* Add the dynamic load of nvrtc, and support runtime compiling of CUDA kernel using nvrtc.
test=develop

* Call CUDA driver api to launch the kernel compiled by nvrtc.
test=develop

* Disable for mac and windows.
test=develop

* Refine the codes to support manually specified num_threads and workload_per_thread.
test=develop

* Refine the CUDA kernel to support large dims.
test=develop

* remove assert.h

* change PADDLE_ASSERT_MSG to PADDLE_ENFORCE

test=develop

* fix tensorrt paddle_enforce

test=develop

* add_distributions_of_normal_and_uniform

* paddle/fluid/API.spec

* modify API.spec

* modified paddle/fluid/API.spec, test=develop

* modify paddle/fluid/API.spec, test=develop

* modify paddle/fluid/API.spec, test=develop

* fix some comment, test=develop

* modify API.spec, test=develop

* Add distributions of Categorical and MultivariateNormal test=develop

* fix pylint codestyle test=develop

* fix conflict file test=develop

* edit API.spec test=develop

* improve sample code test=develop

* modify api.spec test=develop

* enable inplace for affine_channel op, dropout op, test=develop

* remove dropout inplace for ngraph fails, test=develop

* test=develop add a argument for softshrink python api

* test=develop fix doc format 

test=develop fix doc format

* test=develop fix API.spec

test=develop fix API.spec

test=develop

elementwise broadcast function enhancement

Add user-friendly error message in optimizer ops to give a hint about the position sensitive problem of run(startup_program) (#19605)

* add extra error message hint in optimizer ops

* polish format & delete useless change, test=develop

* extract init judue from shape compare, test=develop

* add kernel for flatten_op, test=develop

* add kernel for flatten_op, test=develop

* fix the license and remove redundant code, test=develop

test=develop

* enable ngraph throught build_strategy test=develop

* add unittest test=develop

* put use_ngraph unconditional test=develop

* remove paddle_enforce test=develop

* remove paddle_enforce test=develop

* fix copyright test=develop

* limit for ngraph only test=develop

* add scatter_nd op, test=document_preview test=develop

* fixed the document, test=document_preview test=develop

* modify the notes, test=document_preview test=develop

* remove the ShareDataWith, test=develop

* test=develop
Fix the scatter op bug when use the add mode, and support the int64 data type of scatter_op Index(#18804).

* test=develop
Remove the PADDLE_ENFORCE and use PADDLE_ENFORCE_EQ

* test=develop
Remove the fix bug of scatter_add, and just add the support of int64 in scatter_add

* test=develop
Add the test case for scatter op, the test case just for index int64

test=develop

test=develop

* Add a interface to enable cudnn for inference.

* Add cudnn_placement_pass.
test=develop

* Set the default value of cudnn_enabled_op_types to null.
test=develop

* Write the common basic class, placement_pass_base, to refine the codes.
test=develop

* Call EnableCUDNN in unittest.
test=develop

* Refine cudnn_placement_pass tester.

* Enable the testing of cudnn_placement_pass in inference's unittest.
test=develop

* Add the check of op kernels.
test=develop

* using MKLDNNMemoryFormat = mkldnn::memory::format changes
test=develop

* PADDLE_ENFORCE update
test=develop

* update ngraph ops test=develop

* update unittest test=develop

* increase coverage test=develop

* remove unused PADDLE_ASSERT(_IS_NOT_ERROR)

* replace PADDLE_ASSERT with PADDLE_ASSERT_MSG

test=develop

Change backward_guard to optimize_guard to maximize the allreduce overlap

test=develop, test=document_preview

* add padding in linear_chain_crf op

* modify API.spec

* add linear_chain_crf_op.cc and linear_chain_crf_op.h

* remove useless unit test , test=develop

* modify API.spec, test=develop

* remove some blanks in nn.py , test=develop

* fix some bugs on nn.py and API.spec ,test=develop

* fix nn.py, test=develop

* fix API.spec ,test=develop

* fix bug of CI test in test_linear_chain_crf_op.py

* fix bug of CI test in test_linear_chain_crf_op.py, test=develop

* remove paddle_enforce, test=develop

* remove paddle_enforce, test=develop

* remove paddle_enforce, test=develop

* remove paddle_enforce, test=develop

* remove paddle_enforce, test=develop

* remove paddle_enforce, test=develop

* modify nn.py, test=develop

* fix API.spec, test=develop

* fix unittest bug, test=develop

* add place to reader, remove snappy dependency

* remove snappy dependency from train demo

test=develop

* - First set of modifications

- Compilation fixes

- compilation fix

- Another compilation fix

- Moved AcquireSoftmaxPrimitiveDescriptor call into handler

- MKL-DNN Softmax PD refactor

test=develop

- Compilation fix

test=develop

- another compilation fix

- cosmetcis

test=develop

- Compilation fix

- Fix to crash when softmax backward is created

* - Fixes after review of softmax refactoring

test=develop

* add retry_allocator for gpu, test=develop

* follow chengduoZH's comments, test=develop

* follow huihuang's comments,test=develop

* change f,l in enforce.h to be file,line, test=develop

* increase code coverage by adding unittests, test=develop

* fix CMakeLists.txt, test=develop

* 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

* fixed the code for coverage

* fixed the document,test=document_preview test=develop

- Refactor step 1

- Compilation fix

- Yet another compilation fix

- Even more compilation fix

- Lint fixes

test=develop

- Removed deprectaed PADDLE_ENFORCE occurance

test=develop

- Candidate fix to BN forward

- Lint fixes

test=develop

- Refactoring in data_layout_transform

- compilation fix

- Another comppilation fix

- Step further into darkness

- Yet another compilation fix

- Yet another compilation fix

- missing header

- compilation fix

- Added MKLDNN -> Paddle conversion in fetch op

test=develop

- Compilation fix

test=develop

- Lint

test=develop

- Mul fix

- Fix to MKLDNN MUL op and Elementwise MUL UT

test=develop

- Workaround for diffrent weights with groups representation Paddle vs
  MKL-DNN.

test=develop

- Candidate fix for 5D convolution with groups

- Refactor of fix for conv3d and conv2d in fetch op

test=develop

- Compilation fix

- Still same compilation fix

- Compilation fix

- Compilation fix

- Reverted refactoring of fixes

- Adapted test_conv2d_int8_mkldnn so it exects data in NCHW format
  not NHWC

test=develop

- minor fix in UT

test=develop

- Lint fixes

test=develop

* Modify the dropout op to multi-thread
test=develop

* define parallel
test=develop

test=develop