map.categorical.encoding: Categorical mapping tables
In XanderHorn/lazy: Automated data pre-processing and machine learning

Description Usage Arguments Value Author(s) Examples

View source: R/map_categorical_encoding.R

Creates a list of mapping tables, one for each categorical feature in the dataset. These tables include engineered features which can then be joined back to the original dataset. Feature engineering techniques include: one hot encoding, ordinal proporitonal encoding, weighted noise target mean encoding given parameter y is provided.

1
2
3

map.categorical.encoding(data, x, y = NULL, max.levels = 10,
  min.percent = 0.025, track.features = TRUE, seed = 1,
  progress = TRUE)

`data`	[required \| data.frame] Dataset containing categorical features
`x`	[required \| character] A vector of categorical feature names present in the dataset
`y`	[optional \| character \| default=NULL] The name of the target feature contained in the dataset. If no target is provided mean target encoding will not be calculated.
`max.levels`	[optional \| integer \| default=10] The maximum levels allowed for a categorical feature to create one hot encoded features
`min.percent`	[optional \| numeric \| default=0.025] The minimum proportion a categorical level is allowed to have before it is flagged as a low proportional level
`track.features`	[optional \| logical \| default=TRUE] Creates tracking features that records which categories had low proportional values present
`seed`	[optional \| integer \| default=1] The random number seed for reproducable results
`progress`	[optional \| logical \| default=TRUE] Display a progress bar