Highlights
- Arctic Code Vault Contributor
Create your own GitHub profile
Sign up for your own profile on GitHub, the best place to host code, manage projects, and build software alongside 50 million developers.
Sign upPinned
855 contributions in the last year
Activity overview
Contributed to
PaddlePaddle/Paddle,
chenwhql/diagrams,
PaddlePaddle/FluidDoc
and 4 other
repositories
Contribution activity
December 2020
Created 5 commits in 1 repository
Created a pull request in PaddlePaddle/Paddle that received 3 comments
Support type promote for basic math ops (quantum required)
PR types New features
PR changes OPs
Describe 本PR主要实现了基础二元运算+,-,*,/,@的运算类型提升,类型提升示例:
Tensor('int') + Tensor('float32') -> Tensor('float32')
Tensor…
+536
−255
•
3
comments
Opened 13 other pull requests in 2 repositories
PaddlePaddle/Paddle
4
open
7
merged
1
closed
- [Complex] Add real & imag op and api for complex tensor
- Add complex dtype op (add) test example
- [Cherry-pick] fix train eval set error in static mode (#29540)
- Remove train eval mode of Tracer
- Support adam, fill_constant, scale op for complex case
- Fix Layer train eval setting failed in static mode
- [cherry-pick] support type promote for basic math ops
- change test_imperative_signal_handler to exclusive
- [Cherry-pick] Hot fix complle failed in gcc4.8 caused by complex impl (#29254)
- [Cherry-pick] Hot fix complle failed in gcc4.8 caused by complex impl (#29254)
- [WIP] support tensor scalar type promotion
- Hot fix complle failed in gcc4.8 caused by complex impl
PaddlePaddle/FluidDoc
1
open
Reviewed 29 pull requests in 2 repositories
PaddlePaddle/Paddle 28 pull requests
- update activation op on kunlun
- add conj op for complex types
- [Kunlun] PR1:Support one Kunlun card training in parallel executor
- add python version selection in musl build
- Modify CublasHandleHolder to Fix Random Unittest Failure
- [Feature] one ps (2/4)
- Fix precision problem
- Remove duplicated macro
- Check precision problem
- Enforce the error message of gather_nd
- [Dy2Stat] Support to check the shape and dtype of input_specs and input Tensors
- [Dy2Stat] 1. Fix error in _build_cond_stmt of for-range stmts. 2. Support that step value is negative in for-range stmts
- support deepcopy for Layer/Tensor/Paramerbase
- remove complex module direction (#29419)
- remove complex module direction
- Fix Unit Test: Add Sleep Time for CUDA Retry
- refine reshape grad and double grad kernel, use tensor copy async (#2…
- revert cast eigen kerenl
- remove complexvariable
- Release/2.0 rc1
- Make transpose, trace, kron, reshape, sum op support complex type
- [Cheery-Pick][Dy2stat] Add a decorator paddle.jit.not_to_static to support that not to convert a function in Dynamic-to-Static.
- [Dy2stat] Reduce Exception Type for Better Error Message
- [Dy2Stat] Fix bug: Do not use gast.Subscript to replace gast.Name in when transforming for_enumerate_loop
- fix random failed of complex matmul
- Some pull request reviews not shown.