Improving short-term prandial blood glucose outcomes for people with type 1 diabetes, a complex disease that affects nearly 10 million people worldwide. We aim to leverage semi-supervised learning to identify unlabelled meals in time-series blood glucose data, develop meal-scoring functions, and explore causal machine-learning techniques.
go through individual classes inheriting from BaseDetector (after this PR) and ensure they comply with the new homogenized interface definitions. There will likely be small issues for each class. As #105 progresses, the tests will likely flag up incompatibilities with the API which also need to be fixed.
As described in issue https://github.com/sktime/sktime/issues/7323
I have carried out the most significant parts on the base class level here: https://github.com/sktime/sktime/pull/7342
What remains is:
transform
method, to be discussedBaseDetector
(after this PR) and ensure they comply with the new homogenized interface definitions. There will likely be small issues for each class. As #105 progresses, the tests will likely flag up incompatibilities with the API which also need to be fixed.Once this is done, we can make changes within
skchange
itself:skchange
to use theBaseDetector
fromsktime
as base class where possibleskchange
to use theparametrize_with_checks
utility fromsktime
for testing. This is conditional on #105.