Summary table of (X, y) data — get_data

Provides a summary of the given (X, y) data in table form. Serves as a wrapper function around skimr::skim(), which skims a data frame and returns a broad overview of useful summary statistics. This wrapper can currently handle columns of type "factor", "numeric", "character", "logical", "complex", "Date", and "POSIXct". All other column types are ignored.

get_data_summary(
  X,
  y = NULL,
  skim_out = NULL,
  digits = 2,
  sigfig = FALSE,
  features = NULL,
  max_features = 1000,
  html = knitr::is_html_output(),
  ...
)

Arguments

X: Data matrix or data frame.
y: Response vector.
skim_out: (Optional) cached output of `skimr::skim()`. Specify if the skim output has been pre-computed in order to reduce computation.
digits: Number of digits to display for numeric values
sigfig: Logical. If TRUE, digits refers to the number of significant figures. If FALSE, digits refers to the number of decimal places.
features: (Optional) vector of features to include in summary. Default (NULL) is to include all features.
max_features: (Optional) maximum number of features to include in summary. Only used if features = NULL. Default is 1000. If the number of features in X exceeds `max_features`, the features kept in the summary are chosen randomly.
html: Logical indicating whether or not the output is an html table or a latex table.
...: Additional arguments to pass to vthemes::pretty_DT() if html = TRUE or vthemes::pretty_kable() if html = FALSE.

Value

Returns an html table (i.e., the output of vthemes::pretty_DT()) or a latex table (i.e., the output of vthemes::pretty_kable()), containing a broad overview of summary statistics for each data column.