GerbenBeintema / deepSI

Dynamical System Identification using python incorporating numerous powerful deep learning methods. (deepSI = deep System Identification)
Other
43 stars 16 forks source link

Missing dependency #7

Closed JonasWeigand closed 2 years ago

JonasWeigand commented 2 years ago

Hi Gerben,

Maybe one import is missing. The cnt.dare() function refers to the python control toolbox.

https://github.com/GerbenBeintema/deepSI/blob/05443f6a23cc04b4c0f0aeb3e610840810de13e9/deepSI/fit_systems/ss_linear.py#L146

In the original, it is imported via

https://github.com/CPCLAB-UNIPI/SIPPY/blob/0a542bd259600d486b5a90a8a892e598fe6a0878/sippy/OLSims_methods.py#L15

and there via

https://github.com/CPCLAB-UNIPI/SIPPY/blob/0a542bd259600d486b5a90a8a892e598fe6a0878/sippy/functionsetSIM.py#L8

which refers to the python control package

https://github.com/python-control/python-control

The try / except line catches the error, so it is not critical that Kalman cannot be estimated. Still, it might be better for BLA include the control package dependency or to delete cnt.dare().