Hi,

I noticed that the random seed does not affect the splits of the datasets using the scaffold splitting method.

Thanks!