Using ALS algorithm for writing a code
Afficher commentaires plus anciens
Is ALS algorithm supported by MATLAB R2019a. What is the best source of downloading the alogorithm and the reference to write the code for analysing spectral data?
Réponses (1)
Star Strider
le 26 Avr 2019
I’ve not run this, so I have no experience with it. It’s the only MATLAB code for the algorithm that I could find.
The slightly corrected version of the function:
function z = baseline(y, lambda, p)
% Estimate baseline with asymmetric least squares
m = length(y);
D = diff(speye(m), 2);w = ones(m, 1);
for it = 1:10
W = spdiags(w, 0, m, m);
C = chol(W + lambda * D' * D);
z = C \ (C' \ (w .* y));
w = p * (y > z) + (1 - p) * (y < z);
end
Also consider:
function z = baseline(y, lambda, p)
% Estimate baseline with asymmetric least squares
m = length(y);
D = diff(speye(m), 2);w = ones(m, 1);
for it = 1:10
W = spdiags(w, 0, m, m);
C = chol(W + lambda * (D' * D));
z = C \ (C' \ (w .* y));
w = p * (y > z) + (1 - p) * (y < z);
end
The documentation for the function is in the paper.
Experiment to get the result you want.
5 commentaires
Csaba
le 2 Oct 2019
at the very beginning I would add:
if isrow(y)
y=y';
end
Timothy Loayza
le 23 Août 2020
Thanks a lot, it works very well for Raman's spectras. (lambda = 100'000, p = 0.01)
BTW : I don't know who to cite for it as the link is not working anymore.
Star Strider
le 23 Août 2020
Timothy Loayza — I actually saved it. I am attaching it here. (It was apparently open-access at the time, so I doubt that I am violating any copyright by posting it here.)
Also, with respect to Csaba’s comment, if you want to force it to be a column vector, just do this:
y(:)
.
Raimund
le 4 Nov 2025
Déplacé(e) : Star Strider
le 4 Nov 2025
Hi,
thanks. This implementation works great.
Best regards,
Raimund
Star Strider
le 4 Nov 2025
Déplacé(e) : Star Strider
le 4 Nov 2025
My pleasure!
A Vote would be appreciated!
Catégories
En savoir plus sur Smoothing and Denoising dans Centre d'aide et File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!