search

RobotPsychologist / bg_control

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.
https://blood-glucose-control.streamlit.app/
18 stars 42 forks source link

Add side by side histograms (overlay) #178

Closed Tony911029 closed 12 hours ago

RobotPsychologist commented 2 hours ago

Hey @Tony911029 should you be adding .devcontainer to your gitignore?

Also I discussed with @fkiraly today we should try to extend these plots to a sort of "temporal" histogram called a fan chart.

https://en.wikipedia.org/wiki/Fan_chart_(time_series)

So what we need is another function that allows us to specify the meal length window, and the time delta for first differences, the data will need to be split up or chunked into evenly spaced windows that are the same length as the meal length, then we will plot the temporal histogram like the fan chart where the start of each window corresponds to at t=0.

Tony911029 commented 1 hour ago

ok! I am going to move this conservation to the actually ticket!