* enabel test_parallel_dygraph_se_resnext
test=develop

* add signal message to stderr, test=develop

* add unittests for ugly SignalHandle, test=develop

* supports multiple NCCL communicators preserved in NCCLCommContext
test=develop

* add ut for c_comm_init_all operator and fix cuda resource release problem
test=develop

* save the callstack information to file when exception throws test=develop

* improve sequence_conv api doc test=develop

* add warning for padding param test=develop

modify into deprecated

fix fleet_desc dense_table unsort bug ，not  support format for abacus hotstart yet.

* support tensor input with padding for warpctc op

* merge with develop

* test=develop

* modified python API examples test=develop

* nn.py is modified for code coverage test=develop

* update documents info about warpctc op in API.spec test=develop

* add test_warpctc_with_padding in test_layers test=develop

* add warning log for cuda_version back to warpctc_op.cc

* modify API.spec for warpctc op test=develop

* modify API.spec

* update warpctc test to new CompiledProgram API test=develop

* modify code examples for warpctc op test=develop

* modify API.spec for warpctc op test=develop

* modify API.spec for warpctc op test=develop

* fix optimizer bug
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

add cuda10 support in fast_install.sh and add dynamic get version for release, then remove useless ave check for MacOS install check

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

test=develop

This PR fix problem with issue #18096 , which raise an error for user to specify the error about load dir is wrong

test=develop

* Add lookahead optimizer

* add unittest for lookahead optimizer

test=develop

* add doc string for LookaheadOptimizer

test=develop
test=document_preview

* add API spec for lookahead

test=develop
test=document_preview

* modify api spec

test=develop
test=document_preview

* modified doc string

* modify the test file

test=develop
test=document_preview

* modify doc string

test=develop
test=document_preview

* test=develop,Optimized error reporting information

* test=develop,add importscipy unittest

* test=develop, rename the file and function

* add default getItem

* test=develop, fix has_attr disabled error in Layer

* test=develop, fix GroupNorm and deepcf bug on attrs

* fix sync mode hang in transpiler
* remove sync mode in send/recv
* replace PADDLE_ENFORCE with PADDLE_ENFORCE_NE

* Fix arg do_model_average in param_attr

test=develop

* Update api spec

test=develop

test=develop

 add fleet_desc config feature & multi_sparse table,

* update light_nas_strategy: add latency constraint

test=develop

* update light_nas_strategy: update get_model_latency

test=develop

* update light_nas_strategy: add more check

test=develop

* update light_nas test

test=develop

* update light_nas test

    test=develop

* minor update light_nas test

    test=develop

* minor update light_nas test

test=develop

* update light_nas test

test=develop

* update _constrain_func of light_nas_strategy

test=develop

* update _constrain_func of light_nas_strategy

test=develop

* remove unused code

test=develop

* add pool2d quantization support, only for max-pooling.

* add the partial quantization support.

* Fix data parallel & sequence mask in dygraph

test=develop

* Revert change in data_parallel

test=develop

* use PE for compiler
test=develop

* add local user data conversion into full_pascalvoc_test_preprocess.py
test=develop

* change PADDLE_ENFORCE to PADDLE_ENFORCE_GE
test=develop

* change according to reviews
test=develop

* Implement the operator with sprase matrix multiply

* Update the URL of mklml library.

test=develop

* Disable MKLML implematation when using no-linux.

test=develop

* optimize bp with mkl sparse matrix
test=develop

* 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

* add diag init in Uniform_random op test=develop

* modify api.spec test=develop

* fix unform_batch_size_like maker test=develop

* add diag_num and diag_step assert check test=develop

* update parallel.py
test=develop

* stronger the error message of tensor's mutable_data

test=develop

* update error message

test=develop

* increase test_parallel_executor_seresnext time limit
test=develop

* split test_parallel_executor_seresnext
test=develop

* temporally disable reduce_and_allreduce test because of the random failure.
test=develop

* split gpu and cpu
test=develop

test=develop