Transformer Model for Multivariate Time Series Classification: A Case Study of Solar Flare Prediction

The code is an implementation of the ATransformer Model for Multivariate Time Series Classification: A Case Study of Solar Flare Prediction. It is implemented using Tensorflow.

Overview: This repository provides the implementation of the Transformer-based MVTS Classification Framework. The repository has: The dataset can be downloaded from this link: https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/EBCFKM

Models/ contains the implementation of the Attention-based and Transformer Model for Multivariate Time Series Classification.

Baselines/ contains the baselines implementation and data preprocessing.

requirements.txt contains the libraries and tools needed to execute the models.