Skip to content

Latest commit

 

History

History
18 lines (11 loc) · 931 Bytes

File metadata and controls

18 lines (11 loc) · 931 Bytes

Bubble_Identification_MTS

This is the official repository for JGR paper "Plasma-Sheet Bubble Identification Using Multivariate Time Series Classification".
This code consists of two parts, including "model_construction" and "predict_code".
The best trained model result is in the folder "best_model_result".

How can we use our model to predict outcomes?

One can modify the code to predict new data based on the prediction code in the “predict_code” folder.

Extension

This code is very easily to add more input parameters. One can expand on this to train one's own model.

Acknowledgement

Our models' architecture is based on the "dl-4-tsc" repository. We would like to thank the providers of the “dl-4-tsc” repository (https://github.com/hfawaz/dl-4-tsc). We would also like to thank "sktime" repository (https://github.com/sktime/sktime).

In case of any questions, please do not hesitate to contact us.