Replace zeros with consecutive numbers in a vectors
3 Ansichten (letzte 30 Tage)
Ältere Kommentare anzeigen
Eli Dim
am 17 Jul. 2015
Bearbeitet: Azzi Abdelmalek
am 17 Jul. 2015
I have a large column vector which looks like this:
A=[1;2;3;4;5;1;5;6;..................;2305;0;0;0;0]
I want to replace the zeros in the end so the new vector looks like this:
A=[1;2;3;4;5;1;5;6;..................;2305;2306;2307;2308;2309]
How can I do this without a loop? The vector changes its size after every iteration, but I always have a couple of zeros in the end.
0 Kommentare
Akzeptierte Antwort
Guillaume
am 17 Jul. 2015
Bearbeitet: Guillaume
am 17 Jul. 2015
There are many ways you could do this. If the zeros are always at the end with no zero beforehand, this will work:
A = [nonzeros(A); A(find(A == 0, 1)-1) + (1:sum(A==0))']
1 Kommentar
Guillaume
am 17 Jul. 2015
And if there can be zeros before the end, this only replaces the zeros at the very end:
lastnumberidx = find(A~=0, 1, 'last');
A = [A(1:lastnumberidx); A(lastnumberidx) + (1:numel(A)-lastnumberidx)']
As I said, plenty of ways...
Weitere Antworten (1)
Azzi Abdelmalek
am 17 Jul. 2015
Bearbeitet: Azzi Abdelmalek
am 17 Jul. 2015
This takes in account all zeros
A=[1 ;2; 7;0; 0 ;4;8;0;6;7;18;0;0;0;12;0 ;0 ;13]
A=A'
idx1=A==0 ;
ii1=strfind([0 idx1 0],[ 0 1]);
jj=strfind([0 idx1 0],[1 0])-ii1;
[B,C,CC]=deal(zeros(size(A)));
B(ii1)=[0 jj(1:end-1)];
qq=cumsum(idx1).*idx1;
aa=(qq-cumsum(B)).*idx1;
[C(ii1-1),aa(ii1-1)]=deal(A(ii1-1));
CC(ii1-1)=[0 A(ii1(1:end-1)-1)];
DD=cumsum(C-CC).*idx1+aa;
A(idx1)=DD(idx1)
0 Kommentare
Siehe auch
Kategorien
Mehr zu Linear Algebra 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!