- changed status to resolved
optimise compression of L in PLE (was PLS) decomposition
Issue #23
resolved
We know a priori in PLS how the column swaps look like such that we don't need to consider all these bit operations. This could speed up our code considerably especially for sparse-ish matrices.
Comments (1)
-
reporter - Log in to comment
optimised compression L in _mzd_pls() (fixes
#23)→ Wiki macro error: Changeset 69e8daf61bc2 not found.