* fix typos: is_integer in attribute.py
* add more test cases for fft

* [NPU] add int64 support for scatter op

* [NPU] delete debug codes

* [NPU] optimize codes

* first commit

* more uts

* file name duplicated

* timeout

* Update CMakeLists.txt

change TIMEOUT from 120 to 240

* Update CMakeLists.txt

more time

* Update CMakeLists.txt

timeout

* Update CMakeLists.txt

60s

* dequantize matmul and matmul_v2 Y weights in qat2_int8

* review fix

* split conv and mul tests, add matmul test

* fixup

* fix ci build

* remove unused variables

* formatting fix

* remove extra newline at end of file

* fix inplace addto pass

* update

* fix ut

* improve ci coverage

* fix musl ci compile error

* fix launch_utils.py. test=develop

* fix launch_utils.py. test=develop

* optimizer __call__ to make dygraph faster

* fix return type

* remove cpp Layer

* add prior_box for kunlun

* update

* update CMakeLists

* fix fc_fuse

* modify cmake notest,test=windows_ci

* retrigger all the ci

* optimizer __call__ to make dygraph faster

* fix return type

* add angle_op

* Add new unittests

* Replace I with O channel for filter groups

* Undo changes affecting other operators

* Fix oneDNN namespace typo

* Fix code format error

* add regularation and Nesterov for mergerd_momentum

* refine unittest for use_nesterov attr

* refine op check

* refine code

* fix bug

* refine code of regularization_flag

* delete useless code

* [Auto Parallel] elastic support auto parallel re-launch

* [Auto Parallel] elastic support auto parallel re-launch

* fix ci issue

* fix ci issue

* fix rank mapping unittest

* fix rank mapping unittest

* fix ci issue

* fix ci issue

* fix ci issue

* fix ci issue

* fix ci issue

* fix ci issue

* fix ci issue

* fix ci issue

* fix ci issue

* fix ci issue

* fix ci issue

* fix ci issue

* fix ci issue

* add diff op, test=develop

* rm some notes, test=develop

* update diff doc

* update sample code

* fix diff api params and example code, test=develop

* modify for wincheck-inference case

* modify according to zhouwei's comment

* open with_trt and block failed unittests in windows

* test

* support data_format='NHWC' for prelu channel mode

* [Auto Parallel]  Add the unified cluster representation

* [Auto Parallel] Add the graph class for physical mapping

* [Auto Parallel] Add the simple physical mapper

* Set the timeout of the mapper

* Merge the upstream develop unittests cmake files

* Fix a bug of the process group

* Remove mapper unittest from platforms which is not GPU

* Move the instantiation of process group after resharding

* Add the local id for devices

* Update the rank mapping format

* Add some comments

* Remove the related files about mapping

* Update the unittest for auto mapping

* Remove unused rank_mapping unittest

* Improve the unittest coverage

* Improve the unittest coverage

* add scope_guard

* 1. fix control flow cases 2. fix calc_gradient

* add expand_v2/expand_as_v2 for kunlun

* update expand_as_v2

* update expand_as_v2

* support float16/bool

* update xpu.cmake

* amp.decorate optimizers set to None is ok

* refine unittest

* add unittest and refine example code

* refine unittest

* native commit for triple grad of sigmod

* Updated unittests files

* init functional jacobian api

* Updated trible_test func

* Updated gradient_checker & test_script

* finish test with dtype float32

* add float64 test case

* polish code

* use atol=1e-5 with dtype float64

* fix for ci

* set timeout for test_jacobian

* fix dygraph grad to support high differential

* polish API docstring

* Updated gradient checker and some related files

* fix double grad strip error for high differential

* fix double grad strip error for high differential

* Add Sigmoid triple grad tests

* fix dygraph double grad dtype error when calling for high differential senario

* Updated triple grad teses func

* Use np.random to initialize ddx

* Updated triple_grad_check func

* add todo for gradient checker and refine some comments

* remove additional code

* add test for warnging in backward.py

* format python code

* support multi input in triple gradient checker

* Add matmul triple grad kernel

* Updated comments of TODO

* Supported some special tests

* Change code-format to follow CI std

* Updated gradient_checker.py

* Fix conflicts

* Removed unnecessary printing log

* Change code style to follow CI std

* support batch in jacobian and hessian

* add batch jacobian and batch hessian

* Add batch_jacobian test, draft version

* [New features] Add elementwise_mul triple grad kernel (#37152)

* Add elementwise_mul triple grad kernel

* Removed InplaceInferer and polished code

* Add numerical_batch_jacobian,numerical_batch_hessian and tests

* Support batch_jacobian and batch_numerical

* Use pre-commit to check code format

* Update doc, polish code, add unit test

* Reset the TIMEOUT properties of test_jacobian to pass CI
Co-authored-by: Nlevi131 <limaolin01@baidu.com>
Co-authored-by: NJiabin Yang <360788950@qq.com>

Co-authored-by: NChen Long <1300851984@qq.com>

* [Auto Parallel]  Add the unified cluster representation

* [Auto Parallel] Add the graph class for physical mapping

* [Auto Parallel] Add the simple physical mapper

* Set the timeout of the mapper

* Merge the upstream develop unittests cmake files

* Fix a bug of the process group

* Remove mapper unittest from platforms which is not GPU

* Move the instantiation of process group after resharding

* Add the local id for devices

* Update the rank mapping format

* Add some comments

* Remove the related files about mapping

* Remove unused rank_mapping unittest

* Improve the unittest coverage

* fix data parallel when VOCAB var in program

* test

* test

* rm test

* update

* update

* update

* add unittest

* update

* update save

* Fix bugs when bias is none for static graph for fused_attention op.

reset_inplace_version removes all inplace related records to VarBase/VariableWrapper, the essential purpose of which is to let you use inplace operations as if using its non-inplaced version, which of course will cause unexpected consequences if not used with care.

This is essentially a hack interface to satisfy one specific request