CELL2DSET

Construct dataset from a cell array having one feature per cell.

    A = CELL2DSET(C,F,M,L)

Input
C	Cell array with one cell per feature. Categorical features should  be given by strings in a character array or as cellstrings (one  object per cell) stored in the feature cell.  Numerical features may either be stored as character arrays, cell  strings or as a numeric vector.  In case C is a 2D cell array columns are first combined into a  cellstring or a vector.
F	Format field (e.g. 'ccncnn') for distinguishing cell character  arrays with categorical data ('c') from numeric data ('n').
M	Optional character array with symbols used for missing values. The  empty string ('') will always be interpreted as a missing value.
L	Labels that may be used for labelling A (numbers or strings, see
PRDATASET).	(optional)

Output
A	Dataset with categorical features numerically coded as indices in  the DATA field, pointing to the list on category names stored in the
FEATDOM	field, see SETFEATDOM.

Description

This routine is an alternative for PRDATASET in case data is given by  strings or numeric characters. The parameter F is optional. If not given  all character arrays and cell strings are interpreted as categorical.

Missing data (empty strings), will be coded as a NaNs.

See also

datasets, setfeatdom, feattypes, dset2cell, cat2dset, dset2cell,