IRLS-MAKS

[sunghoon031]

Hi. Nice work!

Could you tell me which part in the code you implemented IRLS you described in the paper?

I was looking for it, but couldn't find it.

[gabmoreira]

Hi, thank you!
At the moment this repo only contains the C++ implementation of MAKS, which was the main focus of the article.

As mentioned in the paper, the IRLS version was implemented in MATLAB. Since it performs multiple spectral decompositions, it is computationally expensive. I will probably be releasing the MATLAB code of both MAKS and IRLS-MAKS in the future :-)

[sunghoon031]

I see. Thanks! When you release your MATLAB code, please let me know by commenting on this issue. I'm very interested in trying it out!

[sunghoon031]

(writing here because I couldn't get to you by email...)

Hi. I'm currently working on the similar problem, and I wanted to compare my results to yours for a benchmark purpose. I followed your text in the paper and tried to replicate the results for IRLS-MAKS, but couldn't get the results as good as the ones reported in your paper. Do you mind sharing your MATLAB code for IRLS? (If you don't want to share it with public, maybe you could share it with me privately?) I will make sure to cite your work if I get to publish my work. Please let me know.