My code on n consecutive elements in a matrix

5 visualizzazioni (ultimi 30 giorni)
Hi, I got a question to figure out the largest sum of n consecutive elements in a matrix. The order can be in the same row, column, diagonal, or reverse diagonal. Here is my code. However, the matlab keeps telling me variables "i","j", "u", "w","z" may be unused.Could you give me a hint how to make these variables valid? Also, I guess there is a way to simply my code a little bit? Thanks!
A numerical example would be: Suppose A=[1 2; 3 4] and n=2. We need to find the largest sum of 2 consecutive elements in all directions. In this case, we have
rows: 1+2, 3+4
columns: 1+3, 2+4
diagonal: 1+4
reverse diagonal: 2+3
Obviously, the output of this function should be 3+4. How could I write a function without loops? Any hint will be appreciated.
% function x=maxsum(A, n)
[p,q]=size(A);
k=zeros(1,q*floor(p/n));
%rows
for i =1:p
j=1:q-n+1;
u=1:q*floor(p/n);
k(u)=sum(A(i,j:j+n-1));
i=i+1;
j=j+1;
u=u+1;
end
K=max(k);
%columns
y=zeros(1,p*floor(q/n));
for w=1:q-n+1
v=1:q;
z=1:p*floor(q/n);
y(z)=sum(A(w:w+n-1, v);
w=w+1;
v=v+1;
z=z+1;
end
Y=max(y);
%diagonal
t=zeros(1,p-n+1);
for h=1:p-n+1
B=A(h:h+n-1,h:h+n-1);
t(h)=diag(B);
h=h+1;
end
T=max(t);
%reverse diagonal
m=zeros(1,p-n+1);
for o=1:p-n+1
c=flip(A(o:o+n-1,o:o+n-1));
m(o)=diag(c);
o=o+1;
end
M=max(m);
%compare all
x=max([M T Y K]);
% end
  2 Commenti
Shantanu Gontia
Shantanu Gontia il 11 Lug 2018
Modificato: Shantanu Gontia il 11 Lug 2018
You are changing the looping variable inside the loop itself. In a MATLAB for, you need not increment the looping variable inside the loop, it is done automatically.
In addition, you have written j=1:q-n+1; and also, k(u)=sum(A(i,j:j+n-1));. This is inconsistent as j is a vector.
KSSV
KSSV il 11 Lug 2018
Your loop index is replaced with same variable inside loop. This is not good code. YOu need to rethink. Give us a numerical example with data, you will get help and this task can be achieved without loop.

Accedi per commentare.

Risposta accettata

Matt J
Matt J il 11 Lug 2018
Modificato: Matt J il 11 Lug 2018
To avoid loops, use convolution. For example
forwardDiags = conv2(yourMatrix,eye(3),'valid')
will compute all forward diagonal sums of length 3. Similarly,
reverseDiags = conv2(yourMatrix,fliplr(eye(3)),'valid')
will compute reverse diagonals, and so on.

Più risposte (0)

Categorie

Scopri di più su Loops and Conditional Statements in Help Center e File Exchange

Tag

Prodotti


Release

R2017b

Community Treasure Hunt

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

Start Hunting!

Translated by