Triton & nvidia Tensor Core for sparse math #289
Answered
by
ptillet
blefaudeux
asked this question in
Q&A
-
|
Open question, does Triton support the specialized A100 hw around sparse tensor operations ? Not an issue really, just for context |
Beta Was this translation helpful? Give feedback.
Answered by
ptillet
Sep 20, 2021
Replies: 1 comment
-
|
It actually doesn't at this point, mostly because my focus has been more geared towards training at this point, and sparse tensor cores have little use there at the moment |
Beta Was this translation helpful? Give feedback.
0 replies
Answer selected by
blefaudeux
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
It actually doesn't at this point, mostly because my focus has been more geared towards training at this point, and sparse tensor cores have little use there at the moment