Import One-Hot Data

Import One-Hot Data

Usage

import_onehot(
  data,
  cols,
  actor,
  session,
  window_size = 1L,
  window_type = "tumbling",
  aggregate = FALSE
)

Arguments

data

A data.frame in wide format.

cols

An expression giving a tidy selection of columns to be considered as one-hot data.

actor

An optional character string giving the column name of data containing the actor identifiers.

session

An optional character string giving the column name of data containing the session identifiers.

window_size

An integer specifying the window size for grouping.

window_type

A character string. Either "tumbling" (the default) for non-overlapping windows or "sliding" for one-step sliding window.

aggregate

A logical value that determines how multiple occurrences of the same event within a window are processed. Option TRUE aggregates multiple occurrences into a single occurrence. Option FALSE keeps all occurrences (the default).

Value

The processed data as a data.frame.

See also

Other data: import_data(), prepare_data(), print.tna_data(), simulate.group_tna(), simulate.tna()

Examples

d <- data.frame(
  actor = gl(100, 5),
  session = gl(10, 50),
  feature1 = rbinom(500, 1, prob = 0.33),
  feature2 = rbinom(500, 1, prob = 0.25),
  feature3 = rbinom(500, 1, prob = 0.50)
)
onehot1 <- import_onehot(d, feature1:feature3)
onehot2 <- import_onehot(d, feature1:feature3, "actor", "session")