* bug fix, test=develop

* Refine the cache of program, context and scope in executor.
test=develop

* Refine the unittest test_executor_and_use_program_cache.

* Add the test the PaddingRNN with use_program_cache=True.
test=develop

* Remove a check.
test=develop

* Refine the unittest to check whether it is correct when setting use_program_cache=True.
test=develop

* fix fetch handler error with pslib
* fix distributed lookup table op with 1 pserver

test=document_fix

* fix en doc train_from_dataset and infer_from_datatset

* fix fluid.data
test=document_fix

* fix doc batch dim
test=document_fix

* polish executor doc
test=develop

Refine error message for new data API

add executor.FetchHandler for train/infer from the dataset

* shape and optimize paddle error message stack, test=develop

* limit exception type & add unittest, test=develop

* fix multi-platform problem, test=develop

* fix related unnitest failed, test=develop

* add doc & fix unittest errors, test=develop

* fix function name error, test=develop

* update tensor test exception msg compare, test=develop

* remove unittest on win32, the dir format is different, test=develop

* remove useless package, test=develop

* add paddle enforce handler unittest, test=develop

* add exception checkout, test=develop

* fix coverage failed, test=develop

* fix op registry test failed, test=develop

* refactor whole pr, test=develop

* remove test in CMakelist, test=develop

* fix coverage, test=develop

The new "fluid.data" changes old "fluid.layers.data":

1. Add shape and dtype check.
2. Remove "append_batch_size" parameter. We won't offer this in the new data layer because other deep learning platforms don't have this kind of data layer pre-processing. It may confuse users.
3. Remove "stop gradient" parameter because the data layer doesn't do back-propagation

TODO：
Now data layer feeded by executor is checked, will we want to check the feed data of readers in the future?

* support change shuffle thread num
* support change train thread num
* fix receive shuffle data of each channel
* data norm stop gradient
* add check thread_tensor type and root_tensor type when merge metric
* remove sleep in shuffle, add config
* add config of pslib client to client communication
* fix xbox str
* add data norm op testcase
* add flush in trainer finalize

test=develop

* update executor feed

* use PE for compiler
test=develop

* fix memory overlapping of fetch var (return of executor.run), test=develop

* fix wrong usage of ParallelExecutor in op_test, test=develop

* remove useless parameter and simplify code

* avoid tensor destruct untimely, test=develop

* add testcase independent of OpTest, test=develop

* update exe.run

* remove legacy memory optimization codes, test=develop

* follow huihuang's comments,test=develop

* follow luotao's comments, test=develop

* feature/buffer_shared_inplace, test=develop

* refine code, test=develop

* fix elementwise_add op cpu inplace and sum inplace bug, test=develop

* add unittest and debug log, test=develop

* fix parallel_executor scope bug, polish code, test=develop

* fix sum op, activation op, single_in_place_inference bug, test=develop

* remove kLocalExecScopeName, test=develop

* fix unittest,test=develop

* fix out_var first version bug, test=develop

* follow comments,test=develop

* enhance execution error info
test=develop

move the thread-check code from train_from_dataset to a single function
add UT for the thread check function

(1) use channel instead of vector/BlockingQueue in Dataset，to keep same with existing implementation, and make code more readable and flexible (dataset single output channel or multi output channel). one previous memory out of limit problem is cause by not release memory after training.
(2) add Record because MultiSlotType costs too much memory (80B)，fix memory out of limit problem.
(3) add Channel, Archive in paddle/fluid/framework
(4) change dataset from shared_ptr to unique_ptr in pybind
(5) move create/destroy readers from trainer to dataset
(6) move shuffle from datafeed to dataset. dataset holds memory, datafeed is only for load data and feed data to network.
(7) fix thread num bug of Dataset when filelist size < thread num
(8) support set_queue_num in InMemoryDataset

* fix relative import error in python3
* fix debug string info

Add Pipeline Concurrency Train Mode:
- Cpp: pipeline_trainer & section_worker
- Python: PipelineOptimizer
- Add a new data_feed type: PrivateInstantDataFeed
- Add a test demo of pipeline trainer and the test model is gnn
- Do not support win32 now

* fix prepare context redundant code problem, optimize executor by caching create_varaiables
test=develop

* cache sub_scope, program, var when use_program_cache=True is set

* make fetch_list runable with variables, add more unittest for use_program_cache

add Run Prepared Ctx, fix pybind problem

* polish_executor_and_add_ctx_cache

* add save/load model, shrink table, cvm, config file & fix pull dense bug
test=develop

* fix global shuffle bug, fix pull dense bug, fix release memeory bug, fix shrink error
add client flush, add get data size
test=develop

* fix global shuffle bug
test=develop

* fix global shuffle bug
test=develop

* fix code style
test=develop

* fix code style & modify pslib cmake
test=develop

* fix error of _role_maker
test=develop

* fix code style
test=develop

* fix code style
test=develop

* fix code style
test=develop

* fix code style
test=develop

* fix code style
test=develop

* fix windows compile error of fleet
test=develop

* fix global shuffle bug

* add comment
test=develop

* update pslib.cmake
test=develop

* fix fill sparse bug
test=develop

* fix push sparse bug
test=develop

* Add conv2d_grad_grad_op
* Extracte the cuDNN conv algo searching code in conv_cudnn_helper.h.
    - Now use it in conv2d_grad_grad.
    - Will simply the searching code in conv2d and conv2d_grad in next PR.
* Enhance and fix bug in unit testing of gradient_checker.
* Support to fetch empty variables，return None in Python.

* fix train_from_dataset and infer_from_dataset example

* add inductive dim for data_reader, example: shape=[-1, 1], then -1 will be inducted through run-time reading of number of elements

Fix the following API examples:

paddle.fluid.scope_guard
paddle.fluid.backward.append_backward
paddle.fluid.cpu_places
paddle.fluid.cuda_pinned_places
paddle.fluid.cuda_places
paddle.fluid.in_dygraph_mode
paddle.fluid.CUDAPlace
paddle.fluid.CPUPlace
paddle.fluid.CUDAPinnedPlace

* polish doc
test=develop

* updata parallel executor doc
test=develop

* update API.spec
test=develop

* polish code
test=develop

* speedup gc and inplace softmax_with_cross_entropy_grad
test=develop

* refine models gpu mem
Merge skip vars and warning messages of mem opt
remove relu mem opt
test=develop

* follow comments
test=develop

test=develop

test=develop

test=develop

test=develop