R textrecipes step_tf 代幣的使用頻率

step_tf(
  recipe,
  ...,
  role = "predictor",
  trained = FALSE,
  columns = NULL,
  weight_scheme = "raw count",
  weight = 0.5,
  vocabulary = NULL,
  res = NULL,
  prefix = "tf",
  keep_original_cols = FALSE,
  skip = FALSE,
  id = rand_id("tf")
)

step_tokenize() 將字符轉換為tokens

來自標記的數字變量的其他步驟：step_lda()、step_texthash()、step_tfidf()、step_word_embeddings()

# \donttest{
library(recipes)
library(modeldata)
data(tate_text)

tate_rec <- recipe(~., data = tate_text) %>%
  step_tokenize(medium) %>%
  step_tf(medium)

tate_obj <- tate_rec %>%
  prep()

bake(tate_obj, tate_text)
#> # A tibble: 4,284 × 956
#>        id artist        title  year tf_medium_1 tf_medium_10 tf_medium_100
#>     <dbl> <fct>         <fct> <dbl>       <int>        <int>         <int>
#>  1  21926 Absalon       Prop…  1990           0            0             0
#>  2  20472 Auerbach, Fr… Mich…  1990           0            0             0
#>  3  20474 Auerbach, Fr… Geof…  1990           0            0             0
#>  4  20473 Auerbach, Fr… Jake   1990           0            0             0
#>  5  20513 Auerbach, Fr… To t…  1990           0            0             0
#>  6  21389 Ayres, OBE G… Phaë…  1990           0            0             0
#>  7 121187 Barlow, Phyl… Unti…  1990           0            0             0
#>  8  19455 Baselitz, Ge… Gree…  1990           0            0             0
#>  9  20938 Beattie, Bas… Pres…  1990           0            0             0
#> 10 105941 Beuys, Joseph Jose…  1990           0            0             0
#> # ℹ 4,274 more rows
#> # ℹ 949 more variables: tf_medium_11 <int>, tf_medium_12 <int>,
#> #   tf_medium_13 <int>, tf_medium_133 <int>, tf_medium_14 <int>,
#> #   tf_medium_15 <int>, tf_medium_151 <int>, tf_medium_16 <int>,
#> #   tf_medium_160 <int>, tf_medium_16mm <int>, tf_medium_18 <int>,
#> #   tf_medium_19 <int>, tf_medium_2 <int>, tf_medium_20 <int>,
#> #   tf_medium_2000 <int>, tf_medium_201 <int>, tf_medium_21 <int>, …

tidy(tate_rec, number = 2)
#> # A tibble: 1 × 3
#>   terms  value id      
#>   <chr>  <chr> <chr>   
#> 1 medium NA    tf_kCw2O
tidy(tate_obj, number = 2)
#> # A tibble: 1 × 3
#>   terms  value     id      
#>   <chr>  <chr>     <chr>   
#> 1 medium raw count tf_kCw2O
# }