mims-harvard/Raincoat

Domain Adaptation for Time Series Under Feature and Label Shifts

40
/ 100
Emerging

When you have time series data from one context (like activity sensor data from one group of people) and need to apply a machine learning model to similar data from a different context (like sensor data from a new group, or a different sensor type), Raincoat helps you adapt that model. It takes your pre-trained model and time series data from both your original and new contexts to produce a more accurate model for the new situation, even if the patterns or categories of events have changed.

133 stars. No commits in the last 6 months.

Use this if you need to apply a time series classification model to new data where the underlying features (how the data looks) or the labels (what the data represents) have subtly or significantly shifted from your original training data.

Not ideal if your data is not time series, if you don't have a pre-trained model, or if you're looking for a simple plug-and-play solution without any configuration.

time-series-analysis sensor-data activity-recognition predictive-modeling healthcare-analytics
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 14 / 25

How are scores calculated?

Stars

133

Forks

16

Language

Jupyter Notebook

License

MIT

Last pushed

Jun 21, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/mims-harvard/Raincoat"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.