I12b unscented kalman edits #180
Conversation
Update the paper reference, the value of kappa (from 1 to 0), the allowed input values to observe, and the SquareRootUKF functions: init, filtered_cholupdate and cholupdate to cope with downdates as well as updates.
- update model parameters to parameter_set - add n_outputs to BaseCost - remove observer from FittingProblem - change solver to _solver - add evaluate and problem inputs to Observer - except all Exception from call to solver - add dataset to UnscentedKalman - update tests to match
Codecov ReportAttention:
Additional details and impacted files@@ Coverage Diff @@
## i12-unscented-kalman #180 +/- ##
========================================================
+ Coverage 94.43% 94.58% +0.15%
========================================================
Files 38 40 +2
Lines 1617 1756 +139
========================================================
+ Hits 1527 1661 +134
- Misses 90 95 +5 ☔ View full report in Codecov by Sentry. |
There was a problem hiding this comment.
Thank you @NicolaCourtier, this looks great :) I have a few suggestion below, they might be more structural things, so happy if we discuss / sort them out later, so I'll hit approve
| self.reset(inputs) | ||
|
|
||
| output = [] | ||
| if hasattr(self, "_dataset"): |
There was a problem hiding this comment.
ideally all attributes of a class should be defined in the init. If some attributes are optional then they should either be set to something, or None. Then you can test for None. Otherwise type checking becomes a pain. This is probably an issue with Problem however, so happy to leave it for later.
There was a problem hiding this comment.
Agree- this can be sorted after merge! The decision is whether we want to allow the UKF to be run without a dataset (e.g. for testing?) or make it compulsory as it will usually be required.
There was a problem hiding this comment.
I know I did make that optional, but it is a bit weird to call observe and not observe anything. If you think it should be compulsory I'm happy to go with that.
Edits to the Observer class and Kalman filter implementation.