* support math operator for variable, test=develop

* polish code, test=develop

* polish code, test=develop

* expose dygraph.grad api, test=develop, test=document_fix

* add more parameter in dygraph.grad API, test=develop

* add only_inputs=True parameter, test=develop

* follow comments, test=develop, test=document_fix

* fix typo, test=develop, test=document_fix

* fix no cvm, when accessor is DownpourSparseValueAccessor, set no_cvm=True
* test=develop

* support feed_python_builtin, test=develop

* add test, test=develop

* support CompiledProgram, test=develop

* support fluid.data, test=develop

* fix ci problems, test=develop

* follow comments, test=develop

* add clear_one_table
* test=develop

* fix bug when grad_op's output not be used in compute kernel in dygraph mode, test=develop

* fix test_data_norm_op.py, test=develop

* polish code style, test=develop

* fix error of data_norm's infershape, test=develop

* polish comment, test=develop

1. Add Decorator 'dygraph_to_static_program'
2. Add corresponding ProgramTranslator.get_program
3. Add ProgramTranslator.save_inference_model
4. Modified some warning information of dy2stat
5. Change program cache to contain startup_program because for users who gets program to run, they may like to initialize startup program

* add unittest test=develop

* add function cache test=develop

* Add parameter(learning_rate) in NoamDecay test=develop

This PR does exact the thing as the title. The reason is that we plan to develop 4 decorators

"dygraph_to_static_code"
"dygraph_to_static_program"
"dygraph_to_static_func"
"dygraph_to_static_output"

 The 4 decorators will emphasize different part when translating dygraph to static graph. Decorator name "dygraph_to_static_graph" is too big for the function it implements.

As the title, we add decorator "dygraph_to_static_code", and add related "get_code", "get_func", "get_output" for ProgramTranslator. Next step will be adding "dygraph_to_static_program"

* Polish error Info in while_loop test=develop

test=develop

* Post_training_quantizaion supports min_max methon

1. Rename AutoTracer to ProgramTranslator
2. Rename cached_program to program_cache
3. Remove some functor style __call__
4. Dict key should be string but not hash code of string

* add dequantize_log_op and make pyramid hash support int8 weight test=develop

* add unittest and update pyramid hash op test=develop

* remove paddle_enforce test=develop

* fix error message test=develop

* remove incorrent commit test=develop

* fix error message in log_dequantize test=develop

* change 2019 to 2020 test=develop

* remove useless check_grad test=develop

* sequential reader stage 1, test=develop

* fix ut, test=develop

* fix iterable=False reset bug, add some logs and polish code, test=develop

* inference feed partial data, test=develop

* Turn on keep_order=True for test, test=develop

* enhance ut to test more cases, test=develop

* test commit for reverting

* Revert "test commit for reverting", test=develop

This reverts commit 80aef42e.

* add ut of merged and unmerged results, test=develop

* add more uts for coverages and add en doc of api, test=develop

* follow comments, test=develop

* change note style, test=develop

update embedding_eltwise_layernorm fuse pass and fused kernel, to support multi input

* add double grad implementation for dygraph, test=develop

* polish code, add uts, test=develop

* fix place bug, test=develop

* polish codes, add more uts for coverages, test=develop

* add no_grad_set, test=develop

* add star gan ut, test=develop

* follow comments, test=develop