* make hapi support amp, and add unittest

* make unittest only support GPU

* update parameters for amp in hapi.Model

* update hapi.Model.prepare interface, and update unittest

* fix test_model.py unittest bug

* add grad clear in dygraph

* use_fp16_guard defaults to True, which could avoid nan

* add input check, and add internal doc link to low level api

* update doc, and decrease the sample num of dataset to avoid timeout

* make hapi amp param  support str 'O1' or 'O2'

* resume calling , modify the code of the check part

* upgrade the usage of Fleet API, and disable 'pure_fp16' param

* fix summary bug

test=document_fix

* remove PrettyTable dependence from paddle.flops

* fix bug in python2.7

* fix flops

* fix flops

* fix bug

* fix bug

* fix warning and no grad

* update condition of logger print

* fix flops

* fix flops

* update EarlyStopping doc

* update EarlyStopping doc, test=document_fix

* add ReduceLROnPlateau

* add reducer

* refine envent for memorycopy

* add concat&split for allreduce

* apply concat & split for fuse tensor

* fix nccl dep

* fix the untest, compile problem and ddp initialize problem

* fix untest for mac & add some comments & solve the repeated param in sublayers

* fix untest for windows & fix document

* solve pretty table dependent in flops api

* add unittest dependent

* temp

* enhance logger callback for benchmark

This reverts commit 63e90ee3.

* add hapi api flops

* fix bug

* fix some bug

* add unit test

* fix unit test

* solve ci coverage

* fix doc

* fix doc

* fix static flops

* delete the comment

* fix some grammar problem in doc

* fix some bug

* fix some doc

* fix some doc

* Fix doc format for callbacks, metrics and Model
* Fix code sample and doc

* add early stopping

* add doc for early stopping

* fix sample code bugs

* update infer of mode, update doc, add unittests to increase coverage rate

* fix sample code for early stopping

* update sample code and unittests

* reduce time cost of test_callbacks unittest

* fix model.py code style error

* update hapi save_inference_model output pathname

* update hapi save_inference_model output pathname

* use new 2.0-api paddle.static.io.load_inference_model

* add unittests to increase coverage rate

* add lr scheduler

* fix summary for rnn gru lstm

* return pil image as default according backend

* fix lr scheduler

* fix code style

* update hapi predict interface

* fix code style

* fix docs

* fix docs

* fix docs

* update docs

* fix codes style

* fix unittest

* fix unittest

* fix coverage

* rename manual_seed to seed

* rename xxx1d-->xxx1D, xxx2d-->xxx2D, xxx3d-->xxx3D

* rename manual_seed --> seed

* do not rename .cc, .cu and .h file

* rename manual_seed --> seed

* rename manual_seed --> seed

* rename manual_seed --> seed

* rename manual_seed --> seed

* disable_static on doc example code

* donot change manual_seed on generator

* add enable_static on sample code

* convert python/paddle/fluid/layers/nn.py to bak

* fix typo

* fix code style

* fix seed to manual_seed when call functions of Generator()

* fix bug

* hapi/model step learning rate on batch end. test=develop

* step lr_scheduler on epoch end in hapi/model.py fit. test=develop

* fix model bugs, inputs can be InputSpec instance

* correct error message

* fix dynamic parallel for hapi

* fix code style

* keep summary mode unchange

* add no grad decorator

* add visualdl callback

* remove input requirment in dygraph Model

* correct unittest

* upadte save inference model in dygraph without input

* fix unittets for test_model.py

* solve conflicts

* solve conflicts

* delete http.log

* fix test_model.py bug, correct initialization of MyModel

* fix unittests bugs

* set paddle manual seed for unittest

* fix Model bugs, because inputs can be list or dict when it is provided.

* add random seed for test_export_deploy_model

* delete redundant codes, because  calls

* Code optimization, error information optimization

* refine vision models

* reproduce summary api