Update pip install paddle-serving-app

原命令 pip install paddle-serving-server-gpu==0.4.0.trt # GPU with CUDA10.1+TensorRT

新命令  pip install paddle-serving-server-gpu==0.4.0.100 # GPU with CUDA10.1+TensorRT

V0.4.0 gbk

fix gbk in win

cherry-pick PR 936

Fix Readme 0.4

V0.4.0 930 923 925

fix win doc and move requirements

[WIP] 0.4 index page readme

Docs fix

Merge pull request #926 from TeslaZhao/develop

Fix invalid variable 'path'

Merge pull request #921 from TeslaZhao/develop

Update documents of Pipeline Serving

This reverts commit cb37efc0.

V0.4.0 cherry-pick #918 #689 #917

rm ocr_detection dir

support encryption model

support windows doc

V0.4.0

Predicting on multi-devices in local predictor

Update test_client.py

Merge pull request #911 from wangjiawei04/example_bugs

fix bugs in examples

V0.4.0

Update INFERENCE_TO_SERVING_CN.md

add dockerfile for tenortrt

Bump junit from 4.11 to 4.13.1 in /java/examples

modify faster_rcnn_model benchmark

Grpcio restrict lower than 1.33.2

cherry pick #785

sync V0.4.0

cherry pick 844 900 902 904

Imagenet pipeline