search

vedant-sanil / ADLES

Implementation of the Adjoint Least-Squares method (ADLES) for estimating an asymmetric vocal fold oscillation model.
1 stars 0 forks source link

Difference in grouping #1

Open neelavd opened 9 months ago

neelavd commented 9 months ago

Hey thank you so much for your implementation of ADLES! I was looking through the code for the adjoint model and I noticed a difference between the equations in the code and the equations presented in the paper. In particular in the paper we have the equations $\ddot{\lambda} + (2\beta\xi_{r}\dot{\xi}_{r} + 1 - \frac{\Delta}{2})\lambda + 2\tilde{c}dR = 0$ (and similarly for $\eta$ and $\xi_{l}$). In the code we seem to have $-2\tilde{c}dR - (2\beta\xi_{r}(\dot{\xi}_{r} + 1 - \frac{\Delta}{2}))\lambda - \ddot{\lambda} = 0$, and what I'm curious about in particular is the difference in grouping between $(2\beta\xi_{r}\dot{\xi}_{r} + 1 - \frac{\Delta}{2})$ and $(2\beta\xi_{r}(\dot{\xi}_{r} + 1 - \frac{\Delta}{2}))$, as these don't seem equivalent. Is this intentional?

vedant-sanil commented 9 months ago

Hi, it's been a while since I implemented this. Give me some time to go over my code to let you know if it was intentional or not. Out of curiosity, are you using ADLES for a downstream application or just looking to replicate results in the paper?

neelavd commented 9 months ago

Ultimately I want to use it for a downstream application, but my first step is to replicate the paper's results. Would you possibly also have a copy of the FEMH dataset? I can't seem to find any downloads of it anymore, but I find a lot of papers referencing it!