提交 · 59bd1e6f810cc8fd7ca09270b2aadbfe2e081288 · 机器未来 / Paddle

09 12月, 2021 1 次提交
- Y
  
  [fleet_executor] set learning rate for each run and set op_role for fetch var (#37945) · 59bd1e6f
  由 Yuang Liu 提交于 12月 09, 2021
  
  59bd1e6f
08 12月, 2021 1 次提交
- Y
  
  [fleet_executor] bug fix for fleet_executor, test=allcase (#37934) · 55b87742
  由 Yuang Liu 提交于 12月 08, 2021
  
  55b87742
07 12月, 2021 1 次提交
- L
  
  [Fleet Executor] Add feed, fetch and check correctness (#37824) · b8793f70
  由 LiYuRio 提交于 12月 07, 2021
  
  b8793f70
02 12月, 2021 1 次提交
- L
  
  [Fleet Executor] Refine runtime graph (#37703) · 0074a3c9
  由 LiYuRio 提交于 12月 02, 2021
  
  0074a3c9
30 11月, 2021 1 次提交
- L
  
  [Fleet_Executor] Passing runtime scope and place (#37603) · 87e65a99
  由 LiYuRio 提交于 11月 30, 2021
  
  87e65a99
26 11月, 2021 1 次提交
- W
  
  use prune program in new executor, test=develop (#37591) · 4201c94a
  由 wanghuancoder 提交于 11月 26, 2021
  
  4201c94a
25 11月, 2021 3 次提交

[cherry-pick 2.2 heterps]bug fix for launch_utils.py (#37521) · 8bb1038c

由 zmx 提交于 11月 25, 2021

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* [heterps]bug fix for _run_from_dataset

* fix heter_server.cc

* fix launch_utils.py

* fix heter_section_worker.cc

* fix. test=develop

* fix. test=develop

8bb1038c

L
[new-exec] fix program cache key (#37500) · e64829e2
由 Leo Chen 提交于 11月 25, 2021
```
* fix program cache key

* bug fix

* fix cache problem

* remove unused code
```
e64829e2
L
[new-exec] skip compiled program (#37512) · 171da2ce
由 Leo Chen 提交于 11月 25, 2021
```
* skip compiled program

* fix ut
```
171da2ce

24 11月, 2021 1 次提交
- T
  [GpuPs]pybind core (#37287) · d69daed1
  由 Thunderbrook 提交于 11月 24, 2021
```
* pybind core

* set use psgpu
```
  d69daed1
23 11月, 2021 1 次提交
- L
  [new-exec] skip compiled program with places > 1 (#37457) · 2dfcdf21
  由 Leo Chen 提交于 11月 23, 2021
```
* skip compiled program with places > 1

* fix corner case and add ut
```
  2dfcdf21
22 11月, 2021 1 次提交
- L
  
  [new feature] add local scope for interpretercore (#37379) · 1f0512be
  由 Leo Chen 提交于 11月 22, 2021
  
  1f0512be
19 11月, 2021 1 次提交
- Y
  
  [fleet_executor] Parse pipeline config (#37319) · ca088f92
  由 Yuang Liu 提交于 11月 19, 2021
  
  ca088f92
18 11月, 2021 2 次提交

[heterps]change default executor for heter trainer (#37314) · c98d175d

由 zmx 提交于 11月 18, 2021

* fix pslib. test=develop

* add device to train_from_dataset. test=develop

* refine fleet.stop_worker. test=develop

* fix ut. test=develop

* fix ut. test=develop

* fix executor & ut. test=develop

* fix executor & ut. test=develop

* fix executor & ut. test=develop

c98d175d

Y

[fleet_executor] Parse runtime graph to start carrier (#37282) · f85bd5c9
由 Yuang Liu 提交于 11月 18, 2021

f85bd5c9

17 11月, 2021 3 次提交

L
[new-exec] Refine standalone executor (#37278) · 6d6642c8
由 Leo Chen 提交于 11月 17, 2021
```
* init

* add feed ops in python side

* import LRScheduler

* update_feed

* refine code format
```
6d6642c8

[heterps]Refactor heterogenous worker (#37244) · 54d2626a

由 zmx 提交于 11月 17, 2021

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix ut. test=develop

* fix ut. test=develop

* fix ut. test=develop

* refactor heter trainer. test=develop

* fix. test=develop

* fix ut. test=develop

* fix ut. test=develop

* fix ut. test=develop

* fix ut. test=develop

* fix ut. test=develop

* fix ut. test=develop

* fix ut. test=develop

* fix ut. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix ut. test=develop

* fix ut. test=develop

* fix ut. test=develop

54d2626a

L

[Fleet Executor] Construct runtime graph (#37158) · 0daa69d4
由 LiYuRio 提交于 11月 17, 2021

0daa69d4

15 11月, 2021 1 次提交
- W
  fix 3 bug of new_executor (#37142) · 8358d614
  由 wanghuancoder 提交于 11月 15, 2021
```
* fix 3 bug, test=develop

* refine, test=develop
```
  8358d614
11 11月, 2021 3 次提交

[Heterps]Refactor Heter Pipeline Parameter Server (#36845) · a2da1efa

由 zmx 提交于 11月 11, 2021

* change username

* fix

* fix

* fix

* fix

* fix

* update

* update

* update unittests

* fix

* update

* fix

* update

* fix

* fix

* fix

* update

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update send_and_recv op. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix ut. test=develop

* fix unit. notest,test=coverage

* fix ut. notest, test=coverage

* update. notest,test=coverage

* fix ut. notest, test=coverage

* fix ut. notest, test=coverage

* fix. notest, test=coverage

* fix. notest, test=coverage

* fix ut. notest, test=coverage

* fix ut. notest, test=coverage

* fix ut. notest, test=coverage

* fix ut. notest, test=coverage

* add func. notest, test=coverage

* fix ut. notest, test=coverage

* fix. test=develop

* fix. test=develop

a2da1efa

L

Get global cluster information (#37084) · 31673a92
由 LiYuRio 提交于 11月 11, 2021

31673a92
W
fix 2 bug: 1.skip lodtensorarray; 2.delete feed op (#37090) · d5df6bdf
由 wanghuancoder 提交于 11月 11, 2021
```
* fix 2 bug: 1.skip lodtensorarray; 2.delete feed op, test=develop

* program clone, test=develop
```
d5df6bdf

10 11月, 2021 1 次提交
- A
  
  Fix inner_program in Executor (#37083) · 8a2ce0f2
  由 Aurelius84 提交于 11月 10, 2021
  
  8a2ce0f2
09 11月, 2021 1 次提交
- A
  
  fix CompileProgram in Executor (#37036) · 77a8c94b
  由 Aurelius84 提交于 11月 09, 2021
  
  77a8c94b
03 11月, 2021 1 次提交
- L
  
  executor framework (#36892) · 10b039b7
  由 LiYuRio 提交于 11月 03, 2021
  
  10b039b7
29 10月, 2021 1 次提交
- W
  fix some bug in new executor (#36822) · b5af9575
  由 wanghuancoder 提交于 10月 29, 2021
```
* fix some bug in new executor, test=develop

* fix error message, test=develop
```
  b5af9575
20 10月, 2021 1 次提交

Add FasterTokenizer Operator (#34491) · 3f2d6a3f

由 Steffy-zxf 提交于 10月 20, 2021

Add Tokenizer related functionalities for Transformer model in order that the process of training and predicting is consistent.

* support the text string as an input Tensor
* support the "VOCAB"unordered_map<wstring, int> as an input Tensor to lookup tokens
* Tokenizer used for BERT. This tokenizer applies an end-to-end, text string to wordpiece tokenization.
* It first applies basic tokenization, followed by wordpiece tokenization.

3f2d6a3f

13 10月, 2021 1 次提交
- H
  Remove RunFromCinn in PE because We Will Call CinnRunner in Compute of SubgraphOp (#36385) · e051bba0
  由 Huihuang Zheng 提交于 10月 13, 2021
```
Remove RunFromCinn method in PE because We Will Call CinnRunner in Compute method of SubgraphOp
```
  e051bba0
11 10月, 2021 1 次提交

Add use_cinn Flag and RunFromCinn in PE (#36107) · 5690666c

由 Huihuang Zheng 提交于 10月 11, 2021

Add use_cinn flag and use it to control whether we run PaddlePaddle using CINN.

Also add:

Replace PaddlePaddle graph with a CINN graph in a pass
PE Method to feed data and run the graph by CINN

5690666c

08 10月, 2021 1 次提交

Support CUDA Graph on ParallelExecutor (#36250) · f9591bb1

由 Zeng Jinle 提交于 10月 08, 2021

* support CUDA Graph on PE

* add ut, fix CI compile

* reduce memory consumption

* fix CUDA 10 CI

* improve coverage

* improve python coverage

f9591bb1

22 9月, 2021 1 次提交
- W
  fix feed for new executor (#35803) · 4c2a06df
  由 wanghuancoder 提交于 9月 21, 2021
```
* fix feed, test=develop

* delete one test case, test=develop
```
  4c2a06df
15 9月, 2021 1 次提交
- A
  Enhance Check mechanism and Support single tuple/list of fetch_list in Executor (#35726) · cae050e8
  由 Aurelius84 提交于 9月 15, 2021
```
* Enhance Check mechanism of fetch_list in Executor

* support single tuple

* fix typo

* fix typo
```
  cae050e8
14 9月, 2021 1 次提交

Intergrate StandaloneExecutor in Static.Executor Interface with... · 4bc08530

由 Aurelius84 提交于 9月 14, 2021

Intergrate StandaloneExecutor in Static.Executor Interface with FLAGS_USE_STANDALONE_EXECUTOR (#35628)

* Intergrate StandaloneExecutor in Static.Executor Interface with FLAGS_USE_STANDALONE_EXECUTOR

* Enhance unittest and clean code in StandaloneExecutor

* polish unittest

4bc08530

28 7月, 2021 1 次提交

graph_to_program save parameter and stop_gradient information (#33771) · 8a7dee31

由 jiangcheng 提交于 7月 28, 2021

This PR added optional boolean is_parameter and stop_gradient in the VarDesc proto, and remove them during save_inference_model

8a7dee31

27 7月, 2021 1 次提交
- W
  
  [hybrid parallel] pipeline support adamw and LRScheduler (#34402) · 6ab0a6a8
  由 WangXi 提交于 7月 27, 2021
  
  6ab0a6a8
09 7月, 2021 1 次提交
- Y
  
  [hybrid performance] pipeline cache trainer (#33998) · 98c7191d
  由 Yuang Liu 提交于 7月 09, 2021
  
  98c7191d
06 7月, 2021 1 次提交
- W
  
  [hybrid performance] pipeline add program cache (#33954) · c9ae1362
  由 WangXi 提交于 7月 06, 2021
  
  c9ae1362
05 7月, 2021 1 次提交
- W
  
  [hybrid performance] optimize pipeline performance · 9914dff7
  由 WangXi 提交于 7月 05, 2021
  
  9914dff7
08 5月, 2021 1 次提交
- D
  【heterps】support cuda11 for heterps; add profiler in oneps (#32640) · beab9563
  由 danleifeng 提交于 5月 08, 2021
```
* add trainprofiler for heterps in oneps; test=develop

* add set_use_ps_gpu; test=develop
```
  beab9563
23 4月, 2021 1 次提交
- B
  solve hccl communicate conflict (#32447) · 0e74eea2
  由 Baibaifan 提交于 4月 23, 2021
```
solve hccl communicate conflict (#32447)
```
  0e74eea2

机器未来 / Paddle 与 Fork 源项目一致

机器未来 / Paddle
与 Fork 源项目一致