How to store columns of different lengths using cells

3 Ansichten (letzte 30 Tage)
Fede C 2018 London
Fede C 2018 London am 11 Mär. 2019
Bearbeitet: per isakson am 12 Mär. 2019
Dear all,
I understand that in order to store column vectors of different length one ought to use cells. I fail to see my mistake in attempting to do this in the code below. I get the following error message: 'Array indices must be positive integers or logical values' but I don't really know what that means. Here's my attempt:
clc;
clear;
[~,~,raw] = xlsread( '21series',8) ;
dates = cellfun( @(x) sscanf( x, 'Q%d %d' ), raw(2:end,1), 'UniformOutput', false ) ;
dates = horzcat( dates{:} ).' ;
headers = raw(1,2:end) ;
select = raw(2:end,2:end) ; % Cell array mixed str/values.
select(cellfun( @ischar, select )) = {NaN} ; % Cell array NaN/values
select = cell2mat( select ) ; % Numeric array.
[nnn,kkk] = size(select);
for colidtrim=1:kkk;
datalogical(:,colidtrim)=~ isnan(select(:,colidtrim));
end
newData=cell(1,kkk);
for colidback2num=1:kkk
trimtempstoreCol = select(~isnan(select(:,colidback2num)),colidback2num);
newData{trimtempstoreCol};
end
Any suggestions?
Many thanks
  1 Kommentar
Bob Thompson
Bob Thompson am 12 Mär. 2019
Which line specifically gives the error?
Is 'trimtempstoreCol' always a postive integer?

Melden Sie sich an, um zu kommentieren.

Akzeptierte Antwort

per isakson
per isakson am 12 Mär. 2019
Bearbeitet: per isakson am 12 Mär. 2019
I assume that you intend to put successive values of trimtempstoreCol into cells of newData.
To do that, replace
newData{trimtempstoreCol};
by
newData( 1, colidback2num ) = {trimtempstoreCol};
or by
newData{ 1, colidback2num } = trimtempstoreCol;
That will make your code run.
PS: datalogical is never used.

Weitere Antworten (0)

Kategorien

Mehr zu Data Type Conversion finden Sie in Help Center und File Exchange

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by