How to split NIPS-TS-SWAN and IPSTS-GECCO datasets?

[qiu69]

May I ask how you split the NIPS-TS-SWAN and IPSTS-GECCO datasets? The dataset you provided has already been processed. How did you split it? I looked at the original sources of these two datasets, as well as the Revising Time Series Outlier Detection: Definitions and Benchmarks, but couldn't find how these two datasets were sharded. I only found the complete and unshared datasets.