2018 06 20

gongwb

Yan Chunwei

Inference
- 子图 for TensorRT 初步跑通，基于图的 analysis 优化框架基本成型
  - OPEN feature/analysis to support subgraph for TRT
    - https://github.com/PaddlePaddle/Paddle/pull/11538
  - OPEN feature/trt add softmax converter
    - https://github.com/PaddlePaddle/Paddle/pull/11526
  - MERGED bugfix/trt-op
    - https://github.com/PaddlePaddle/Paddle/pull/11487
  - MERGED feature/pass manager
    - https://github.com/PaddlePaddle/Paddle/pull/11440
- 高层 API output buffer 统一管理
  - MERGED inference/unify output buffer management
    - https://github.com/PaddlePaddle/Paddle/pull/11569
- 图像3个demo 完成一个，其他基本类似
  - OPEN WIP demo/mobilenet inference
    - https://github.com/PaddlePaddle/Paddle/pull/11510
- bugfix
  - MERGED bugfix/add_inference_lib_to_release
    - https://github.com/PaddlePaddle/Paddle/pull/11482
  - MERGED bugfix/anakin-ci
    - https://github.com/PaddlePaddle/Paddle/pull/11473
文档
- https://github.com/PaddlePaddle/Paddle/pull/11469
reviews

https://github.com/PaddlePaddle/continuous_evaluation/pull/70 https://github.com/PaddlePaddle/Paddle/pull/11466#pullrequestreview-128691509 https://github.com/PaddlePaddle/continuous_evaluation/pull/52#pullrequestreview-129019309 https://github.com/PaddlePaddle/continuous_evaluation/pull/69#pullrequestreview-129045735 https://github.com/PaddlePaddle/Paddle/pull/11525/files#r195896649 https://github.com/PaddlePaddle/Paddle/pull/11537#pullrequestreview-129570511

Bai Yifan

Train Pyramidbox model
Refine infer to use the pyramid method
- https://github.com/PaddlePaddle/models/pull/986
Evaluate Pyramidbox model

luotao

Training Benchmark of Fluid CPU multi-thread (with @zhaochengduo)，similar with V2: https://github.com/PaddlePaddle/Paddle/issues/11620
OCR: batch inference on OpenBlas version is similar to the online version.
doc:
- add doc of sequence_softmax and parallelDo: https://github.com/PaddlePaddle/Paddle/pull/11488
- add doc for inference_transpiler: https://github.com/PaddlePaddle/Paddle/pull/11521
- add url of cuda9.0_cudnn7_avx_mkl library：https://github.com/PaddlePaddle/Paddle/pull/11595
fix compiler error in contrib/tape: https://github.com/PaddlePaddle/Paddle/pull/11466
code review:
- bugfix/add_inference_lib_to_release: https://github.com/PaddlePaddle/Paddle/pull/11482
- Add a ctest hung tool: https://github.com/PaddlePaddle/Paddle/pull/11471
- MKLDNN:
  - [Merge] remove mkldnn flag from gtest strdup for cpu: https://github.com/PaddlePaddle/Paddle/pull/11395
  - [merge] MKLDNN layout: Support for activation operator: https://github.com/PaddlePaddle/Paddle/pull/11124
  - [merge] MKLDNN layout: Gaussian random layout: https://github.com/PaddlePaddle/Paddle/pull/11523
  - [merge] MKLDNN layout: Support for sum operator: https://github.com/PaddlePaddle/Paddle/pull/11102
  - [review] bnorm+relu fuse for mkldnn (inference) : https://github.com/PaddlePaddle/Paddle/pull/11434

wanghaoshuang

guosheng

NMT:
- Add weight sharing into Transformer and refine code to adapt to WMT14' en-de settings.
  - https://github.com/PaddlePaddle/models/pull/990
- Add README for Transformer (WIP).
  - https://github.com/PaddlePaddle/models/pull/994
- Experiments on WMT14 en-fr dataset
Review:
- hsigmoid_op
  - https://github.com/PaddlePaddle/Paddle/pull/11063

Dang Qingqing

Face Detection:
- Fix concat_op CUDA kernel bug: https://github.com/PaddlePaddle/Paddle/pull/11541
- Expose crop op into Python API: https://github.com/PaddlePaddle/Paddle/pull/11546
- Pyramidbox inference: https://github.com/PaddlePaddle/models/pull/992
- Train and evalute model
- [WIP] speed training:
  - 1. use memory optimization to increase batch size: https://github.com/PaddlePaddle/Paddle/issues/11605
  - 1. more profiling
Batch Norm fix:
- https://github.com/PaddlePaddle/Paddle/pull/11578
Doc:
- layers: https://github.com/PaddlePaddle/Paddle/pull/11495
- non-layers: https://github.com/PaddlePaddle/Paddle/pull/11559
Code revew:
- fake_quantize_op: https://github.com/PaddlePaddle/Paddle/pull/11359#pullrequestreview-130230036
- ICNet: https://github.com/PaddlePaddle/models/pull/975#pullrequestreview-128714172
- SSD Doc: https://github.com/PaddlePaddle/models/pull/987
- pyramidbox eval: https://github.com/PaddlePaddle/models/pull/986

Yan Xu

performance
- [merged] overlap memcpy in multiple rpc ops, improve 15% performance on vgg16 model,https://github.com/PaddlePaddle/Paddle/pull/11221
- [WIP] independent cuda stream in rpc op
stability and code polish
- ocr distributed model, https://github.com/Yancey1989/paddle_fluid_benchmark/tree/master/ocr_recognition
bug fix and code polish
- dist train on ce valid diff value, https://github.com/PaddlePaddle/paddle-ce-latest-kpis/pull/48
- new distribute op details folder opertors/distributed, https://github.com/PaddlePaddle/Paddle/pull/11603
- hidden bcast_params while dist train,https://github.com/PaddlePaddle/Paddle/pull/11575
- fix pserver sub-blocks bug, https://github.com/PaddlePaddle/Paddle/pull/11585

Yu Yang

Polish Documentation
Add script to check API changes
- https://github.com/PaddlePaddle/Paddle/pull/11573

Zeng Jinle

Design Python data feeding method
- https://github.com/PaddlePaddle/Paddle/pull/11592
[WIP]Implement Python data feeding in both C++ and Python side, and self-testing is about to be carried out.
Find a bug in tensor_py.h, which may cause "multiple definition" error when compiling if it is included in two or more compilation unit. It is caused by implementing fully specialized template functions in header file tensor_py.h. Adding inline before these functions would fix it. Ready to create a PR.

Qingsheng Li

Seq2seq Attention Model:
- Fixing sequence_expand_op https://github.com/PaddlePaddle/Paddle/pull/11618
Help Chan with book chapter

qiaolongfei

paddle fluid framework
- add keep_kids flag for executor https://github.com/PaddlePaddle/Paddle/pull/11524
paddle document
- update split_lod_tensor, create_array and array_length doc
  - https://github.com/PaddlePaddle/Paddle/pull/11383
- Add none layers api doc https://github.com/PaddlePaddle/Paddle/pull/11532
distributed lookup table
- Feed Ctr模型配置完成，基于线上数据的data reader完成，训练跑通。
- 用小数据跑训练任务，能正常收敛，还需要在大数据上验证。
- concat_op support data layer as input https://github.com/PaddlePaddle/Paddle/pull/11527 https://github.com/PaddlePaddle/Paddle/pull/11579
- support Ftrl optimizer https://github.com/PaddlePaddle/Paddle/pull/11547
- optimize auc calculate https://github.com/PaddlePaddle/Paddle/pull/11574

minqiyang

MERGE: Fix hang up bug when the 2nd GPU's memory is full
- issue
  - https://github.com/PaddlePaddle/Paddle/issues/11375
- PR
  - https://github.com/PaddlePaddle/Paddle/pull/11377
MERGE: Fix the bug that sub blocks in lr decay op
- issue
  - https://github.com/PaddlePaddle/Paddle/issues/11429
- PR
  - https://github.com/PaddlePaddle/Paddle/pull/11520
OPEN: unit test about gracefully shutdown in paddlepaddle/paddle:latest-dev takes no effect
- issue
  - https://github.com/PaddlePaddle/Paddle/issues/11509
OPEN: Update Doc: Parameter Split Strategy in Distribute Transpiler
- issue
  - https://github.com/PaddlePaddle/Paddle/issues/11250
- PR
  - https://github.com/PaddlePaddle/Paddle/pull/11283
OPEN: Update Design Doc: Distributed Block Design
- issue
  - https://github.com/PaddlePaddle/Paddle/issues/11609

Xin Pan

detection_map文档修复
- https://github.com/PaddlePaddle/Paddle/pull/11563
其他API文档注释
- https://github.com/PaddlePaddle/Paddle/pull/11511
Fix thread-safety
- https://github.com/PaddlePaddle/Paddle/pull/11608
Compatibility
- https://github.com/PaddlePaddle/Paddle/issues/11528
debug transformer 多机
- https://github.com/panyx0718/transformer-nist/commit/936c80d853fbf635d5cd302921f73385fb53e26b

zhaochengduo

Parallel Executor
- Enhance Parallel Executor stable
  - https://github.com/PaddlePaddle/Paddle/pull/11616
Doc
- DataReader, fluid.nets, Parallel Executor, ParameAttr
Small fix
- Fix concat compile warning
  - https://github.com/PaddlePaddle/Paddle/pull/11607
- Enable more type for SplitOp and ConcatOp
  - https://github.com/PaddlePaddle/Paddle/pull/11476
- Add FAQ(libmkldnn.so is not found when importing paddle.fluid)
  - https://github.com/PaddlePaddle/Paddle/pull/11607
Review
- small thread-safety fix and doc improvemen
  - https://github.com/PaddlePaddle/Paddle/pull/11608
- Make the CUDA kernel of concat correct and fix unit tests
  - https://github.com/PaddlePaddle/Paddle/pull/11541
- concat support data as input
  - https://github.com/PaddlePaddle/Paddle/pull/11527

guochaorong

CE duty problem following
- https://github.com/PaddlePaddle/paddle-ce-latest-kpis/issues/54
- https://github.com/PaddlePaddle/Paddle/issues/11615
CE platform migration
- https://github.com/PaddlePaddle/continuous_evaluation/issues/76
- https://github.com/PaddlePaddle/continuous_evaluation/wiki/Deploy-CE-platform-on-qianmo
python api monitor script
- pr: https://github.com/PaddlePaddle/Paddle/pull/11562
- review: https://github.com/PaddlePaddle/Paddle/pull/11573/files
paddle machine management
- http://wiki.baidu.com/pages/viewpage.action?pageId=502089396

Yibing Liu

Fix some problems in 10+ api reference
- https://github.com/PaddlePaddle/Paddle/pull/11491
Add python api for argsort operator
- https://github.com/PaddlePaddle/Paddle/pull/11174
Transformer: Setup model training of en-fr translation model
DeepASR:
- Update open sourced training data
  - https://github.com/PaddlePaddle/models/pull/989
- Support multi-thread parallel decoding
  - https://github.com/PaddlePaddle/models/pull/967

Code Review:

https://github.com/PaddlePaddle/models/pull/990

qiuxuezhong

NMT:
- fix convergence bug for distributed transformer:need broadcast parameters among GPUs and learning rate operater bug
- performance problem with distributed transformer, multi-machine is lower than single machine.
- issue： https://github.com/PaddlePaddle/Paddle/issues/11568
CTR:
- discussion on distributed training of ctr model

jetfuel

Communicating with ShanYi and Wang Guan to prepare PaddlePaddle.org new release. Targeting release next Friday.

PRs:
- Rearrange the feature layout: https://github.com/PaddlePaddle/PaddlePaddle.org/pull/498
- Add a new "Community" links on the top navigation bar : https://github.com/PaddlePaddle/PaddlePaddle.org/pull/499
Reviews:
- https://github.com/PaddlePaddle/book/pull/548#pullrequestreview-129664492
Project and Issues Management - PaddlePaddle.org
- https://github.com/PaddlePaddle/PaddlePaddle.org/projects/5
- Update the homepage background section: https://github.com/PaddlePaddle/PaddlePaddle.org/issues/495: 4 Nicky
- Update the homepage with a "Feature” section: https://github.com/PaddlePaddle/PaddlePaddle.org/issues/496: 4 Jeff
- Update the homepage with a "Model" section: https://github.com/PaddlePaddle/PaddlePaddle.org/issues/491: 8 Daming
- Update the homepage with a "Resource" section: https://github.com/PaddlePaddle/PaddlePaddle.org/issues/492: 8 Jeff
- Update the homepage with a "First Deep Learning Experience": https://github.com/PaddlePaddle/PaddlePaddle.org/issues/490: 16 Daming
- Update the homepage with an new installation guide: https://github.com/PaddlePaddle/PaddlePaddle.org/issues/489: 4 Jeff
- Add "About" on to the top navigation: https://github.com/PaddlePaddle/PaddlePaddle.org/issues/486: 8 Nicky
- Add "Community" on to the top navigation: https://github.com/PaddlePaddle/PaddlePaddle.org/issues/487: 2 Jeff
- Add "Blog" to the top navigation: https://github.com/PaddlePaddle/PaddlePaddle.org/issues/488: 24
- Consolidate Documentations: https://github.com/PaddlePaddle/PaddlePaddle.org/issues/493: 24 Varun
- Version select: https://github.com/PaddlePaddle/PaddlePaddle.org/issues/494: Varun

kexinzhao

Add lazy execution feature in tape: https://github.com/PaddlePaddle/tape/pull/18
Add data feeder in tape: https://github.com/PaddlePaddle/tape/pull/22
Review:

varunarora

New version of deployment of docs on PRs on other repos: https://github.com/PaddlePaddle/PaddlePaddle.org/pull/481
Working on designing / feedback on new version of PaddlePaddle.org

tonyyang-svail

Dynamic Graph: https://github.com/PaddlePaddle/tape
- Overview: https://github.com/PaddlePaddle/tape/issues/10

nicky

Update chapter 6 https://github.com/PaddlePaddle/book/pull/539
Rewrite book chapter 8 machine translation https://github.com/PaddlePaddle/book/pull/552
Review:

daming-lu

Finished 3 chapters for the new book version (Fluid)
Started working on PaddlePaddle.org new version
- https://github.com/PaddlePaddle/PaddlePaddle.org/pull/500
Review:
- https://github.com/PaddlePaddle/book/pull/546
- https://github.com/PaddlePaddle/PaddlePaddle.org/pull/498

helinwang

Learn imitation learning
- read paper
- setup race game training environment for imitation learning
- think and discuss how to apply it on our boat
- need a switch between self drive and guided drive.
- if current/wind is big, need to make that part of the state.
- need accurate sensor, perhaps GPS is a good one.

Lei Wang

Reading papers Yi referred for imitation learning
Reading book chapters to get basic knowledge about machine learning

kexinzhao

Tape
- Add dropout, batchnorm, and pool2d op: https://github.com/PaddlePaddle/tape/pull/34
- Add adam optimizer: https://github.com/PaddlePaddle/tape/pull/36
- Add accuracy function and test on mnist: https://github.com/PaddlePaddle/tape/pull/40
- Update paddle submodule to a fork: https://github.com/PaddlePaddle/tape/pull/25
- Review:

Release Notes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

2018 06 20

gongwb

Yan Chunwei

Bai Yifan

tangwei

fengjiayi

wuyi

tensor-tang

dongzhihong

luotao

wanghaoshuang

guosheng

Dang Qingqing

Yan Xu

Yu Yang

Zeng Jinle

Qingsheng Li

qiaolongfei

minqiyang

Xin Pan

zhaochengduo

guochaorong

Yibing Liu

qiuxuezhong

jetfuel

kexinzhao

varunarora

tonyyang-svail

nicky

daming-lu

helinwang

Lei Wang

kexinzhao

Clone this wiki locally