Граф коммитов

5 Коммитов

Автор SHA1 Сообщение Дата
cao lei 95d65e4ec0
sync to flash attention kernel 2.5.9 and add document of how to write custom op (#757)
* sync to flash attention kernel 2.5.9

* support users to overload GetMayInplace and ReleaseMayInplace

* Undo the change for pybind11 dependency
2024-07-10 07:09:40 -07:00
Wenbing Li 97ee9eb56f
Refactor OrtxStatus to be header-only implmentation. (#720) 2024-05-17 15:40:11 -07:00
Tang, Cheng 3b889fc42f
update custom op v2 struct to be able to invoke from eager mode (#700)
Co-authored-by: Cheng Tang <chenta@a100.crj0ad2y1kku1j4yxl4sj10o4e.gx.internal.cloudapp.net>
Co-authored-by: Wenbing Li <10278425+wenbingl@users.noreply.github.com>
2024-04-30 13:53:39 -07:00
Wenbing Li f9290e8bac
Add a status class for future tokenizer API implementation (#690)
* Add a status class for future API implementation

* Update bpe_kernels.cc

* fix the ios package pipeline

* update mistral test model name
2024-04-18 21:12:14 -07:00
Wenbing Li 646462790b
Refactor the header file directory and integrate the eager tensor implementation (#689)
* refactor the header file in include folder

* fix the basic-token eager unit test case

* a more flexible way to handle string tensor shape.

* fix the unit test path issue

* remove the multi-inherits to avoid issue during pointer casting

* add api cmake build support

* undo some temporary changes

* code refinement

* fix variadic arg

* only expose the context for ort version >= 17

* fix a shape bug

* fix the cuda build issue

* change ifdef condition of GetAllocator

* finalize the ort c abi wrapper file name

* fix the iOS build break

* align gtest version with triton

* Update ext_apple_framework.cmake for iOS header files

---------

Co-authored-by: Cheng Tang <chenta@a100.crj0ad2y1kku1j4yxl4sj10o4e.gx.internal.cloudapp.net>
2024-04-17 12:58:19 -07:00