Art Pelling

Wow, I can't wait to get this merged :) Great idea to adapt according to the aspect ratio! A couple of thoughts: Is there any reason why you chose linearly-spaced...

> It's just the simplest thing I could think of. Note that in `mag_plot`, the log scale is applied to the input frequency, such that in the end the initial...

> I would say that it is not too much trouble for a user to call `np.linspace` and then `mag_plot`. I am oftentimes comparing discrete-time systems with different sampling times...

> Just to clarify, is the issue you found with `freq_resp` that it works with angular frequencies instead of standard frequencies, or that it doesn't choose frequencies adaptively? Sorry, I...

> Thanks for the detailed explanation. As I see it, the main issue is that all the methods assume that `w` is given in `rad/s`, even with `Hz=True`. How about...

> That sound good @artpelling, @pmli ? Yes sounds good, will get on that soon. This really is low priority, everything should continue to work fine even if we do...

@renefritze I just wanted to test the changes locally, but all the lyapunov related tests are skipped, because I don't have pymess. The decorators ``` @skip_if_missing('SLYCOT') @skip_if_missing('PYMESS') ``` make it...

I guess this can be merged now? Do we have some workflow for reminding ourselves to add deprecation warnings for Slycot in `2023.1` or something?

Maybe the Eigensystem Realization Algorithm should be added to this list. Moreover, QuadBT http://arxiv.org/abs/2104.01006v2 provides a very interesting approach that seems to unite identification of continuous-time/discrete-time systems from frequency-domain/time-domain data....

When envisioning the integration of these data-driven MOR methods into pymor, I do not think that they fit nicely into the current ecosystem. Since they are MOR methods they should...