Time series classification with random convolution kernels based transforms: pooling operators and input representations matter

Code

• github.com/ANR-MYEL/SelF-Rocket
  Officialnone★ 15
• github.com/msd-irimas/multi_comparison_matrix
  none★ 22

Abstract

This article presents a new approach based on MiniRocket, called SelF-Rocket, for fast time series classification (TSC). Unlike existing approaches based on random convolution kernels, it dynamically selects the best couple of input representations and pooling operator during the training process. SelF-Rocket achieves state-of-the-art accuracy on the University of California Riverside (UCR) TSC benchmark datasets.

Tasks

Reproductions