Jacob Stevens-Haas

I'm sorry, you would need to troubleshoot further. That code involves heavy use of sympy, which I've never used. Specifically, can you see where pysindy provides something unexpected?

Are you using `model.predict()`? If not, it's a new issue. Regardless, I'd need to know what is meant by "resolutions", what is the exact error you get, and do you...

Yeah, I think `SINDyPI` does not provide the ability to simulate. You would need to extract the `coef_` that correspond to the model you want, rearrange terms symbolically into a...

Great question. You're on the right approach - either directly changing the SR3 code or subclassing and overriding the _objective() method. I don't know why your change to that line...

To add so I don't forget - apparently b is never padded with extra axes to fit A, but A does get padded with extra axes to fit b. So...

Ok thank you, the examples in there give a better sense of the issue. @rgommers The point in NEP 56 **does** fix this, as choosing whether `b` is a stack...

1. Sorry, I meant Parallel - Implicit (SINDy-PI). I only have a basic understanding of SINDy-PI, which is that, (a) differentiation $\dot x$ is added as a variable input to...

Discrete SINDy might also fit better as a subclass of `SINDy`. I believe it's incompatible with Trapping and Weak form, if not PI. In the code, most logic is couched...

@znicolaou I'm happy to lead and do this incrementally. It's partially driven by how tests fail in somewhat unpredictable places, partially driven by a goal to simplify the `SINDy` class...

@akaptano > A reminder to make these changes on a separate branch and then to PR it. :100: > By discrete SINDy, you mean fitting equations looking like e.g. x_{k+1}...