* move elementwise raw

* fix

* fix

* mv paddle/fluid/platform/device/xpu/tests 2 test/xpu/cpp

* add missing cmake

* update

* update

* update

1. add type caster for paddle's complex type, to allow pybind to automatically cast it with python's complex type;
2. add complex64 and complex128 data type for `libpaddle.Tensor`'s element get and set(which is required to perturb an element to get the numerical derivative)
3. add support for cuda pinned place in `libpaddle.Tensor` element get and set

---
4. fix a bug in op code generation.(Creation of output folder in concurrent with parsing op yamls.)

* [AMP] Add python API for collecting operator stats.

* Fix import and polish codes.

* Add more unittest.

* Add doc for the new APIs.

* add autogen code support for spectral_norm

* bug fixed

* fix PR-CI-Static-Check fail

* support run haokanctr model in heterps-models

* polish setup.py

* polish JVM_LIB in evn_dict

* align infer auc with DistPsArch pre-stable

* async and multi thread data feed

* rewrite dense tensor intialization

* async infer shape and reuse memory

* add scatter composite rule.

* add public_python_api

* add python unit16 support.

* fix code style.

* add cinn to makelist

* cinn unsupport uint16, forbidden cinn when dtype==uint16.

* update assign_pos

* update attention_lstm

* update barrier

* update batch_fc

* update beam_search

* update beam_search_decode

* update bilateral_slice

* fix bug

* Handle Structure kernel for InterpreterCore::RunOperator

* fix bug

* fix rocm compile

* fix rocm compile

* Revert "fix rocm compile"

* test

* revert test and update cmake

---------
Co-authored-by: Nchenruibiao <chenruibiao@baidu.com>

* Support ignore_index for c_softmax_with_cross_entropy_op.

* Polish code. Remove useless comments and add Testcase.

* Polish code for TestCase.

* Polish code.

* Polish code style.

* Polish code.

* Change loss calculation formula and ignore_index dtype.

* Polish TestCase.

* Fix bug of c_softmax_with_cross_entropy_op_xpu_op. Attribute 'ignore_index'
dtype is int64_t.

* add FP16 for multinomial

* fix input data

* update code

* fix FP16

* fix code

* support auto generate for prelu

* op_compat 中增加输入参数

* del attrs ; add kernel data_type

* add PreluGradInferMeta

* stride slice fp16 and bf16 unitest

* fix code style

* add self.dtype

* add autogen code support for sigmoid_cross_entropy_with_logits

* add inplace

* add autogen code support for merge_selected_rows

* bug fixed

* force sync batch norm grad sequential

* fix bug in proto

* add utest

* change op with xpu

* change range yaml

* fix bug in generate_op.py

* [CINN] pass global seed to CINN

* fix cu not include cinn/runtime/flags.h bug

* fix DefaultCUDAGenerator should has device id bug

[CodeStyle][C416][C417] rewrite unnecessary comprehension with function call and use generator instead of map (#52140)

* codestyle c416 c417

* fix error

* fix inc

* unify all C4 rules into one

* fix inc

---------
Co-authored-by: NSigureMo <sigure.qaq@gmail.com>

* add gloo  send_recv

* fix code_stype

* fix CI bug

* fix send_recv.cc

* add send_recv without sync_op

* fix send_recv test

* fix gather.cc

* pad3d add unittests of fp16 and bf16

* pad3d add unittests of fp16 and bf16

* fix cuda place

* fix random to uniform

* fix class name

* fix fp16 max relative error to 1.5e-3

* add dytpe register for onednn

* add pad uint16 check of common.py

* remove check_eager

* test_check_grad --> test_check_grad_normal

* Clear the infrt-related code

* remove tools/infrt