Introduction

This repo contains the experiment code belongs to the paper "An end-to-end machine learning approach with explanation for time series with varying lengths".

It shows some code snippets for python, to use cnn with masking layer for timeseries with varying length and use CAM for visualisation of the activation result of the Gloabel Average Pooling (GAP) layer.

Requirements

• python 3.8.10
• pandas 1.3.3
• numpy 1.23.5
• matplotlib 3.4.3
• keras 2.6.0
• tensorflow 2.6.0+nv

Journal ResearchGate Neural Computing and Applications

Citation

If you find CNN code snippets useful to your research, please cite our work:

@Article{Schneider2024,
  author    = {Schneider, Manuel and Greifzu, Norbert and Wang, Lei and Walther, Christian and Wenzel, Andreas and Li, Pu},
  journal   = {Neural Computing and Applications},
  title     = {An end-to-end machine learning approach with explanation for time series with varying lengths},
  year      = {2024},
  issn      = {1433-3058},
  month     = feb,
  doi       = {10.1007/s00521-024-09473-9},
  publisher = {Springer Science and Business Media LLC},
}