Vectorising piecewise quadratic interpolation function

5 次查看(过去 30 天)
I am struggling with the vectorisation of the following code. Could you please help me?
function v = polyinterp(x,y,u)
n = length(x);
v = zeros(size(u));
for k = 1:n
w = ones(size(u));
for j = [1:k-1 k+1:n]
w = (u-x(j))./(x(k)-x(j)).*w;
end
v = v + w*y(k);
end
end
  2 个评论
darova
darova 2021-8-11
Maybe in this case the code is more readable without vectorising
Su-Yeon Chung
Su-Yeon Chung 2021-8-11
Thanks for your comment! Even it's a better code, I want to know what the vertorised version would be.

请先登录,再进行评论。

回答(1 个)

darova
darova 2021-8-20
Here is the vectorized version (not tested)
xkj0 = x-x';
xkj0 = xkj0.*eye(size(xkj0)) + eye(size(xkj0)); % make diagonal elements all ones (dividing by zero)
Xkj = repmat(xkj0,[1 1 length(u)]); % 3D matrix of differences xk-xj
U = repmat(reshape(u,1,1,[]),length(x)*[1 1]); % 3D matrix u
X = repmat(x(:),1,length(x),3); % 3D matrix x
W = (U-X)./Xkj;
Y = repmat(y(:),1,length(y),3); % 3D matrix y
temp = prod(W.*Y,2); % don't know direction of multiplication (rows?columns?)
v = sum(temp,1); % summing rows (maybe columns)?

类别

Help CenterFile Exchange 中查找有关 Programming 的更多信息

标签

产品


版本

R2020a

Community Treasure Hunt

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

Start Hunting!

Translated by