When does one need to index for plotting

1 visualización (últimos 30 días)
am
am el 28 de Mzo. de 2019
Comentada: am el 28 de Mzo. de 2019
I refer to this question:
I previously never indexed when I filled with values in a while loop and the values were saved in the vector.
But in that case I needed to do :
iterations(iter) = iter
error(iter) = norm(x-x_ref)
to make sure the values were saved in the vector and adapt automatically.
Can someone explain why is that, when manipulating vectors, sometimes you need to iterate with an index and sometimes not?

Respuesta aceptada

Walter Roberson
Walter Roberson el 28 de Mzo. de 2019
You do not need indexing when you have vectorized code. For example,
x = linspace(0, 2*pi, 500);
y = sin(x) .* cos(x.^2);
plot(x, y);
All of the values are produced at the same time.
You do need indexing when you are producing only part of the answer at a time. For example,
x = linspace(0, 2*pi, 500);
y = zeros(size(x));
for K = 1 : numel(x)
y(K) = sin(x(K)) .* cos(x(K).^2);
end
  2 comentarios
am
am el 28 de Mzo. de 2019
Thank you.
So I need indexing in every loop?
Walter Roberson
Walter Roberson el 28 de Mzo. de 2019
No. Consider
t = linspace(0, 2*pi, 500);
x = zeros(size(t));
for K = 1:2:5
x = x + sin(2*pi*t*K);
end
This is a loop, but each iteration of the loop produces a full vector of results and so does not need any indexing.

Iniciar sesión para comentar.

Más respuestas (1)

am
am el 28 de Mzo. de 2019
I am sorry, I don't understand. Does each iteration rewrites x?
  5 comentarios
Walter Roberson
Walter Roberson el 28 de Mzo. de 2019
The first line creates a 15 x 2 matrix of normally distributed random numbers (with mean 0 and standard deviation 1)
The second line calculates the sine of the first column and replaces the first column with those sines. Indexing was needed on the right side of the = in order to fetch only the first column, and indexing was needed on the left side of the = in order to assign to only the first column.
The third line calculates the exponential of the second column and replaces the second column with those exponentials. Indexing was needed on the right side of the = in order to fetch only the second column, and indexing was needed on the left side of the = in order to assign to only the second column.
The end result is 15 x 2.
am
am el 28 de Mzo. de 2019
Thank you, it's clearer.

Iniciar sesión para comentar.

Categorías

Más información sobre Matrix Indexing en Help Center y File Exchange.

Etiquetas

Community Treasure Hunt

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

Start Hunting!

Translated by