* Fix "tar" to "pdparams" to adapt to dygraph

* Update the download link of Paddle Inference Library

* support fp16 training

* Use compiled training program

* Change timing ips.

* Use dali

* add pure fp16 training

* fix a bug, which will not use fuse pass using pure fp16 training.

* modify code as review

* modify loss, so that it will use different loss when using pure fp16 training.

* remove some fluid API

* add static optimizer.

* fix eval script

* fix dali shell

* fix pred

* fix resnest

* fix hrnet se

* fix se for export

fix resnest, ghostnet typo

* fix dpn

* fix se resnext101 metrics

* Update mobilenet_v3.py

* Update densenet.py

* Update resnest.py

* Update hrnet.py

* Update vgg.py

* Update xception.py

* add support for multi cards eva

* add static running in dygraph

add resnest101

* unify model and model_name

* rm unsed archs fix bug of se_resnext

* fix distillation model for distributed.launch

* fix export model for googlenet

* add densent init

* fix export model

fix log benchmark

Add InceptionV3 model

polish codes and docs to adapt Paddle2.0-rc

* Create regnet.py

* Update __init__.py