Reputation: 11218
I read interesting article about correct memory usage in MATLAB. Here it is: Link at official website And here I see example:
If your data contains many zeros, consider using sparse arrays, which store only nonzero elements. The following example compares the space required for storage of an array of mainly zeros:
A = diag(1e3,1e3); % Full matrix with ones on the diagonal
As = sparse(A) % Sparse matrix with only nonzero elements
I tried to implement it in my code and find interesting moment:
A = diag(1e3,1e3)
does not create matrix with ones on the diagonal! It creates matrix of zeros with only one nonzero element:
clear A
A = diag(1e3,1e3);
find(A);
ans =
1001001
A(1001001)
ans =
1000
Ok. I read about diag
function in help and see this:
D = diag(v) returns a square diagonal matrix with the elements of vector v on the main diagonal.
Ok! So it really doesn't create diagonal matrix if v
consist of 1 element! Is it mistake at help?
BUT. One more question: why it works this way?
diag(5,5)
ans =
0 0 0 0 0 5
0 0 0 0 0 0
0 0 0 0 0 0
0 0 0 0 0 0
0 0 0 0 0 0
0 0 0 0 0 0
I expect to get matrix 5x5 with 5
value at (1,1) or (5,5). Why it creates 6x6 matrix and why 5
is a (1,6) element?
Some time ago they fix documentation:
Upvotes: 2
Views: 81
Reputation: 36710
Now reading your question again, I understood it really and have to rewrite my answer. You are refering to this part of the documentation:
A = diag(1e3,1e3); % Full matrix with ones on the diagonal
As = sparse(A) % Sparse matrix with only nonzero elements
whos
Name Size Bytes Class
A 1001x1001 8016008 double array
As 1001x1001 4020 double array (sparse)
That example is definitely wrong, probably it should be:
A=eye(1e3,1e3)
As=sparse(A);
Which creates a 1000x1000 matrix with ones on the main diagonal.
The bug is reported to mathworks
Upvotes: 1
Reputation: 2379
The article is incorrect.
A = diag(1e3,1e3);
does not produce a matrix with ones on the diagonal. The code should instead read:
A = eye(1e3,1e3);
Upvotes: 2
Reputation:
Manual: diag you are using the 2nd overloaded version of diag:
D = diag(v,k) places the elements of vector v on the kth diagonal. k=0 represents the main diagonal, k>0 is above the main diagonal, and k<0 is below the main diagonal.
So your command A = diag(5,5) will construct a matrix where the diagonal elements of 5th diagonal above the main diagonal will be equal to the vector [5]. Thus the resulting value where only A(1,6) has a value.
If you want to have a 1e3x1e3 Matrix with ones on the diagonal try
A = diag(ones(1,1e3));
Upvotes: 3