* fix compile error on jetson platform

* remove unused head file

* rm decode_jpeg op on jetson platform

 [CHERRY-PICK2.1]Remove paddle_custom_op dynamic libraries, and link to FLUID_CORE on windows (#32583) (#32769)

* Remove paddle_custom_op dynamic libraries, change link to FLUID_CORE on windows, and check copy_to

* fix CI

* Add casting initializers for bf16 training

* Changes after review

* Correct test and add comment
Co-authored-by: Njoanna.wozna.intel <joanna.wozna@intel.com>

* Add deprecated warning info.

* Add unittest for deprecated decorator.

* Add warning info for tensor.grad

修复了py_layer_op由于没有析构PyLayerContext造成内存(显存)泄露的问题。

原始pr：#32707

* clear 'BasicEngine' when an exception occurs in the backward. (#32546)

* clear 'BasicEngine' when an exception occurs in the backward.

* deal with conflict.

* deal with conflict.

* forward return any type. (#32661)

* revert data_generator

* add setup.py

* base changes for fix

* minor change

* fix for bwd kernel

* removed unnecessary import

* implemented reviewers suggestions

* CI fix

cherry-pick:change softmax_with_cross_entropy_op's parameter name from softmax_switch to use_softmax (#32750)

* change parameter name from softmax_switch to use_softmax, test=develop

* cherry-pick:change parameter name from softmax_switch to use_softmax, test=develop

avoid polluting logging's root logger

* add to_grayscale, normalize

* add rotate

* add vfip and hflip

* add crop center_crop

* add utils

* add utils

* update utils, add raise for some cases

* add padding, support constant, reflect, replicate, circular same as paddle.pad

* update rotate

* using utils func in [v|h]flip

* add get-image-[n,c,w,h] axis utils

* add get-image-[n,c,w,h] axis utils

* align

* update

* remove default value in utils func

* add assert for pad

* update assert paddle image

* support rotate fill func

* raise valueerror for pad

* remove typing, py2 dont support

* init uinttest for transforms tensor

* add resize op

* register [normalize hflip crop center_crop resize transpose] imagenet

* register [normalize hflip crop center_crop resize transpose] imagenet

* fix bugs, (w, h) getter and import

* add _get_image_size for tensor image

* add pad vflip for tensor image

* add unittest for tensor transforms

* update transforms unittest for converage CI probelms, test=develop

* update

* update

* update

* fix `get_shape` for tensor backend

* update

* update

* add more resize tests

* update

* update for ci test

* update

* remove redundancy code

* update uinttest, and set tensor image to hwc by default

* add tensor backend

* fix copyright doc, rm comment code, add pil unittest

* update data_format to `chw` for tensor

* coverage notest,test=coverage

* update

* update

cherry-pick #32561

Dy2stat failed when user writes return paddle.to_tensor(xxx), the reason is that visit_Expr doesn't work when the Expr is in return. Some other statements may trigger same bug. To fix it, we re-wrote a transformer to transform paddle.to_tensor to paddle.assign for all Call nodes.

* update, test=develop

* add relu6_ hardsigmoid_ leaky_relu_ Inplace APIs

* add softmax_with_cross_entropy_ Inplace API

* add clip_ scale_ add_ subtract_ Inplace APIs

* add wlist

* fix parameter of scale api

* add add_n_ Inplace API and remove log_ Inplace API

* fix elementwise_add_ and elementwise_sub_ broadcast problem

* elementwise inplace api give error message before run the op

* use broadcast_shape in elementwise inplace op

* add 8 inplace apis that is auto generated

* add unittest for all inplace apis

* add decorator for inplace apis in static mode

* fix windows blas fail of exp inplace api, change array_equal to allclose

* add flatten inplace api

* add flatten unittest

* fix flatten unittest

* add decorator

* fix grad.numpy in test_pylayer_op

* unsupport softmax_with_cross_entropy_

* add test_inplace_softmax_with_cross_entropy to static_mode_white_list

* delete __all__ in inplace_utils

* delete activation inplace function and add Tensor.inplace_func

* change paddle.inplace_ to Tensor.inplace_

* fix little problem

* add paddle in inplace_utils

* add op read_file and decode_jpeg

* Add dlnne engine runtime

* Remove <const_cast> and remove unrelated modify with dlnne, +clang-format

* Add copyright message

* Add some paddlepaddle_pass to support more networks

* Add delete dropout_op pass
Co-authored-by: Ndenglin-github <82362191+denglin-github@users.noreply.github.com>

* Add bf16 uniform random initializer

* Remove duplicated section

* Change UT to CPU place only

* Put detail functions into anonymous namespace

This is cherry-pick of #32281

cherry-pick of #32666

* add fake interface for hook in static mode

* add unittests

* fix failed unittests

* jit.save/load support function.

* delete unnittest test_jit_load_model_incomplete.

* edit code according to CI

* Modify the documentation.

* add note to doc.

* edit paddle.save/load API

* Update io.py

edit doc

* delete cpython-37.pyc

* Update io.py

edit doc

* Update io.py

recommit

* Update io.py

recommit

* Update io.py

recommit

* Update io.py

recommit

后续修复计划是啥

- Executor is nt always having FLAGS_use_mkldnn set to true