Cleaning cells
in [Matlab]
|
Prev: different matrix operations on columns without loops
Next: Angle Recovery in Fourier Mellin Image Registration
From: james bejon on 26 May 2010 09:13 Dear All, I'm sure this is really easy, but it's driving me nuts. I can't find a way of replacing the NaNs in a cell array like c = {'geasgeas', NaN, 1, 'feafea'}; with blanks, i.e. with {''}. My first attempt was to use cellfun(@isnan, c, 'Uni', 0) as an index, but unfortunately this expression interrogates every element of every cell in the array. And cellfun(@isnumeric, c, 'Uni', 0) returns 1 for NaNs, so doesn't help much either.
From: Steven Lord on 26 May 2010 09:25 "james bejon" <jamesbejon(a)yahoo.co.uk> wrote in message news:htj6p1$1i$1(a)fred.mathworks.com... > Dear All, I'm sure this is really easy, but it's driving me nuts. I can't > find a way of replacing the NaNs in a cell array like > > c = {'geasgeas', NaN, 1, 'feafea'}; > > with blanks, i.e. with {''}. My first attempt was to use > > cellfun(@isnan, c, 'Uni', 0) > > as an index, but unfortunately this expression interrogates every element > of every cell in the array. And > > cellfun(@isnumeric, c, 'Uni', 0) > > returns 1 for NaNs, so doesn't help much either. Your first attempt was close -- you just need an ALL in there. c = {'geasgeas', NaN, 1, 'feafea'}; r = cellfun(@(x) all(isnan(x)), c); c(r) = {''} -- Steve Lord slord(a)mathworks.com comp.soft-sys.matlab (CSSM) FAQ: http://matlabwiki.mathworks.com/MATLAB_FAQ To contact Technical Support use the Contact Us link on http://www.mathworks.com
From: james bejon on 26 May 2010 10:53 Very nice. Thanks.
From: us on 26 May 2010 11:41
"james bejon" <jamesbejon(a)yahoo.co.uk> wrote in message <htj6p1$1i$1(a)fred.mathworks.com>... > Dear All, I'm sure this is really easy, but it's driving me nuts. I can't find a way of replacing the NaNs in a cell array like > > c = {'geasgeas', NaN, 1, 'feafea'}; > > with blanks, i.e. with {''}. My first attempt was to use > > cellfun(@isnan, c, 'Uni', 0) > > as an index, but unfortunately this expression interrogates every element of every cell in the array. And > > cellfun(@isnumeric, c, 'Uni', 0) > > returns 1 for NaNs, so doesn't help much either. one of the many solutions c = {'a',nan,1,'bb',[1,nan,2]}; % <- note: added one more CELL... r1=c(~cellfun(@(x) all(isnan(x)),c)) % <- ALL() % r1 = 'a' [1] 'bb' [1x3 double] % -and- r2=c(~cellfun(@(x) any(isnan(x)),c)) % <- ANY() % r2 = 'a' [1] 'bb' us |