Skip to content

GitLab

Switch branch/tag
  1. 25 Aug, 2020 1 commit
    • Paul Fultz II's avatar
      Improve layernorm performance (#613) · 56b3bf58
      Paul Fultz II authored 
      * Use increment instead of division to compute register offset

* Formatting

* Limit layernorm to 1024 elements

* Formatting

* Add verification to driver

* Formatting

* Remove early return

* Use block_size 256

* Vectorize the kernel

* Formatting

* Convert to vector type

* Add layernorm tests

* Formatting

* Formatting

* Refactor layernorm to run both algos

* Formatting

* Fix compile error

* Fix tidy warnings

* Formatting

* Add layernorm function

* Formatting
      56b3bf58
  3. 27 Nov, 2018 1 commit
  5. 14 Nov, 2018 1 commit
  7. 06 Nov, 2018 4 commits
  9. 18 Sep, 2018 1 commit
  11. 17 Sep, 2018 3 commits
  13. 30 Aug, 2018 1 commit
  15. 29 Aug, 2018 2 commits