* update the demo code and the doc of varbase.backward.

* remove the dependence of  BackwardStrategy

As the title

* add DataParallel cn doc

* polish details

* fix method declarative error

* polish format

* fix sample code error

* append to white list

* Add Chinese documents for api paddle.fluid.dygraph.enabled, test=develop, test=document_fix

* Add Chinese documents for api paddle.fluid.dygraph.enabled 2, test=develop, test=document_fix

* Add Chinese documents for api paddle.fluid.dygraph.enabled 3, test=develop, test=document_fix

* Add Chinese documents for api paddle.fluid.dygraph.enabled 4, test=develop, test=document_fix

* Update `no_grad` docs

add alias `paddle.no_grad`
use `paddle.no_grad()` for decorator usage
update code sample

* Fix alias

* Reword decorator usage note

* fix wrong in doc

* fix docs math style

* fix docs

* fix conv2d

* fix conv2d

* add English doc rst

* add chinese doc rst

add doc of new API:StepDecay, API:MultiStepDecay

fix doc of fluid.dygraph.to_variable,test=develop

* add register buffer zh doc

* refine zh doc

* add chinese doc of new api(ReduceLROnPlatear)

* add chinese doc of new api(ReduceLROnPlatear)

* add chinese doc of new api(ReduceLROnPlatear)

* fix doc of ReduceLROnPlatear,test=develop

* fix description of mode

* fix doc of ReduceLROnPlatear,test=develop

fix sample code error of fluid.dygraph.NaturalExpDecay

As the title.

fix the diff of sample code in chinese doc with english doc

* add new apis
quote> test=develop

* refine doc
test=develop

* use p
test=develop

* rm no_grad_guard
test=develop

* add register_forward_hook api for Layer, test=develop

* fix sample code, test=develop

* fix build strategy cn doc sample code diff

* fix code error

* fix execution strategy error

* fix LayerNorm error

* Add learning_rate in NoamDecay test=develop

* Update sample code test=develop

* remove FC, test=develop

* delete en_doc for fc, test=develop

* polish code style, test=develop

* change FC to Linear, test=develop

add clear_gradient、current_step_lr api and parameter_list parameter for optimizer in dygraph (#1694)

* add clear_gradient function and parameter_list parameter for optimizer, test=develop

* fix AdagradOptimizer_cn.rst, test=develop

* add sample code, test=develop

* polish sample code, test=develop

* polish description for clear_gradient, test=develop

* add current_step_lr api for optimizer, test=develop

* improve doc of module evaluation and CosineDecay API

* tweak doc of CosineDecay

add the doc of prepare_context

fix save_dygraph cn doc error and sample code error

test=develop

* add two named api
test=develop

* minor fix
test=develop

* update PRelu doc
test=develop

* refine doc
test=develop

* split arg
test=develop

* fix alpha format
test=develop

* del left name_scope docs
test=develop

* minor fix
test=develop