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.

* 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 bf16 uniform random initializer

* Remove duplicated section

* Change UT to CPU place only

* Put detail functions into anonymous namespace

* add input EpsilonTensor for adam

* update python api

* add unit test

* add npu test

* add more ut

* add fake interface for hook in static mode

* add unittests

* fix failed unittests

Reduce the time cost for the elementwise_add test case (#32628)

* add alltoall api, test=develop

* update 2.0 public api in nn

* replace Chinese character cause error in ci;synchronization with pr:#32588 to avoid 'ascii' codec in python2

* numbers used in paddle.nn.functional.norm but not imported

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

* deal with conflict.

* deal with conflict.

* jit.save/load support function.

* delete unnittest test_jit_load_model_incomplete.

* edit code according to CI

* Modify the documentation.

* add note to doc.

* fix cross_entropy calculation error

* add unittest and fix static

* Support list and tuple for parameters of layer_norm, multiprocess_reader, DatasetFolder and ImageFolder.

* add unittest for layer_norm.

* add require gpu for example.

* add sendrecv, test=develop

* fix bn docs

* fix unittest

* Modified params of some APIs to support tuple and list.

* fixed bug.

* Modified the return value of tensor.grad from numpy as tensor.

* Modify unittests.

* fixed bugs.

* Add warning info for x.grad

* fixed unittests which used x.grad

* fixed bug.

* change prepend to append

* fix ut

* add testcase

* fix ut

* fix test_custom_relu_model

* fix model parallel

* rm parallel_help.py

* add embedding

* support backward return None.

* edit unittest.

* edit code according to CI

* Improve error information

* skip op has no fp16 kernel

* add ut

paddle.to_tensor will be translated to paddle.assign in Dy2stat, however paddle.assign doesn't support int, float, bool. This PR added the supports.

* add pp dygraph, test=develop

* Dygraph reocmpute

* unitest for Dygraph reocmpute

* dy recompute remove unitest for win and mac

* support save/load binary format tensor

* Fix error when create cudaplace

* Fix error when create cudaplace

* Fix error when create cudaplace

* get devive context from pool.

* move define of 'SerializeToStream' and 'DeserializeFromStream' to 'lod_tensor.cc' and 'selected_rows.cc'.

* support complex object

* improve coverage.

* improve coverage

* improve coverage.

* fix a bug.

* polish API

* save/load program

* paddle.save/load: layer

* deal with conflict

* if PY2, block test_paddle_save_load.TestSaveLoadLayer

* polish code.

* polish code

* edit unnittest

* The condition for object to be identified as state_dict becomes strict

* use 'core._cuda_synchronize'

* add pipeline layer

* update

* fix bug: when x.dim < y.dim, the result of compare_op is inverse to expected result

* support the cuda for fix the compare broadcast bug