Learn R Programming

sparklyr (version 1.8.5)

ft_bucketizer: Feature Transformation -- Bucketizer (Transformer)

Description

Similar to R's cut function, this transforms a numeric column into a discretized column, with breaks specified through the splits parameter.

Usage

ft_bucketizer(
  x,
  input_col = NULL,
  output_col = NULL,
  splits = NULL,
  input_cols = NULL,
  output_cols = NULL,
  splits_array = NULL,
  handle_invalid = "error",
  uid = random_string("bucketizer_"),
  ...
)

Value

The object returned depends on the class of x. If it is a spark_connection, the function returns a ml_estimator or a ml_estimator object. If it is a ml_pipeline, it will return a pipeline with the transformer or estimator appended to it. If a tbl_spark, it will return a tbl_spark with the transformation applied to it.

Arguments

x

A spark_connection, ml_pipeline, or a tbl_spark.

input_col

The name of the input column.

output_col

The name of the output column.

splits

A numeric vector of cutpoints, indicating the bucket boundaries.

input_cols

Names of input columns.

output_cols

Names of output columns.

splits_array

Parameter for specifying multiple splits parameters. Each element in this array can be used to map continuous features into buckets.

handle_invalid

(Spark 2.1.0+) Param for how to handle invalid entries. Options are 'skip' (filter out rows with invalid values), 'error' (throw an error), or 'keep' (keep invalid values in a special additional bucket). Default: "error"

uid

A character string used to uniquely identify the feature transformer.

...

Optional arguments; currently unused.

See Also

Other feature transformers: ft_binarizer(), ft_chisq_selector(), ft_count_vectorizer(), ft_dct(), ft_elementwise_product(), ft_feature_hasher(), ft_hashing_tf(), ft_idf(), ft_imputer(), ft_index_to_string(), ft_interaction(), ft_lsh, ft_max_abs_scaler(), ft_min_max_scaler(), ft_ngram(), ft_normalizer(), ft_one_hot_encoder_estimator(), ft_one_hot_encoder(), ft_pca(), ft_polynomial_expansion(), ft_quantile_discretizer(), ft_r_formula(), ft_regex_tokenizer(), ft_robust_scaler(), ft_sql_transformer(), ft_standard_scaler(), ft_stop_words_remover(), ft_string_indexer(), ft_tokenizer(), ft_vector_assembler(), ft_vector_indexer(), ft_vector_slicer(), ft_word2vec()

Examples

Run this code
if (FALSE) {
library(dplyr)

sc <- spark_connect(master = "local")
iris_tbl <- sdf_copy_to(sc, iris, name = "iris_tbl", overwrite = TRUE)

iris_tbl %>%
  ft_bucketizer(
    input_col = "Sepal_Length",
    output_col = "Sepal_Length_bucket",
    splits = c(0, 4.5, 5, 8)
  ) %>%
  select(Sepal_Length, Sepal_Length_bucket, Species)
}

Run the code above in your browser using DataLab