Merge pull request #1801 from TeslaZhao/v0.9.0

Merge pull request #1800 from TeslaZhao/develop

Merge pull request #1695 from TeslaZhao/v0.8.3

cherry-pick

Merge pull request #1671 from TeslaZhao/v0.8.2

Merge pull request #1670 from TeslaZhao/develop

fix dead link

fix dead link

Merge pull request #1526 from ShiningZhang/v0.7.0

v0.7.0 pick 1 PR to v0.7.0

Update benchmark.py

Merge pull request #1257 from bjjwwang/v0.6.0

V0.6.0 pick 4 pr

Merge pull request #1070 from wangjiawei04/v0.5.0

V0.5.0 pick 2 pr

- New Features

  - Support Java Client
  - Support TensorRT, add docker image for cuda10.1 and TensorRT 6
  - Modify the LocalPredictor interface to align with the RPC interface usage
  - Add Pipeline Serv·ing Dag Deployment
  - Support Windows 10 (Only Web Service and Local Predictor)
  - Add built-in Serving model converter
- Improve of Compatibility
  - Release cuda10.1 version of paddle_serving_server_gpu
  - Release Python 3.5 version of paddle_serving_client
  - Remove serving-client-app circular dependencies
  - Modify version of dependencies
- Improve of Framework
  - Pipeline Dag support multi-gpu
  - lower RPC thread restriction to 1
- Documents
  - Modify "COMPILE"
  - Add "WINDOWS_TUTORIAL"
  - Add "PIPELINE_SERVING"
  - Modify "BERT_10_MIN"
- New Demo
  - pipeline demos
  - Java Demo
- Bug fixes
  - Fix subprocess CUDA ERROR3 bug
  - Fix pip install dependencies
  - Fix import error in windows
  - Fix bugs in web service

Merge pull request #839 from MRXLT/release/0.3

[cherry pick to 0.3.2] debugger predict && bug fix