First, let's learn how to use the .Bucketizer(...) transformer. Here's the snippet that allows us to transform the Horizontal_Distance_To_Hydrology column into 10 equidistant buckets:

import pyspark.sql.functions as f
import pyspark.ml.feature as feat
import numpy as np

buckets_no = 10

dist_min_max = (
    forest.agg(
          f.min('Horizontal_Distance_To_Hydrology')
            .alias('min')
        , f.max('Horizontal_Distance_To_Hydrology')
            .alias('max')
    )
    .rdd
    .map(lambda row: (row.min, row.max))
    .collect()[0]
)

rng = dist_min_max[1] - dist_min_max[0]

splits = list(np.arange(
    dist_min_max[0]
    , dist_min_max[1]
    , rng / (buckets_no + 1)))

bucketizer = feat.Bucketizer(
    splits=splits
    , inputCol= 'Horizontal_Distance_To_Hydrology'
    , outputCol='Horizontal_Distance_To_Hydrology_Bkt'
)

(
    bucketizer
    .transform(forest)
    .select(
         'Horizontal_Distance_To_Hydrology'
        ,'Horizontal_Distance_To_Hydrology_Bkt'
    ).show(5)
)