You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
has_na() to check if variables or observations in a data frame contain NA, NaN or Inf values. Convenient shortcuts for this function are complete_cases(), incomplete_cases(), complete_vars() and incomplete_vars().
total_mean() to compute the overall mean of all values from all columns in a data frame.
prcn() to convert numeric scalars between 0 and 1 into a character-percentage value.
numeric_to_factor() to convert numeric variables into factors, using associated value labels as factor levels.
Changes to functions
set_na() now also replaces different values per variable into NA.
Changed behaviour of row_sums() and missing values. row_sums() gets a n-argument and now computes row sums if a row has at least n non-missing values.