Apply lightweight preprocessing to signal columns — filter_gazepoint

Applies conservative base-R preprocessing to selected numeric signal columns. Supported methods are moving average, rolling median, and linear detrending. The function returns new columns by default and records a processing log.

Usage

filter_gazepoint_signal(
  data,
  signal_cols,
  method = c("moving_average", "rolling_median", "detrend"),
  group_cols = NULL,
  time_col = NULL,
  window = 5,
  suffix = NULL,
  overwrite = FALSE,
  na_rm = FALSE
)

Arguments

data: A data frame.
signal_cols: Character vector of numeric signal columns.
method: One of "moving_average", "rolling_median", or "detrend".
group_cols: Optional grouping columns.
time_col: Optional numeric time column used for ordering within groups.
window: Window length for moving-average and rolling-median methods.
suffix: Optional suffix for created columns. If NULL, a suffix is derived from method.
overwrite: Logical. If TRUE, overwrite input signal columns.
na_rm: Logical. If TRUE, ignore missing values inside rolling windows.

Value

A data frame with class gazepoint_filtered_signal.