Why does the datatype automatically change to categorical while importing a spreadsheet?

While importing a spreadsheet using the Import Tool, the datatype of all the columns in the spreadsheet changed to "categorical". Why does this happen?

Answer 1

[MathWorks Support Team]

In the Import Tool, a column of data is interpreted as categorical if it:

1. is detected as text by the detectImportOptions function, and
2. the uniqueness of the text in that column is below a threshold, in which case, using a categorical value may be a better option for the user.

If the spreadsheet contains dashes or any such characters in cells as a placeholder for empty cell, it causes the data type detection to treat the column as text. You can observe this at the command line by executing the following commands to the attached spreadsheet "Book1.xlsx":

>> opts = detectImportOptions("Book1.xlsx");

>> opts.VariableTypes

ans =

1×10 cell array

{'char'} {'char'} {'char'} {'char'} {'char'} {'char'} {'char'} {'char'} {'char'} {'char'}

A workaround for this is to force these columns to be numeric. For example:

>> opts = detectImportOptions("Book1.xlsx");

>> opts = setvartype(opts, "double");

>> t = readtable("Book1.xlsx", opts);