Datasets used in the demo script

[manner]

We're currently porting and integrating MStream to the online machine learning framework river and were trying to reproduce the results with the given datasets. Unfortunately, the run.sh script doesn't work out of the box but requires preprocessed variants of the used datasets, which are split up into categorical, numerical, and timestamp parts. Do you have those preprocessed files available somewhere?

[bhatiasiddharth]

Thank you for your interest in integrating MStream into river. Sharing the processed datasets over email.