ROVI-org / auto-soh / 11021043140

95%

push

github

First draft of an Offline Esimator (#87)

* Introduce interface for offline objective functions

* Cache the result of len(state)

It never changes

* Fix sign convention in test data

* Finish implemeting the MSE loss

* Add a SciPy estimator

* Move the initial guess to the objective

Keep the estimators agnostic to the problem being solved

* Initial draft of documentation

* Add offline estimation to example notebook

* Documentation and terminology changes

* Ensure scipy is OK with negative loss values

Still keep the normalization strategy

(Nice catch, @vventuri!)

* Update nomenclature in notebook

* Emphsaize initial transient state is being fit

1. Add it into the figure more prominantly
2. Identify it as one of the outputs
3. Mention that we use a starting guess for it

* Incorporate Noah's feedback

65 of 67 new or added lines in 4 files covered. (97.01%)

1352 of 1423 relevant lines covered (95.01%)

0.95 hits per line