* optimizer conv alog speed

* code polish

* remove useless code

* fix compile error

* fix cpu compile error

* not use cudnn alog t

* add search cache max number

* polish code

* fix cache test bug

* add groups data format to conv args

* fix cache test bug

* fix cudnn_deterministic bug

* fix test switch auto tune bug

* fix test swith autotune bug;

* fix conv cache bug

* fix cache test error

* fix cache test bug

* fix windows mac compile error

* fix workspace search error

* update cudnn cache

* fix cache test bug; test=develop

* fix autotune swith test error

* polish code

* oplish code

* fix unique_kernel bugs

* fix unique kernel cu bugs

* add python final state

* fix bug

* fix bugs

* fix bug

* fix bug

* revert impl, final state mul not support selected rows

* fix softmax use cudnn error

* add softlable false unitest

* revert loss.py

* make tensor_util contains no cuda code

* refine isfinite

* revert ut

* move isfinite function to its op

* fix test

* fix compile

* std::isnan is not defined for int type on windows

* fix windows compile

* fix fp16

* fix rocm compile

* revert gradient node

* fix arm compile problem

* fix format

* fix ci error

* add legacy_api.yaml

* set merged_momentum inplace only

* support inplace optional<vector<tensor>>

* add dygraph_mode api

* optimize TensorToConstDenseTensorPtr

* 【Hackathon No.34】优化 poisson op

* [poisson] code style fix

* modify code style

* prevent from big number

* modify code style

* modify code style

* modify import

* modify import

* modify code style

* Adapt minlength attr for bincount

* while support for python container.
It is convenient to convert more dynamic graph codes into static graphs.

* cond support python container

* rename the member function of SparseTensor

* use shorter function names

* erfinv

* fix some tiny issues

* Update scope.h

* typo

* Update dense_tensor.inl

* move distribute_fpn_proposals

* fix some code

* fix yaml bugs

* add set dtype

* move proposal_impl to funcs

* fix compile bugs

* [CustomDevice] add profiler apis

* migrate CalculateEstOccupancy into cuda_tracer

* update

* add ut

some python c api use final state

* add forward inplace final state api

* fix bug

* fix reshape

* fix coverage

* add inplace info for erfinv, lerp, put_along_axis

* fix put_along_axis infer_meta

* fix format

* update yaml

* fix

* add group_norm dyanmic plugin

* transfer trilinear op to phi, change name from trilinear_interp_v2 to trilinear_interp

* reserve linear_interp param

* change testcase scale if-branch

* testcase test_imperative_case

* fix trilinear testcase

* import paddle in test_trilinear_interp_v2

* [OpAttr]Squeeze axes support Tensor

* add support_tensor

* fix unittest

* fix coverage

* change to async mode for xpu multi-card training in static graph mode

* minor bugfix

* irrelevant. move to another pr

* move change to other pr

* fix stream issue

* fix 'stream not meet with current context' error

* fix branch diverge, test=kunlun

* transfer bilinear op to phi, change bname from bilinear_interp_v2 to bilinear_interp

* reserve linear_interp param

* fix cross device import

* use addKernel

* fix compile

* remove elementwiseAddto

* add return

* fix custom place

* xpu unittest grad compute supports more types, *test=kunlun

* add instance norm xpu, *test=kunlun

* transfer bicubic_interp op to phi, change name from bicubic_interp_v2 to bicubic_interp

* test final_state_bicubic_interp api

* testcase match imperative case

* fix squared_l2_norm bug

* update buffer.h

* move check finite and unscale kernel into phi

* move infershape into phi

* move update_loss_scaling kernel into phi

* remove original kernels

* move update loss scaling infershape into phi

* add header for xpu and npu

* solve coverage failed

* fix npu test failed

* remove mutable data in cu file

* fix new executor failed

* add valid check for meta tensor output