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
This is a request tracker to add an API for sparse + sparse = sparse matrix addition API in the sparse BLAS domain in the oneMKL Specification.
The operation performed is, $C = \alpha\cdot\text{op}(A) + \beta\cdot\text{op}(B)$, where:
$A$, $B$, and $C$ are sparse matrices,
$\alpha$ and $\beta$ are scalars, and
$\text{op}(\cdot)$ is one non-transpose/transpose/conjugate-transpose.
Going by the recent overhaul in [oneMKL][spblas] Update sparse blas API #522, in all likelihood, the APIs will likely need 3 or 4 stages. The APIs could be a modified version of the soon-to-be-added Intel oneMKL sparse::omatadd APIs:
Thank you for the issues Gajanan. Do you know if there is any order of priority between the different issues #578, #584, #585 and this one?
We'll update the issues when we get to them and if no one else start working on them first.
Thank you for the issues Gajanan. Do you know if there is any order of priority between the different issues #578, #584, #585 and this one? We'll update the issues when we get to them and if no one else start working on them first.
The priorities are coincidentally in the order of their issue numbers: #578, #584, #585, and this issue, (#586). I would add that the first two are high priority and the last two are lower priority.
This is a request tracker to add an API for sparse + sparse = sparse matrix addition API in the sparse BLAS domain in the oneMKL Specification.
sparse::omatadd
APIs:The text was updated successfully, but these errors were encountered: