You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I was checking the DBSCAN algo , where mostly computing pairwise distance it use -2X*X**T, so for this operation currently sklearn uses the blas _gemm. where it consuming much time. If we can use the syrk which also level3 blas (https://pyclblas.readthedocs.io/en/latest/SYRK.html). Which is more optimized for this type of operations.
Is there any similar approach any one seen or tried.
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
Hello
I was checking the DBSCAN algo , where mostly computing pairwise distance it use -2X*X**T, so for this operation currently sklearn uses the blas _gemm. where it consuming much time. If we can use the syrk which also level3 blas (https://pyclblas.readthedocs.io/en/latest/SYRK.html). Which is more optimized for this type of operations.
Is there any similar approach any one seen or tried.
Thank you.
Beta Was this translation helpful? Give feedback.
All reactions