* add composite rule for squeeze

* fix pre commit

* fix pre commit

* simplify rules

* arrange code

* fix int axis

* simplify squeeze axis rules

* bugfix

* fix pre commit

* update codes about pad3d

* add codes about Tensor type Padding

* update

* 更新单测文件

* format code style

* update and to &&'

* rewrite codes about pad3d

* add codes about converting paddle pad format to tensorrt pad format

* fix some errors

* 指定trt版本范围

* 修正dims初始化方式

* fix code style

* update test pad values

* 指定pad3d trt版本

* 更新 单测 文件范围

* 更新单测文件

* update pad3d paddings convert codes

* update pad3d

* add static mode support

* update test file

* fix bugs about dynamic mode test codes

* fix bug and add limite in op_teller

* use a new padding convert method[ITensor* padding with using Slice to split the pre_pad and the  post pad]

* fix PADDLE_THROW grammaly error

* update test codes

* 添加对于Tensor padding 的 size 判断

* use '=' replace import submodule in partial_program.py

* roll back and test

* remove all 'from paddle.amp.auto_cast import' from partial_program.py

shape support bf16

* Add fp16 and bf16 to the checking dtype list of rand apis.

* Remove the checking of raising TypeError.

* migrate fill_constant to paddle.tensor

* move fill_constant to paddle.tensor and repalce the reference

* add missing fill_constant replacement

* fix typro

* remove unused import fill_constant

* fix zeros import error

* fix circle import

* fix layers.zeros

* fix unitest

* fix unitests

* fix unitest

* use paddle.full replace fill_constant in samplecode

* fix sample code

* recovery xpu test

* recovery xpu test

* fix circle import

* fix utils import error

* fix utils error

* fix circle import

* redo

* fix circle import

* fix prim fill constant import

* fix type error

* fix increase error

* fix test error

* fix fill_constant

* [Custom Op] Support extra_library_paths of load

* change API order

* change unittest parameter order

* [Tensor Operants & Prim-Relevant] Tensor supports compare operants

* fix dependence of test_comp_static

* fix unit test

* fix: fix broadcast object list

* fix: fix scatter object list

* remove fluid.layers.collective and related unittests.

* fix layers.init import

* fix all_gather method

* remove unused args in all_gather method

* fluid clean: remove fluid.ir to framework.ir and some funcs form fluid.layer.io to incubate.

* delete fluid.ir

* add_fp16_for_diff

* fix doc_for_fp16

* finish task

* fix some question.

* fix error

* change unittest:zeroDim.

* limit protobu_version > 3.20.0

* limit protobu_version > 3.20.0

* update unittest of fp16 for slice op

* set atol to default value

* update max_relative_eror

* support fetch empty tensor on CPUPlace

* fix the shape in unittest of empty output

* add bn vjp

* fix example

* fix code

* fix code

* fix cinn case

* fix code

* fix example

* fix code

* fix example

* fix example

* added paddle.polar to paddle

* added paddle.polar unitest

* failed to use pytorch for evaluating results, and changed with numpy implementation

* updated code style

* updated __init__.py

* updated code style

* fixed unitest code

* lkh test polar

* polar add

* fixed errors and optimized code

* fixed error

* optimized polor api

* updated code style

* updated code style

---------
Co-authored-by: Ndiadestiny <1247889154@qq.com>

* add sqrt composite rule/test

* add sqrt composite rule/test

* fix ops/sqrt, add cinn test

* fix sqrt_comp

* fix sqrt_comp

* fix sqrt_comp

* fix

* fix codestyle

* fix codestyle

* add fp16 test

* add ops/sqrt

* fix

* fix

* fix unitest

* fix

* fix

* fix

* init unit test commit, contains register thinking

* support inplace

* get inplaced x.grad

* Try support inplace and hook at the same time

* Support inplace, need debug

* Support inplace successfully

* Inplace use Tensor&, consistent with Tensor*

* fix MapPlainOutputs bug

* fix double grad inplace error

* rename flash_attn_raw to flash_attn_unpadded

* fix static api

* fix static return

* add dynamic support

* add more test

* fix bug

* change test

* change test

* add fp32 grad plus fp16 param in adamw

* add python UT

* fix test case

* in test_adamw_op py file, force the moment2 value LE 0

* add a compare option

* remove bf16 fused adam kernel case

* update env setting

* update pass logic

* dist op support bf16

* backward cast update

* update setting

* update backward

* revert amp pass

* update fp16 backward logic

* register c_embedding bf16

* revert engine

* add unitest

* add unitest

* update unitest

* update cmake

* update math

* update math.py

* update unitest

* update unitest

* revise unitest

* revise unitest

* update unitest

* update unitest

* update unitest

* split pass

* fix compile

* fix ut

* more time

* modify ut

* reduce dim

* fix compile

* reshape weight

* tensor

* remove enforce

* static shape ut

* batchsize

* reorder pass

* minus test cases

* windows timeout

* windows time out

* remove test for windows

* correct

* sssss

* xxx