提交 · 755438a7372991ffc8574bf8f04f264954939e43 · 机器未来 / Paddle

02 7月, 2022 2 次提交

unify cpu context, part2 (#44012) · 755438a7

由 Leo Chen 提交于 7月 02, 2022

* fix init()

* delete test_device_context

* replace CPUDeviceContext with CPUContext

* fix test_scalar

* remove dot_op.cc

* fix compile

755438a7

unify cpu context (#43989) · 09096aeb

由 Leo Chen 提交于 7月 01, 2022

* unify cpu context

* fix init()

* delete test_device_context

* fix test_scalar

09096aeb

01 7月, 2022 9 次提交
- Z
  
  template GetWeightCPUData (#44002) · b4fef397
  由 zhoutianzi666 提交于 7月 01, 2022
  
  b4fef397
- Q
  
  [DCU] fix compiling error on DTK22.04, test=develop (#43999) · eb8fc759
  由 Qi Li 提交于 7月 01, 2022
  
  eb8fc759
- S
  Make inference_api_test compile with shared lib in Windows (#43946) · b9640034
  由 Sing_chan 提交于 7月 01, 2022
```
* Revert "Revert "make inference_api_test compile with dynamic linking library (#41944)" (#43882)"

This reverts commit e6d81ddf.

* modify third_party cmake

* move SKIP_CPP_TEST return in the begining
```
  b9640034
- A
  
  [Err msg] optimize kernel call error message (#44001) · 37f2151f
  由 Aganlengzi 提交于 7月 01, 2022
  
  37f2151f
- C
  [Phi] Move all yaml to phi subdirectory (#43986) · 5dfc0cd6
  由 Chen Weihang 提交于 7月 01, 2022
```
* move yaml to phi

* fix undef var error
```
  5dfc0cd6
- L
  Addition of switch_auto_tune option for transpose op (#43310) · 53d5abe3
  由 limingshu 提交于 7月 01, 2022
```
* 2nd part of transpose update

* add switch_auto_tune option.

* add some changes according to Ci

* refine the structure of auto_tune_base.

* merge develop changes

* reset the switch_set_range and change unittest of transpose auto-tune

* change the kernel auto-tune logits
```
  53d5abe3
- Z
  [inference TRT]template GetWeightCPUData (#43993) · 76156d12
  由 zhoutianzi666 提交于 7月 01, 2022
```
* template GetWeightCPUData
```
  76156d12
- [MLU] add mlu kernel for fill_constant_batch_size_like (#43820) · 88e27a07
  由光明和真理提交于 7月 01, 2022
  
  88e27a07
- C
  
  [MLU] add rnn backward kernel. (#43969) · 3a59ede9
  由 Chenxiao Niu 提交于 7月 01, 2022
  
  3a59ede9
30 6月, 2022 14 次提交
- Z
  Move apis(digamma, dist, dot) from legacy_api.yaml to api.yaml (#43956) · f33763e3
  由 zyfncg 提交于 6月 30, 2022
```
* move standard apis to api.yaml

* revert erfinv

* delete dot_op.h

* fix dot

* rerun ci
```
  f33763e3
- Z
  
  remove decrease_axis in op_teller.cc , support them in slice (#43963) · 56ddd7c2
  由 zhoutianzi666 提交于 6月 30, 2022
  
  56ddd7c2
- W
  fused_gate_attention manual code in eager (#43897) · 73f957cf
  由 wanghuancoder 提交于 6月 30, 2022
```
* fused_gate_attention manual code in eager

* refine

* refine

* refine

* refine

* refine

* refine
```
  73f957cf
- [MLU] add mlu kernel for masked_select (#43816) · d29a1214
  由光明和真理提交于 6月 30, 2022
  
  d29a1214
- Z
  
  [MLU] add exp and exp_grad kernel (#43852) · 59d50468
  由 zhaoying9105 提交于 6月 30, 2022
  
  59d50468
- C
  Add statistic code for memory (#43960) · 52d43ca2
  由 chenjian 提交于 6月 30, 2022
```
* add code

* add unit test
```
  52d43ca2
- Z
  Revert "[GPUPS]Optimize dymf kernel (#43911)" (#43958) · 35ca3009
  由 zmxdream 提交于 6月 30, 2022
```
* Revert "[GPUPS]Optimize dymf kernel (#43911)"
```
  35ca3009
- L
  [new-exec] support runing with different scope and the same program using scope_guard (#43962) · 99a4ff8f
  由 Leo Chen 提交于 6月 30, 2022
```
* support scope_guard

* fix test
```
  99a4ff8f
- R
  Remove boost::variant for FetchResultType (#43932) · f720e231
  由 Ruibiao Chen 提交于 6月 30, 2022
```
* Remove boost::variant for FetchResultType

* Fix pybind errors
```
  f720e231
- J
  modify graph_pattern to thread_local (#43942) · 6467ca0d
  由 JingZhuangzhuang 提交于 6月 30, 2022
```
* modify graph_pattern to thread_local

* modify graph_pattern to thread_local
```
  6467ca0d
- Z
  Add new attr of fused_multi_transformer (#43730) · c2a5bb91
  由 Zhang Zheng 提交于 6月 30, 2022
```
* Add new attr of fused_multi_transformer

* fix format

* add note

* add in layer

* fixfixfixfix
```
  c2a5bb91
- C
  [phi]add relu6 kernel and yaml (#43549) · a9bba5ba
  由 chentianyu03 提交于 6月 30, 2022
```
* add relu6 kernel and yaml

* format files

* format code and fix bug

* fix build failed
```
  a9bba5ba
- C
  
  [MLU] add rnn forward kernel. (#43894) · 2616d51a
  由 Chenxiao Niu 提交于 6月 30, 2022
  
  2616d51a
- J
  
  fix deriv with inplace (#43930) · 1efc80c6
  由 Jiabin Yang 提交于 6月 30, 2022
  
  1efc80c6
29 6月, 2022 9 次提交
- Z
  
  [GPUPS]Optimize dymf kernel (#43911) · a7a4843c
  由 zmxdream 提交于 6月 29, 2022
  
  a7a4843c
- W
  
  convert to mixed model python api (#43881) · cbaebb04
  由 Wilber 提交于 6月 29, 2022
  
  cbaebb04
- Z
  Update the lock logic used in CinnCompiler::Compile. (#43876) · ccfde2da
  由 Zhen Wang 提交于 6月 29, 2022
```
* Update the lock logic used in CinnCompiler::Compile.
```
  ccfde2da
- C
  add equal trt converter (#43461) · 1dbbe20e
  由 ccrrong 提交于 6月 29, 2022
```
* add comparisons trt converter
```
  1dbbe20e
- L
  [new-exec] remove variable scope, stage 1 (#43865) · 9f74363f
  由 Leo Chen 提交于 6月 29, 2022
```
* separate variable scope and scope

* hot fix for lod_tensor_blocking_queue

* fix bug that variable exists in global scope
```
  9f74363f
- C
  
  fix device context init error (#43910) · d1ac85e5
  由 Chen Weihang 提交于 6月 29, 2022
  
  d1ac85e5
- W
  inference support mixed-precision model [1]. (#43814) · c7694b82
  由 Wilber 提交于 6月 29, 2022
```
* inference add convert to mixed model ability.
```
  c7694b82
- Z
  Move apis(cross, diagonal) legacy_api.yaml to api.yaml (#43893) · 8fa8e17e
  由 zyfncg 提交于 6月 29, 2022
```
* move cross form legacy_api.yaml to api.yaml

* move diagonal to api.yaml
```
  8fa8e17e
- R
  
  fix custom_device log (#43890) · fb1a93a8
  由 ronnywang 提交于 6月 29, 2022
  
  fb1a93a8
28 6月, 2022 6 次提交
- S
  
  make eager_utils.h in the begining of all headers (#43896) · 72116696
  由 Sing_chan 提交于 6月 28, 2022
  
  72116696
- T
  Enable Bert on bfloat16 datatype (#43455) · 6d31dc93
  由 Tomasz Socha 提交于 6月 28, 2022
```
* Remove output arguments from functions.
Replace pointers with references

* Name used bool flags

* Reorder functions

* Enable bfloat16 data type

* Give declarations some space

* Style

* Style
```
  6d31dc93
- Z
  
  [MLU]: add roi_align and roi_align_grad kernel (#43757) · 99ea0a9c
  由 zhaoying9105 提交于 6月 28, 2022
  
  99ea0a9c
- F
  
  [MLU]add mlu kernel for where_index op (#43720) · ec5f8cfd
  由 fuyou765 提交于 6月 28, 2022
  
  ec5f8cfd
- 石
  
  fixes a bug, test=develop (#43884) · 5369378b
  由石晓伟提交于 6月 28, 2022
  
  5369378b
- 【Sparse】add SparseTensor mv kernel(csr*dense_vec->dence_vec, coo*dense_vec->dense_vec) (#43668) · 5161a047
  由 zhouweiwei2014 提交于 6月 28, 2022
```
* [Sparse]add SparseTensor mv kernel(csr*dense_vec->dence_vec, coo*dense_vec->dense_vec)

* fix CI
```
  5161a047

机器未来 / Paddle 与 Fork 源项目一致

机器未来 / Paddle
与 Fork 源项目一致