Borrar filtros
Borrar filtros

Speed up plotting of multiple lines

23 visualizaciones (últimos 30 días)
Askeladden2
Askeladden2 el 9 de Feb. de 2023
Comentada: Voss el 10 de Feb. de 2023
Dear All Community members,
I have a question concerning speeding up the function plot.
I want to plot several (72835) lines in order to create a hexagonal pattern, where each line will be colored depending on its value represented in a third vector.
The coordinates of the lines are given in two matrices, ‘x’ and ‘y’, while the color depends on the matrix ‘yo’. These matrices are attached.
Example: Line 1
X-coordinates are given in x11 and x12 in matrix 'x'.
Y-coordinates are given in y11 and y21 in matrix 'y'.
The line has a blue-violet color palette. See Figure below.
The code I am using is given below, in addition to a plot of all the lines colored.
figure
for i=1:size(x,2)
plot(x(:,i), y(:,i), 'LineWidth', 2.5, 'Color', yo(i,:))
axis square ;
hold on
end
hold off
colormap(map)
colorbar('Position', [0.8 0.25 0.04 0.6]);
caxis([11 334])
It is very time-consuming to run, and the plot is really slow to respond if I want to zoom or etc.
Is it possible to plot this differently? E.g., Using a patch objective?
Any help is deeply appreciated.

Respuesta aceptada

Voss
Voss el 9 de Feb. de 2023
Here's an idea: plot all the line segments of a given color together as one line object. That way you have 2745 lines instead of 72835. That speeds it up quite a bit. You could speed it up more by reducing the number of unique colors in yo. The colorbar suggests you only need 12 unique colors.
load Data
nans = NaN(1,size(x,2));
xx = [x; nans];
yy = [y; nans];
[uyo,~,jj] = unique(yo,'rows');
for ii = 1:size(uyo,1)
idx = jj == ii;
xx_p = xx(:,idx);
yy_p = yy(:,idx);
line(xx_p(:),yy_p(:),'LineWidth',2.5,'Color',uyo(ii,:));
end
axis square
colormap(map)
colorbar('Position', [0.8 0.25 0.04 0.6]);
caxis([11 334])
Note that I call axis square once after the loop instead of inside the loop. That helps with speed too.
I'm also using line instead of plot for speed. If using plot as in your original, you could speed it up by calling hold on once before the loop and hold off once after the loop. But hold doesn't apply when using line.
  2 comentarios
Askeladden2
Askeladden2 el 10 de Feb. de 2023
Works like a charm.
Thank you very much.
Voss
Voss el 10 de Feb. de 2023
You're welcome!

Iniciar sesión para comentar.

Más respuestas (0)

Categorías

Más información sobre Graphics Performance en Help Center y File Exchange.

Productos


Versión

R2020a

Community Treasure Hunt

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

Start Hunting!

Translated by