Avoid loop for equality

I have a variable
h = 2:2:8
I want to have 4 output variables depending on the 4 differnt h
A = y(s==h(1),:)
As you can see if I have it like this I need to do it 4 times for each h. Is there a possibility to do it like this in one step?
[A1, A2, A3, A4] = y(s==h,:)

6 Kommentare

Adam
Adam am 3 Jul. 2018
What are y and s?
Stef
Stef am 3 Jul. 2018
y is a matrix and s is a column of the matrix
Rik
Rik am 3 Jul. 2018
How would you code this in a loop? If you show that in code, it will be easier to find a solution without a loop (or explain why that doesn't exist).
Stef
Stef am 3 Jul. 2018
Bearbeitet: Stef am 3 Jul. 2018
A=zeros(1,4)
for h=1:4
A(h) = y(s==h,:)
end
Rik
Rik am 3 Jul. 2018
Try to write a MWE. This way we can recreate your desired result. The code you wrote is probably not what you want (as it overwrites the result every loop iteration), nor is it self-contained. We can't run it, and even if we could, we would not get the result you want.
Rik
Rik am 3 Jul. 2018
Despite your edit, this is still not a MWE. We don't have y or s, so we still cannot run this. I would also agree with Guillaume that it is a bad idea to number variables: writing A{1} A{2} A{3} instead of A1 A2 A3 will allow you to easily loop over all elements.

Melden Sie sich an, um zu kommentieren.

Antworten (1)

Guillaume
Guillaume am 3 Jul. 2018

1 Stimme

Do not number variables. Creating variables A1, A2, ... is a bad idea, it will make for slow, hard to understand and difficult to debug code. If you start numbering variables it means that they're somehow related and would be better stored together in a single container that you can index easily.
It turns out you already have such a container, your y and you're probably better off keeping it as is. If you really need to you can split y into a cell array according to s but it's probably going to make subsequent calculations harder rather than easier:
[~, ~, rowid] = unique(s);
splity = accumarray(rowid, (1:numel(s))', [], {@(rid) y(rid, :)})

Kategorien

Mehr zu Performance and Memory finden Sie in Hilfe-Center und File Exchange

Gefragt:

am 3 Jul. 2018

Kommentiert:

Rik
am 3 Jul. 2018

Community Treasure Hunt

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

Start Hunting!

Translated by