Page MenuHomePhabricator

[mlir][sparse] use consistent type for COO object and sparse tensor storage
ClosedPublic

Authored by aartbik on Aug 2 2021, 2:13 PM.

Details

Summary

There was a slightly mismatch between the double COO and actual numerical
type in the final sparse tensor storage (due to external formats always
using double). This minor revision removes that inconsistency by using a
properly typed COO and casting during the "add" method instead. This also
prepares alternative ways of initializing the COO object.

Diff Detail

Event Timeline

aartbik created this revision.Aug 2 2021, 2:13 PM
aartbik requested review of this revision.Aug 2 2021, 2:13 PM
gussmith23 accepted this revision.Aug 2 2021, 3:18 PM
This revision is now accepted and ready to land.Aug 2 2021, 3:18 PM