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.
abdshd
commented
2 hours ago
Parent Issue: #106 skchange integration - registry
@abdshd I'm creating a meal identification project-specific issue for tracking this, but your development will be within sktime.
Your task: adding placeholder entries to skchange estimators in sktime: https://github.com/sktime/sktime/issues/7351