lijinpeiandLeiWang1999 242cb45787 [Example] Optimize online_softmax example (#934) ...

* [Example] Optimize online_softmax example

- Y should be output in float16.
- BN needs to be equal to N to be really online.
- On my H100 machine, this increase speedup from 1.424x to 2.788x.

* enhance

---------

Co-authored-by: LeiWang1999 <leiwang1999@outlook.com>

2025-10-04 21:59:47 +08:00

..

online_softmax.py

[Example] Optimize online_softmax example (#934)

2025-10-04 21:59:47 +08:00