* [AMP] Add python API for collecting operator stats.

* Fix import and polish codes.

* Add more unittest.

* Add doc for the new APIs.

* add pure fp16 major function in auto_cast & tracer

* support master weight in dygraph for pure fp16

* check mix dtype of fp16&fp32 for check_finite_and_unscale op

* change pure fp16 funtion name

* refine some bug in auto_cast

* refine auto_cast interface logic

* add param _casted_by_pure_fp16 for class Layer

* support state_dict hook for save model by user appointed dtype in pure_fp16_decorator

* refine pure_fp16_decorator as decorator

* add unittest

* add comment

* add comment

* support recompute

* add comment for auto_cast and decorator

* support to_static_state_dict for paddle.jit.save

* unlimite models num and optimizers num

* add lookup_table in black_list

* fix momentum and layer state_dict

* fix bug in layer state_dict

* fix bug in layer state_dict_helper

* refine unittest

* refine test_momentun_op

* refine interface and some code

* refine amp_decorator interface

* refine pure fp16 interface

* refine master weight interface

* update 2.0 public api in all left files

* reverse device.py all list;
fix some flake8 errors

* move dygraph amp api to paddle

* refine code and add unit test

* Add RNN related apis in paddl.nn
test=develop

* new rnn api, cell almost done

* add new progresses in rnn APIs for 2.0

* refine rnn APIs and docstrings.

* add unittets

* disable gpu tests when paddle is not compiled with cuda support

* remove unnecessary imports

* fix docstring

* add to no_sample wlist

* backport to python2 to avoid yield from

* add **kwargs, fix typos

* update docstrings for birnn

* rename argument for SimpleRNN and SimpleRNNCell, fix sample code

* add default value for initial_states in fluid.layers.birnn
Co-authored-by: Nguosheng <guosheng@baidu.com>

* define fluid alias in paddle.tensor and paddle.nn dir test=develop

* from layer import loss in paddle.nn dir test=develop

* fix high level API(Inference) bug

* patch the unit tests

* Use for_test=True in the Fluid Trainer to clone the test program

* fix typo

* Should do the samething to the inferencer

* add Inference.infer

* optimize code

* update no_test_word2vec_new_api.py

* update trainer

* split check_and_get_place

* use inference_program to save inference model in Trainer

* update demo

* update save_inference_model

* clean code

* Load/save the params from the params_path

* Switch to use load_persistables and save_persistables

* Instaed of setup the executor to run program and scope. Pass the program to the load_persistables

- The trainer and inferencer will load params from disk if param_path
  argument is not None in their constructor.

- Remove params.py, we will expose core.Scope to the user if needed
  (e.g., for GAN). Currently we will not expose it, unless we clearly
  know doing so can support GAN.

- Add `save_params` to Trainer (a TODO item).

- rename "network" to "program"

* "fix decode bug"

* "follow commnet"

* "fix error"

* "fix hook bug"

* fix based comment

* fix copyright

* fix based on comment

* add copyright hook

* add copyright hook

* refine copyright hook

* "test copyright hook"

* fix check style

* fix ci

* Fix lots of trainer_config_helpers bug, and complete unittest for `layers.py`