Skip to content

Conversation

@gongchensu
Copy link
Collaborator

@gongchensu gongchensu commented Jan 8, 2026

  1. 将add_binary分支的div,pow,mod,min,max算子迁移
  2. 将add_unary分支的abs,acos,acosh,asin,asinh,atan,atanh,ceil,cos,cosh,erf,floor,log,neg,reciprocal,round,sign,sinh,sqrt,tan算子迁移
    总共增加25个算子,重构后增加4972行代码,增加一个算子大概198行代码,其中包括测试脚本,比重构前每个算子增加减少900行代码
image image binary和unary只需要运行图中两个命令,可以分别测试完binary的5个算子的cpu和nv,以及unary的20个算子的cpu和nv

@gongchensu gongchensu requested review from a team, Ziminli and whjthu January 8, 2026 12:48
@gongchensu gongchensu force-pushed the feature/operators_add branch 3 times, most recently from 794b5aa to 8ca4b32 Compare January 15, 2026 09:09
…f,floor,log,neg,reciprocal,round,sign,sinh,sqrt,tan operator with CPU and NVIDIA implementations.
@gongchensu gongchensu force-pushed the feature/operators_add branch from ddec997 to fc92aea Compare January 19, 2026 02:21
@gongchensu gongchensu force-pushed the feature/operators_add branch from fc92aea to 2ed76b9 Compare January 19, 2026 02:43
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant