As we meanwhile have functionality for loading slices from hdf5 and zarr, it makes sense to extend incremental SVD in the way that the slices being processed in incremental way are directly loaded from such a file.
use-case: you have a large data set which you do not want to load completely, but you load groups of columns (with split=0) to be processed incrementally by incremental SVD
As we meanwhile have functionality for loading slices from hdf5 and zarr, it makes sense to extend incremental SVD in the way that the slices being processed in incremental way are directly loaded from such a file.
use-case: you have a large data set which you do not want to load completely, but you load groups of columns (with split=0) to be processed incrementally by incremental SVD