Benchmarking time series classifiers for ECG data using fast.ai

We recently open-sourced PTB-XL, a large clinical ECG dataset, see data descriptor and dataset on Physionet.

We also wrote a first benchmarking paper based on the dataset, where we compare different time series classification algorithms (implemented in fast.ai). We released our code on github. Feel free to adapt our code and to benchmark your favorite time series classification algorithm applied to ECG classification. We are happy to accept PRs with more models and/or results.

1 Like