* add reduce_sum grad kernel

* add reduce_grad

* modify reduce grad

* update reduce grad functions

* fix build error

* add argument mapping

* move cast input after grad

* add dims.size=1 cpu reduce_sum grad compute method

* update reduce grad GPU

* remove raw reduce_sum_grad kernel

* modify header files

* add namespace funcs for reduce_grad_funcstions

* first commit

* fix

* fix

* fix

* fix

* fix

* fix xpu and npu

* fix

* mv erf op to phi

* fix as review

* fix as review

* fix format

* [Phi] Fix infershape if encounter TensorList and Attr("XXX")

* add InferShapeArgumentMappingContext

* pad infershape

* fix code

* fix

* add set dtype

* transfer nll_loss infershape into phi

* Fix issues with intermediate kernels

* Fixed CI issues

* transfer the infershape of accuracy op into phi

* add set_dtype

* add setdtype

* add onnxruntime predictor

* Add code comments

* support link paddle2onnx onnxruntime

* support onnxruntime with python

* support onnxruntime with python

* support onnxruntime with windows

* paddle2onnx compile with windows

* supoort windows compile

* supoort windows compile with onnxruntime

* supoort windows compile with paddle2onnx

* supoort mac compile

* compile with mac

* compile with mac

* add code comments

* fix remind word

* code optimization

* add test case

* add test case

* add inference demo_ci test case

* fix compile paddle2onnx with no python

* add inference demo_ci test case

* add inference demo_ci test case

* add inference infer_ut test case

* support c go api and test cases

* add converage test case

* add converage test case

* add capi test case

* add capi test case

* [phi] move matrix_power op

* MatrixInverse fluid -> phi

* modify the CMake to fix compile bug

* delete useless comment

* mutable memory -> phi Alloc

* modify the include file

* modify the include file

* fix bug in CI compiler

* [PHI] Migrate where_index op

* [PHI] Fix where_index infermate

* [Phi] set where_index out data type

* update reshard for while sub block

* fix code format error

* add selected row op and fix bug in ctest

* modify the date

* fix bug in npu and xpu

* modfiy the include file

* [NPU] Fix fill_constant int64 for CANN 5.0.4 (FillD does not support int64)

* [NPU] fix set_value for CANN 5.0.4

* first commit

* cpu kernel

* first version

* fix compile error

* fix compile error

* delete v2

* fix

* fix

* add alias

* fix

* fix

* fix

* fix error

* fix

* fix

* fix

* fix format

* move segment_pool to phi.

* mark summed ids as optional tensor.

* fix as reviews.

* update. test=develop

* fix. test=develop

* fix. test=develop

* add the infer shape meta for the graph_send_recv

* move the infershape code to another file

* move argsort to phi

* refine files

* remove mutable_data

* fix unittest

* fix unittest

* remove infershape

* update infershape

* fix ci

* fix ci

* fix ci

* fix

* fix

* fix

* fix

* fix

* fix

* fix

* move arg min max to phi.

* move infermeta.

* fix as reviews.

* add tril_triu for xpu, *test=kunlun

* add tril_triu for xpu, *test=kunlun

* add tril_triu for xpu, *test=kunlun

* add tril_triu for xpu, *test=kunlun

* add tril_triu for xpu, *test=kunlun

* add trt.execute

* merge trt.engine type

* update return op

* update comments

* fix style

* fix style

* move dropout to phi; test=develop

* fix xpu, npu compile error; test=develop

* move viterbi to phi

* move infershape to phi

* update infershape

* fix

* resolve conflicts

* save code

* fix bug of set_value

* add coverage test

* support code auto-gene for sparse backward api

* fix bug of intermediate api and name of return var