提交 · a0363d180500be25d9c3a7412a6bd7e18883d661 · PaddlePaddle / Paddle

13 6月, 2022 18 次提交
- C
  
  [MLU] add UTs for mlu interp_v2(bilinear). (#43386) · a0363d18
  由 Chenxiao Niu 提交于 6月 13, 2022
  
  a0363d18
- Q
  
  [MLU]add lookup_table_v2 op and fix amp feature of bert with mlu device (#43366) · 67bd5d9c
  由 qipengh 提交于 6月 13, 2022
  
  67bd5d9c
- W
  
  fused_attention fused_feedforward api support Model Tensor Parallel (#42985) · 31ddaae2
  由 WangXi 提交于 6月 13, 2022
  
  31ddaae2
- Z
  
  add only split (#43424) · 30b10630
  由 zhoutianzi666 提交于 6月 13, 2022
  
  30b10630
- 津
  
  [inference]add topk/topk_v2 trt convertor (#43368) · 65e86580
  由津提交于 6月 13, 2022
  
  65e86580
- G
  
  Add progress bar and speed up Quantization Pass (#43398) · 1c7e35dc
  由 Guanghua Yu 提交于 6月 13, 2022
  
  1c7e35dc
- Z
  [AutoParallel] fix fetch list (#43412) · 562b184c
  由 zhaoyingli 提交于 6月 13, 2022
```
* fix fetch list

* fix unittest
```
  562b184c
- L
  Fix some en docs of paddle. and paddle.nn.initialize. (#42916) · 4f006636
  由 Liyulingyue 提交于 6月 13, 2022
```
* calculate_gain; test=document_fix

* Constant; test=document_fix

* KaimingNormal; test=document_fix

* KaimingUniform; test=document_fix

* randint; test=document_fix

* squeeze;test=document_fix

* argmin; test=document_fix

* argmin; test=document_fix

* triu; test=document_fix

* add_n;test=document_fix

* unique; test=document_fix

* topk; test=document_fix

* squeeze;test=document_fix

* randint;test=document_fix

* argmin; test=document_fix

* constant; test=document_fix

* constant; test=document_fix

* KaimingNormal; test=document_fix

* kaiming; test=document_fix

* unique; test=document_fix
```
  4f006636
- W
  Add default config and update dataset in gpups (#43327) · 24ea1dd8
  由 wangguanqun 提交于 6月 13, 2022
```
* gpups default config and dataset

* codestyle

* add unittest

* code style
```
  24ea1dd8
- Y
  【Hachathon No.30】 (#40545) · 5d48528f
  由 yangguohao 提交于 6月 13, 2022
```
* 'TripletMarginDistanceLoss'

* 'test_file'

* '2022_03_27'

* 2022-03-31

* 2022-04-05

* 2

* 2022-04-17

* 2022-04-17_2

* 2022-04-17_3

* 2022-04-17_4

* 2022-04-25

* 2022-05-02_V1

* 2022-05-06_V1

* 2022-05-07_V1

* Update loss.py

* Update loss.py

* Update loss.py

* Update loss.py

* Update loss.py

* Update loss.py

* Update loss.py

* Update loss.py

* Update loss.py

* Update loss.py

* 2022-06-01_pre-commit

* 2022-06-05

* 2022-06-06

* 2022-06-07

* 2022-06-07_V2
```
  5d48528f
- Y
  
  change saved file dir for save_load unittest (#43416) · 0d0258f8
  由 YuanRisheng 提交于 6月 13, 2022
  
  0d0258f8
- W
  [Dy2St]Refine covnert_var_shape in dy2st (#43348) · 88504892
  由 WangZhen 提交于 6月 13, 2022
```
* Refine covnert_var_shape in dy2st

* Fix UT error

* Return paddle.shape only when has_negative and fix test_tensor_shape error

* Fix test_jit_save_load error
```
  88504892
- J
  
  support dcu in eager (#43371) · ee605673
  由 Jiabin Yang 提交于 6月 13, 2022
  
  ee605673
- J
  [Eager] Support set_grad_ivar for eager (#43378) · a0d0bb63
  由 Jiabin Yang 提交于 6月 13, 2022
```
* support set_grad_ivar for eager

* support set_grad_ivar for eager

* support set_grad_ivar for eager
```
  a0d0bb63
- R
  
  Fix cmakelint errors for some files (#43428) · edf69ae0
  由 Ruibiao Chen 提交于 6月 13, 2022
  
  edf69ae0
- Z
  
  gradient add support SparseCooTensor (#43352) · 97af8516
  由 zhangkaihuo 提交于 6月 13, 2022
  
  97af8516
- Z
  sparse convertion kernel support secondary dispatch (#43345) · 5752643b
  由 zhangkaihuo 提交于 6月 13, 2022
```
* use GpuMemcpy and GpuMemset

* sparse convert kernel support double dispatch by indices dtype

* cudaMemcpyKind->gpuMemcpyKind
```
  5752643b
- A
  [Dy2Stat]Remove fluid.layers in transformer code and return_out_vars in convert_ifelse (#43372) · c4c30e6f
  由 Aurelius84 提交于 6月 13, 2022
```
* [Dy2Stat]Remove fluid.layers in transformer code

* remove useless return_name_vars

* fix unittest

* fix unittest

* fix unittest
```
  c4c30e6f
11 6月, 2022 1 次提交
- C
  
  fix add_n incompatible error (#43395) · 3800f192
  由 Chen Weihang 提交于 6月 10, 2022
  
  3800f192
10 6月, 2022 13 次提交
- C
  [Phi] Fix depthwise conv yaml error (#43379) · f551d9fe
  由 Chen Weihang 提交于 6月 10, 2022
```
* fix depthwise conv yaml error

* fix depthwise conv double grad error
```
  f551d9fe
- L
  
  optimize bwd layer_norm kernel with fast method (#42491) · b4a93884
  由 limingshu 提交于 6月 10, 2022
  
  b4a93884
- [MLU] add mlu kernel for clip (#43229) · 798e2e7e
  由光明和真理提交于 6月 10, 2022
  
  798e2e7e
- F
  
  [MLU]add mlu kernel for scatter op (#43292) · 9ad05afd
  由 fuyou765 提交于 6月 10, 2022
  
  9ad05afd
- Y
  [BugFix]Fix dims mismatch when run rec_svtrnet model in eager mode (#43373) · cdeb3167
  由 YuanRisheng 提交于 6月 10, 2022
```
* change tensor name

* fix unittest bugs
```
  cdeb3167
- T
  
  [Hackathon No.28] implement logcumsumexp (#42267) · 19a7524f
  由 tiancaishaonvjituizi 提交于 6月 10, 2022
  
  19a7524f
- C
  
  [MLU]add mlu kernel for sqrt op (#43326) · 6d3a68cb
  由 cambriconhsq 提交于 6月 10, 2022
  
  6d3a68cb
- A
  add unary ops (#773) (#43363) · 8045fcfd
  由 Allen Guo 提交于 6月 10, 2022
```
* add unary ops

* move to activation_ops
```
  8045fcfd
- W
  Add option for executor profiler (#43355) · 6aee6410
  由 Weilong Wu 提交于 6月 10, 2022
```
* Add option for test executor profiler

* Change option for test executor_profiler
```
  6aee6410
- E
  Re-implemented check_finite_and_unscale_op with newly added xdnn api (#42960) · 6197fbf6
  由 enzodechine 提交于 6月 10, 2022
```
* Re-implemented check_finite_and_unscale_op  with newly added xdnn api
*test=kunlun

* Re-implemented check_finite_and_unscale_op  with newly added xdnn api

*test=kunlun
```
  6197fbf6
- F
  
  [MLU] add randperm kernel and reduce_prod kernel (#43357) · b07f469b
  由 fwenguang 提交于 6月 10, 2022
  
  b07f469b
- K
  fix structure infos conflict in static return_list mode (#43291) · 41bb70fb
  由 Kaipeng Deng 提交于 6月 10, 2022
```
* fix structure infos conflict in static return_list mode. test=develop
```
  41bb70fb
- Q
  
  add cleanup code, test=develop (#43305) · c43b54ad
  由 Qi Li 提交于 6月 10, 2022
  
  c43b54ad
09 6月, 2022 8 次提交
- F
  
  [MLU] add mlu meshgrid kernel (#43271) · 0d719718
  由 fwenguang 提交于 6月 09, 2022
  
  0d719718
- F
  
  [MLU]add mlu kernel for range op (#43296) · 1a80b484
  由 fuyou765 提交于 6月 09, 2022
  
  1a80b484
- C
  
  add mlu gather_nd kernel (#43344) · 0454b777
  由 cifar10 提交于 6月 09, 2022
  
  0454b777
- S
  Add nproc_per_node for DistributedFusedLamb (#43295) · 6678def9
  由 sneaxiy 提交于 6月 09, 2022
```
* add nproc_per_node for DistributedFusedLamb

* fix nproc_per_node communicator bug

* fix ring_id = 1 init bug

* fix ci

* fix test_parallel_executor_mnist.py
```
  6678def9
- Z
  
  use tempfile to place temporary files (#43316) · 2c8739e8
  由 zhaoyingli 提交于 6月 09, 2022
  
  2c8739e8
- A
  [Dy2Stat]Add apply_optimization for @to_static and remove... · 00ce09e6
  由 Aurelius84 提交于 6月 09, 2022
```
[Dy2Stat]Add apply_optimization for @to_static and remove BreakTransformOptimizer by default (#43320)

* [Dy2Stat]Add apply_optimization for @to_static and remove BreakTransformOptimizer by default

* fix unittest
```
  00ce09e6
- C
  
  [MLU]add mlu kernel for conv2dtransposed op (#43233) · c96f7a29
  由 cambriconhsq 提交于 6月 09, 2022
  
  c96f7a29
- C
  Implement dropout_nd operator to optimize dropout with axis not None. (#42463) · caa57498
  由 crystal 提交于 6月 09, 2022
```
Co-authored-by: NLiu Yiqun <liuyiqun01@baidu.com>
```
  caa57498

PaddlePaddle / Paddle 大约 1 年 前同步成功

PaddlePaddle / Paddle
大约 1 年前同步成功