Fully Convolutional Networks with Shapelet Features for Time Series Classification

This work has been accepted by "Applied Softe Computing". paper link

FCN-SF based on the code for "Hassan Ismail Fawaz, Germain Forestier, Jonathan Weber, Lhassane Idoumghar, and Pierre-Alain Muller. Deep learning for time series classiffication: a review. Data Mining and Knowledge Discovery, 33(4):917-963, 2019".

The experiment results are in the file "/results/results.xlsx".

The transformed datasets are stored in the fold "/data/archives/Shapelet". And our trained models are stored in the fold "/data/results".

Acknowledgements

This work was supported by the Innovation Methods Work Special Project under Grant 2020IM020100, and the Natural Science Foundation of Shandong Province under Grant ZR2020QF112.

We would like to thank Eamonn Keogh and his team, Tony Bagnall and his team for the UEA/UCR time series classification repository.