How can I convert a numeric matrix(a) to a 0 and 1 matrix(b)?
6 Ansichten (letzte 30 Tage)
Ältere Kommentare anzeigen
Hi everyone,
suppose I have a matrix:
a = [3;
1;
4;
2]
Then I want it to be:
b = [0 0 1 0;
1 0 0 0;
0 0 0 1;
0 1 0 0]
Explanation of first row in matrix "b" (how it's created, manual):
if a(1)=1
b(1,1)=1
elseif b(1,1)=0
end
if a(1)=2
b(1,2)=1
elseif b(1,2)=0
end
if a(1)=3
b(1,3)=1
elseif b(1,3)=0
end
if a(1)=4
b(1,4)=4
elseif b(1,4)=0
end
and so on for rows 2,3 and 4 in matrix a.
I'm looking for a automatic loop or function in the Matlab, because real size of matrix a is 1000 rows.
Please help me, Thanks.
0 Kommentare
Akzeptierte Antwort
Geoff Hayes
am 25 Apr. 2014
Hi Mohammed,
Why not just use a simple loop?
r=size(a,1); % get the number of rows in a (first dimension,1)
b=zeros(r,r); % initialize the b matrix to all zeros
for i=1:r
b(i,a(i))=1; % set the ith row of b with the a(i) column set to one
end
Geoff
5 Kommentare
Geoff Hayes
am 25 Apr. 2014
Hi Mohammed,
So this isn't much different from before - your b now has 8 rows (since a has 4 rows) and all rows of b are paired. The first row of a either populates row 1 or row 2 of b, the second row of a populates either row 3 or row 4 of b, etc. with the ith row of a populating either row 2i-1 or 2i of b:
for i=1:r
if a(i,2)==1
b(2*i-1,a(i,1))=1;
else
b(2*i,a(i,1))=1;
end
end
That should do it!
Geoff
Weitere Antworten (1)
Jos (10584)
am 25 Apr. 2014
No need for for-loops. This shows the power of linear indexing:
a = [3 1 4 2] % these specify column indices
b = zeros(numel(a), max(a))
idx = sub2ind(size(b), 1:numel(a), a(:).') % linear indices
b(idx) = 1
As for your second question:
a = [3 2; 1 1; 4 2; 2 1]
b = zeros(2*size(a,1), max(a(:,1)))
colix = a(:,1).'
rowix = 2*(1:size(a,1)) - (a(:,2)==1).'
idx = sub2ind(size(b), rowix, colix)
b(idx) = 1
Siehe auch
Kategorien
Mehr zu Loops and Conditional Statements 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!