* add paddle new save load doc

* move static save apis & polish multiple details

* add api mapping

* move apis place

* fix example code error

* polish model save load guide

* polish example code

* fix details example error

* polish guide doc

* change title format

* remove history reason

* remove other history reason

* add enable_static for old save/load

* add enable_static for old api

* add cn doc of trace API

* modified doc of warpctc  ctc_loss and CTCLoss, add support float64 of log_probs

test=develop

* add docs

* add Tensor concept introduction

* add Tensor concept introduction

* add Tensor concept introduction

* add Tensor concept introduction

* add Tensor concept introduction

* fix example code

* update conv kl-loss doc

* refine doc of stack

* update paddle api

* fix typo

* fix doc

* fix sample code

* fix code-block

* 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 batchnorm ,test=develop

* add the chinese doc of binary_cross_entropy_with_logits.

* 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 bug cn doc, test=develop

* fix norm bug cn doc, test=develop

fix the argmax, argmin doc, change the default value for the argmax/argmin

* for new optimizer

* update en doc

* update en doc

* update en doc

* update cn doc

* fix cn doc

* update doc

* fix code style

* fix firmula of adamw

* update doc

* update doc

* fix sample code

* fix sample code

* fix rmsprop sample code

* change learningratedecay to lrscheduler

* fix doc

* fix return

* fix sample code

* fix sample code

* fix sample code

* fix sample code

* add distributed dir and doc

* remove start_processes & move DataParallel

* add get rank and world size

* add cn docs

* add index to distributed_cn

* add white list api

* add api white list

* add cn to api name

* polish details

* add rst suffix

* add py function

* update doc fix for some ops, test=develop

* udpate the doc for asin, cos, atan, test=develop

add the doc for the logic ops, add the broadcasting ops

* add functional_one_hot doc

* fix some bug

* fix some bug

* fix some bug

* fix some asc bug

* fix code bug

* fix text

* add example code

* fix bug

* fix bug

* fix bug

* fix some bug

* add InputSpec zh doc

* fix sample code

* modify declarative into to_static

* sync with 2.0 api

Add dynamic/static learning rate doc for the api2.0

* 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