* Modify data download function, and support unittests of inference APIs on windows, test=develop

* The import error compatible with py2 and py3, and fix unittests problems of inference APIs on Windows, test=develop

* add lstm cudnn of padding data and refine cudnn codes

update the check for the dtype check for the argmin, argmax

* - Added fix to mobilenet

* - compilation fix

* - Fix to conv2d grad oneDNN with groups

test=develop

support heter-xpu-ps

test=develop

* support load infer model format state dict

* add unittests

* remove keep name table

* recolve circle inport

* fix compatible problem

* recover unittest

* polish doc and comment

show unittest time even if unittest failed

* fix dropout bug in backward when input is 1d tensor, test=develop

* add test case and refine error message, test=develop

* refine error message, test=develop

* Add mkldnn bfloat16 option to C-API

* Add test for bfloat16 gpu

* Change coverage test

* Repair capi_gpu test

test=develop

* optimized transformation form tensor to numpy, test=develop

* Modify fetch op handle, from memcpy Sync to memcpy Async, test=develop

* modify CUDAPinnedPlace to CPUPlace, test=develop

* modify CPUPlace to CUDAPinnedPlace, and set default inplace to false, test=develop

* revert fetch_op_handle, add fetch_async_op_handle, test=develop

* revert fetch_op_handle, add fetch_async_op_handle, test=develop

* fix error msg report, test=develop

* fix bug in cpuplace, test=develop

* fix bug in unmerge and tensorarray modle, test=develop

* fix bug, double copy gpu memory, test=develop

* fix chenweihang¡¯s review advice, test=develop

* fix eigen in push sparse; fix hadoop command
test=develop

* add log in load_combine_op
test=develop

test=develop

* fix the argmin,argmax op for the paddlepaddle 2.0， add checkPoint for the argmax/argmin

* Show more possible problems with build_and_check in file paddle_build.sh, test=develop

* Remove test codes modified in file device.py for build_and_check, test=document_fix

* Fix missing blank space in file device.py, test=document_fix

* Final process via summary_check_problems function, test=document_fix

* fix fluid.embedding

* add stack_op to CMakeLists

* add dim=3 support for scale op

* add trt stack op, test=develop

* remove debug message

* add stack plugin serialize

* remove slice, scale op, will add later

* enhence error message

* revise trt ernie test to conver the stack op CI testi, test=develop

* add stack op serialization

* fix test shape after adding stack op

* remove slice op, will add after implementing serialization

* roll back to min_graph=5 to avoid using slice op

* fix scale op output layer

* implement stack op createPlugin

* use workspace and move the defination to .cu

* move stack plugin creator definition to .cu, test=develop

* add check for bernoulli and register bool for unsqueeze

* follow comments

This reverts commit 02083bda.