Closed briandesilva closed 4 years ago
X = [n,t] is how I would write my data matrix for DMD to make sure I have access to left multiplication, AX. I think PySINDy should make its choice explicit because of this unfortunate discrepancy in the two major algo's the group uses. Is there a change we should make to derivative to make a kind of global setter/getter or is the current access enough?
I think the current implementation is fine, though I think I found a bug. On this line of derivative.py, I think you forgot to pass the axis
argument to d
.
I noticed that the default value for
axis
throughoutderivative.py
is 1. What's the rationale for using 1 instead of 0? The only reason I ask is because we've got to setaxis=0
when usingderivative
objects in PySINDy (to conform to sklearn's conventions and those in the original SINDy paper, the 0 axis corresponds to time and the 1 axis to features).