guess_data_types: Guess data types

Description

This function removes any columns where there is no data and makes guesses on the data type. However, this relies on the data not being coded already. If the data has been coded, the metadata generated can be used as a template that can be modified and provided as input for process_data.

Usage

guess_data_types(data_file_path)

Value

outcome: Whether the operation was successfully performed
message: Any information, particularly when the operation fails.
data: The data after removing the columns without any data.
metadata: Automated metadata is created based on the data. However, this relies on data not previously coded, for example, if the status is coded as 0 and 1 rather than 'absent' and 'present', the variable will be recognised as a quantitative variable rather than categorical variable.
any_type: All fields with data.
quantitative: Fields recognised as quantitative.
numerical: Fields recognised as continuous.
count: Fields recognised as count. Count data is recognised from the field name. If a field name starts with 'Number of', it is considered as count data.
categorical: Fields recognised as categorical data.
nominal: Fields recognised as nominal data. All categorical data with more than two levels are recognised as nominal data.
binary: Fields recognised as binary data. All categorical data with only two levels are recognised as binary data.
ordinal: Fields recognised as ordinal data. Any categorical data with more than two levels and with the second character of all the levels being an '_' are recognised as ordinal data.
date: Fields recognised as date.
time: Fields recognised as time.

Arguments

data_file_path: The path for the data file where guessing the data types of columns is necessary.

Author

Kurinchi Gurusamy

Examples

Run this code

data_file_path <- paste0(tempdir(), "/df.csv")
write.csv(penguins, data_file_path, row.names = FALSE, na = "")
guessed_data_types <- guess_data_types(data_file_path = data_file_path)
guessed_data_types