.0000000, 0.00…
## $ tfidf_title_5br <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_9 <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_900ft2 <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_a <dbl> 0.0000000, 0.0000000, 0.0000000, 0.4072621, 0.00…
## $ tfidf_title_and <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_apartment <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_available <dbl> 0.4101348, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_ba <dbl> 0.0000000, 0.2924249, 0.0000000, 0.0000000, 0.58…
## $ tfidf_title_bath <dbl> 0.0000000, 0.0000000, 0.0000000, 0.1696280, 0.00…
## $ tfidf_title_bathroom <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_bed <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_bedroom <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_br <dbl> 0.0000000, 0.3076011, 0.0000000, 0.0000000, 0.61…
## $ tfidf_title_car <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_condo <dbl> 0.000000, 0.000000, 0.000000, 0.000000, 0.000000…
## $ tfidf_title_cruz <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_cupertino <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_for <dbl> 0.0000000, 0.0000000, 0.0000000, 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, 0.0000000, 0.3253362, 0.00…
## $ tfidf_title_gorgeous <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_great <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_har <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_hide <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_home <dbl> 0.0000000, 0.0000000, 0.4101348, 0.0000000, 0.00…
## $ tfidf_title_house <dbl> 0.0000000, 0.2570565, 0.3427420, 0.0000000, 0.00…
## $ tfidf_title_huge <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_in <dbl> 0.00000000, 0.00000000, 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_kitchen <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_large <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_location <dbl> 0.000000, 0.000000, 0.000000, 0.000000, 0.000000…
## $ tfidf_title_luxury <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_map <dbl> 0.4620981, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_near <dbl> 0.5430161, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_new <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_now <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_oca <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_open <dbl> 0.0000000, 0.0000000, 0.5430161, 0.0000000, 0.00…
## $ tfidf_title_parking <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_patio <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ 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_private <dbl> 0.0000000, 0.0000000, 0.0000000, 0.4072621, 0.00…
## $ tfidf_title_ramon <dbl> 0.000000, 0.000000, 0.000000, 0.000000, 0.000000…
## $ tfidf_title_remodeled <dbl> 0.0000000, 0.3253362, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_rent <dbl> 0.000000, 0.000000, 0.000000, 0.000000, 0.000000…
## $ 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_room <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_schools <dbl> 0.0000000, 0.3728942, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_south <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_spacious <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_storage <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tfidf_title_the <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ 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_townhome <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ 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_w <dbl> 0.0000000, 0.0000000, 0.0000000, 0.0000000, 0.00…
## $ tfidf_title_with <dbl> 0.0000000, 0.0000000, 0.0000000, 0.2675083, 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…
## $ nhood_cupertino <dbl> 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ nhood_san.jose.east <dbl> 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ nhood_san.jose.south <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ nhood_other <dbl> 1, 0, 0, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, …
## $ city_cupertino <dbl> 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ city_san.francisco <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ city_san.jose <dbl> 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, …
## $ city_other <dbl> 1, 0, 0, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, …
## $ county_alameda <dbl> 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ county_contra.costa <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, …
## $ county_marin <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ county_napa <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, …
## $ county_san.francisco <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ county_san.mateo <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, …
## $ county_santa.clara <dbl> 0, 1, 1, 1, 0, 1, 0, 1, 1, 0, 0, 1, 1, 0, 0, 1, …
## $ county_santa.cruz <dbl> 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ county_sonoma <dbl> 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …# Specify model
xgboost_spec <-
boost_tree(trees = tune(), min_n = tune(), tree_depth = tune(), learn_rate = tune()) %>%
set_mode("regression") %>%
set_engine("xgboost")
# Combine recipe and model using work
xgboost_workflow <-
workflow() %>%
add_recipe(xgboost_recipe) %>%
add_model(xgboost_spec)
# Tune hyperparameters
set.seed(30220)
xgboost_tune <-
tune_grid(xgboost_workflow, resamples = data_cv, grid = 5)## → A | warning: A correlation computation is required, but `estimate` is constant and has 0
## standard deviation, resulting in a divide by 0 error. `NA` will be returned.## There were issues with some computations A: x1There were issues with some computations A: x2There were issues with some computations A: x3There were issues with some computations A: x4There were issues with some computations A: x5There were issues with some computations A: x6There were issues with some computations A: x7There were issues with some computations A: x8There were issues with some computations A: x9 → B | warning: ! There are new levels in `county`: "napa".
## ℹ Consider using step_novel() (`?recipes::step_novel()`) before `step_dummy()`
## to handle unseen values.
## There were issues with some computations A: x9There were issues with some computations A: x9 B: x1There were issues with some computations A: x9 B: x2There were issues with some computations A: x9 B: x3There were issues with some computations A: x9 B: x4There were issues with some computations A: x9 B: x5There were issues with some computations A: x10 B: x5Evaluate Models
tune::show_best(xgboost_tune, metric = "rmse")## # A tibble: 5 × 10
## trees min_n tree_depth learn_rate .metric .estimator mean n std_err
## <int> <int> <int> <dbl> <chr> <chr> <dbl> <int> <dbl>
## 1 1000 2 1 0.0750 rmse standard 0.356 10 0.0438
## 2 500 21 15 0.316 rmse standard 0.364 10 0.0359
## 3 2000 40 8 0.0178 rmse standard 0.476 10 0.0492
## 4 1500 11 11 0.001 rmse standard 1.75 10 0.0700
## 5 1 30 4 0.00422 rmse standard 7.38 10 0.0607
## # ℹ 1 more variable: .config <chr># 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 0.342 Preprocessor1_Model1
## 2 rsq standard 0.604 Preprocessor1_Model1tune::collect_predictions(data_fit) %>%
ggplot(aes(price, .pred)) +
geom_point(alpha = 0.3, fill = "midnightblue") +
geom_abline(lty = 2, color = "gray50") +
coord_fixed()