* fix AdamOptimizer, test=develop, test=document_fix

* fix AdamOptimizer, test=develop, test=document_fix

* fix AdamOptimizer, test=develop, test=document_fix

* fix AdamOptimizer, test=develop, test=document_fix

* fix python code

* fix norm api cn, test=develop

* fix norm api cn, test=develop

* fix norm api cn, test=develop

* fix norm api cn, test=develop

* fix norm api cn, test=develop

* fix norm api cn, test=develop

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

* remove the dependence of  BackwardStrategy

Co-authored-by: Nwangchaochaohu <wangchao66@baidu.com>

* add float16 dtype, test=develop

* add type for unique

* add unique doc

* add cn for set_default_dtype/get_default_dtype/scatter/matmul

* add distributed collective api_cn; test=develop

* add 1d pooling

fix the doc of dot

As the title

* api2.0, Add Bilinear doc(eng,cn), test=develop

* fix some description, test=develop

* fix some description, test=develop

* fix some description, test=develop

* fix some describtion, test=develop
Co-authored-by: NChen Weihang <chenweihang@baidu.com>

* add docs

 update the doc for the topK

* add unique doc

* Test SmoothL1Loss

* fix SmoothL1Loss doc

* fix some error

* fix to_tensor and add name=None,test=develop

* add bernoulli

* refine manual_seed

* add bernoulli

* refine bernoulli

* refine squeeze

* refine unsqueeze

* refine stack

* fix manual_seed

* fix manual_seed

* add cn doc of dropout series

* fix doc in quto and details

* fix doc of AlphaDropout

* fix cn doc details

* refine doc of functional dropout

* refine details in cn doc

* refine order

* fix sample code in Dropout2D/3D

* add cross_entropy_loss_cn doc to functional_cn and loss_cn

* fix bug in doc/fluid/api_cn/nn_cn/functional_cn/cross_entropy_loss_cn.rst

* fix format for functional_cn/cross_entropy_loss_cn.rst and loss_cn/CrossEntropyLoss_cn.rst

* Add chinese and english documents of paddle.prod

test=develop

* add cn doc of trace API

* add cn doc of functional ctc_loss and CTCLoss class.

* modified the format of doc.

* add en doc of ctc_loss and CTCLoss

* modified paddle.to_variable to paddle.to_tensor

 add doc of isnan, isinf, isfinite

* Add the argmax, argmin api doc for the api 2.0, test=develop