-
It would be great to add ADMM as an algorithm.
-
**Commit:**
bdfc973d4ed46ff2b3ac5765c2063543b6425a9a
To reproduce the issue:
```
dynamatic> set-dynamatic-path ./dynamatic; set-src ./dynamatic/integration-test/admm/admm.c; set-clock-perio…
-
Hi, are there any plans to port [OSQP](https://jaxopt.github.io/stable/_autosummary/jaxopt.OSQP.html#jaxopt.OSQP), or other ADMM solvers from JaxOpt to Optax soon?
Thanks :)
-
The `admm_single_lambda()` function would be better as `void`. That way it can update `theta`, `alpha`, `u` as well as return an objective value and count iterations. This would save some other comput…
-
@nHackel:
I am not sure if I am overlooking something, but you changed the output of the ADMM `converged` function to always return `false`:
https://github.com/JuliaImageRecon/RegularizedLeastS…
-
Hi, thanks for your code sharing. I simply search how you use ADMM in all code files, but I only find ADMM class definition, it is not used anywhere in all the code files, a bit confused, you may can …
-
Support has already been added for (DFT) diagonalizable operators, but block diagonalizable operators, and explicit matrix operators, as discussed in the solver design document, have yet to be impleme…
-
It appears that we re-code the objective function for each source signal in ADMM to be sum square errors with no regularization. Below is a line 217 from CSSS.py, it defines the objective function for…
-
I have noticed that in the current version of the code, the ADMM algorithm does not seem to converge. However, in the arXiv version of the paper, the ADMM algorithm is reported to converge successfull…
-
Are you interested in adding support for the complex PSD cone? It shows up all the time in quantum information problems, and is much more efficient than mapping the problem to the real PSD cone via
`…