* support preload thread
* sleep before fleet wrapper exit for pslib core dump
* optimize hdfs log
* fix master+patch bug

* Open fuse all reduce op
test=develop

* Add Fuse optimization op log

* Add log in fuse_optimizer op pass and fuse all_reduce op pass

* replace with boost::optional<bool>
test=develop

* Polish code
test=develop

* fix code coverage
test=develop

* paddle::framework::vectorize() templatization

test=develop

* update pybind/imperative.cc

test=develop

* revert update on unsqueeze_op.cc and warpctc_cudnn_op.cu.cc

test=develop

* refactor dygraph,test=develop

* fix failed unittest,test=develop

* polish code,test=develop

* check windows ci error,test=develop
try to fix windows ci error by np.allclose,test=develop

* polish vlog and profiler, test=develop

* try to fix preceding ops order,test=develop

* test transformer in windows ci, test=develop

* use python c-api to speed up tracer.trace,test=develop

* test=develop, fix docker with paddle nccl problem

* test=develop, add ut for debug string and gradient_accumulator

* test=develop, add tests for layer/gradient_accumulator/prepared_op

* test=develop, fix complie error for test_prepared_op

* test=develop, add more ut for dygraph

* test=develop, create API.spec for dygraph api change

* test=develop, refoctor name to make it easier to understand

* test=develop, refoctor name to make it easier to understand

* test=develop, fix multi-gpu failed problem , add Tracer tests, change PADDLEENFORCE to PADDLEENFORCE_EQ

* test=develop, fix ut failed on parallel se-resnext

* test=develop, change one more PADDLE_ENFORCE

* Fix bug: add feed_vars to the prune function

* Support looking up embeddings from BoxPS.
* Add a _pull_box_sparse op, for now this op is not exposed to users.
* Add a BoxHelper class, providing 'BeginPass', 'EndPass', 'FeedPass' functions and so on.
* Add 'BoxPSDataset' in python code.
* Add a compile options WITH_BOX_PS and a MACRO PADDLE_WITH_BOX_PS.
* Add UT.
* More concrete information pls refer to: https://github.com/PaddlePaddle/Paddle/pull/18982

* dump slot

* test

* proto

* dump slot

* test

* proto

* code style

* code style

* code style

* style

* add delete after unseen days

* add unseen days

* code style

* conflict solve
test=develop

* add clear model

* code style
test=develop

* code style
test=develop

* support debug tensor of each ins
test=develop

* support debug tensor of each ins
test=develop

* learning rate

* code style

* code style

* code style

* code style

* code style

* code style

* code style

* code style

* code style

* code style

* code style

* code style

* code style
test=develop

* code style
test=develop

* unitest

* style

* style

* multi phase

* add channel

* code style

* style

* style

* unitest

* style

* define

* define
test=develop

* style
test=develop

* rm define
test=develop

* linux

* linux
test=develop

* style
test=develop

* output format
test=develop

* windows ci
test=develop

* fix bug of getting bool Flags from os.environ, test=develop

* add empty loss_name in CompiledProgram for inplace grad test, test=develop

* python inference api supports numpy and add unit test, fix unit test fail in test_slim_int8_googlenet and test_slim_int8_mobilenet

* add pybind interface to get all inplace ops, test=develop

* enhance OpTest to check whether the consistency of operator when using and not using inplace, test=develop

* handle corner cases in op_test, test=develop

* support outputs without tensor holder_, like XShape in reshape_op, test=develop

* fix bug, some op has GradOpMaker, but actually no grad_op in OpInfoMap, test=develop

* use reshape_grad instead of reshape in FlattenGradOp, test=develop

* fix error debug dims info for variables like XShape, test=develop

* change computational order in sum_op to relieve computation difference using inplace, test=develop

* add inplace_atol to check group_norm, and skip inplace_grad for mkldnn, test=develop

* follow sneaxiy's comments, test=develop

* remove unused DefaultGradOpDescMaker in mkldnn op, test=develop

test=develop

test=develop

add save cache model api in fleet& add slots shuffle in dataset module & add metric op to calculate ctr related metrics (#18871)

* add ctr related metric layer test=develop

* add save cache and slots shuffle test=develop

* add save cache and slots shuffle test=develop

* fix error

* fix error

* fix style for ci

* fix for comments

* change SlotsShuffle input to std::strinf for generality

* fix style

* fix style

* fix style

* fix style

* fix style

* fix style

* fix stylr

* fix style

* fix style

* fix style

* fix style

* fix style

* fix style

* fix style

* fix style

* fix style

* fix style

* fix style

* fix style

* fix style

* change non-const reference to pointer

* fix style

* fix style

* fix style test=develop

* fix style  test=develop

* add return ins num in ctr metric op

* change dtype to float in metric_op.py

* fix error test=develop

* fix style test=develop

* fix API spec

* fix API spec

* fix API spec test=develop

* add UT test=develop

* add fleet util (fleet/utils/fleet_util.py): functions for users' convenience
* add some interface in hdfs util : hdfs is_file、hdfs cat

* 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

* fix warpctc.dll not found issue, test=develop

* revert the linux platform change, test=develop

* delete warpctc_lib_path.h.in, test=develop

* add SetPySitePackagePath function

* fix warpctc.dylib not found issue on Mac, test=develop

* improve the paddle lib path setting logic, test=develop

* fix mac ci issue caused by test_warpctc_op unittest, test=develop

* tweak code, test=develop

python inference API support enable_memory_optim

* Fix Mask rcnn predictor
    1. refine memory optim algorithm to support the model with the block op.
    2. output diff : modify the affine channel fuse
    3. add condition_block_infer op
add interface for setting trt calib table dir
test=develop

* add the missing files.
test=develop

* 1 add trt fp16 support
test=develop

* update parallel.py
test=develop

* remove legacy memory optimization codes, test=develop

* follow huihuang's comments,test=develop

* follow luotao's comments, test=develop

* dump slot

* test

* proto

* dump slot

* test

* proto

* code style

* code style

* code style

* style

* add delete after unseen days

* add unseen days

* code style

* conflict solve
test=develop

* add clear model

* code style
test=develop

* code style
test=develop

test=develop

（1）support patch data （merge slots of instances of same line id, modify dense layer which
changes its size）
（2）add fleet load_one_table interface, support load from paddle model and load from pslib model
（3）fix push sparse bug which cause push sparse cost more time（about 10% in my testcase）
（4）when some slots are not in one of your network (join/update, etc.)，data feed、collect label info、push/pull sparse will skip these slots， instead of throw error.
（5）add more debug info in TrainFilesWithProfiler

* support sparse gradients
test=develop

* feature/auto_growth_allocator, test=develop

* add unittest of AlignedAllocator, test=develop

* try to turn on auto_growth to test on CI, test=develop

* fix segmentation fault in mixed_vector.h, test=develop

* add unittests, test=develop

* remove async executor and add data_feed.proto to the deps of train demo

* fix int64_t

* update fill constant op unittest

* add empty line

* 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

* Fix Mask rcnn predictor
    1. refine memory optim algorithm to support the model with the block op.
    2. output diff : modify the affine channel fuse
    3. add condition_block_infer op
add interface for setting trt calib table dir
test=develop

* add the missing files.
test=develop

1. Since allreduce op has 4 reduce types, We split these four reduce types into four ops
2. We also refined the collective op code, e.g. we separated the collective op kernel into CPUKernel and CUDAKernel, and remove the device specified DeviceContext parameter in template as we already knew the target DeviceContext
3. We remove the newly added Collective op role to reduce the complexity of program and graph analysis

Fix dygraph show style for FluidDoc.

* add is_runnning in communicator, test=develop

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

* supports collective training in executor

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

* fix comment
test=develop

* use unique name for nccl_id

* supports output to stream in program_to_code

* insert sync_comm_stream before regularization; add skip_op_callstack capability in program_to_code

* set op role in collective training

* add collective op role

* remove orig file

* add build optimizer by strategy

* add collective strategy

* refine collective strategy

* add multi-process role maker

* refine strategy building factory so that we can easily plugin more strategy

* scale loss grad in collective sgd transpiler

* add support for distributed fc

* code format

* revert some features for dist fc

* add support for distributed fc training

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

* supports collective training in executor

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

* use unique name for nccl_id

* supports output to stream in program_to_code

* insert sync_comm_stream before regularization; add skip_op_callstack capability in program_to_code

* set op role in collective training

* add collective op role

* fix comment
test=develop

* remove orig file

* add build optimizer by strategy

* add collective strategy

* refine collective strategy

* add multi-process role maker

* refine strategy building factory so that we can easily plugin more strategy

* scale loss grad in collective sgd transpiler

* add support for distributed fc

* code format

* revert some features for dist fc

* add support for distributed fc training

* test=develop
add collective op unittest standard

* test=develop
remove the test_collective directory

* test=develop
remove the test_collective directory

* remove slicegather test

* code format for reducescatter

* update attr of shard_index_op

* Modify macro nccl_helper

* remove test without distribute

* macro collective_helper

* marcro update

* test=develop
update support python3.5

* test=develop change gpu memory use to 0.1 when test

* test=develop
update ut equal func

* test=develop
set flags to 1.5

* test=develop fix pickle dumple  py35

* test=develop
fix divide in slice and add sync_comm_stream
update atol and rtol to 1e-05
rm shard_index op and test
modify read input from file to read from memory
remove origin_program in framework and add i/o in c_sync_calc_stream

* test=develop update unittest sync operator I/O

* refine cuda place error msg, test=develop

* use LOG(ERROR)+exit(-1), test=develop

(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