* fix multiple documentation error, test=document_fix

* fix more rst syntax errors, test=document_fix

* fix format issues in docstring, test=document_fix

* fix tensorrt output shape error

* fix unittest tensorrt_engine_op_test

* fix code style for unitest

* use has_grad instead of train_mode

* add vlog for debug

* fix ut

* fix ut

Reduce exception type so that if covert_to_static failed, it reports right error message.

[cherry-pick 2.0rc1][inplace] Add ShareHolderWith for class Variable and SharePlaceholderWith in VarBase.detach() to share the same Tensor/SelectedRows (#29267) (#29359)

[cherry-pick 2.0rc1][Dy2Stat] Fix bug: Do not use gast.Subscript to replace gast.Name in when transforming for_enumerate_loop (#29310) (#29361)

* basic impl of type promote

* add comment & another testcase

* fix complex bugs & support python op promote type

* fix failed unittests & polish code

* add unittest for coverage

* change to only promote complex type

* polish code details

* polish several comments

[Cheery-Pick 2.0.0-rc1][Dy2stat] Add a decorator paddle.jit.not_to_static to support that not to convert a function in Dynamic-to-Static. (#29253) (#29340)

Usage scenarios：A function could have run successfully in static mode,  you can use it to decorate a function in the following cases:
  1. An unknown error occurs in the dynamic-to-static conversion process of the function;
  2. In the internal implementation of the function, it has two branches: dynamic branch and static branch;
  3. Users don't want to convert the function in the process of dynamic to static.

* use has_grad instead of train_mode

* add vlog for debug

* fix ut

* fix ut

* move temporal_shift to functional

* fix the warning of reducer (#29323)

* fix warning of fleet (#29317)

* Fix doc of fleet api (#29282)

* Change the api of DataParallel and Fleet (#29224)

* Add pure fp16 training with master weights. (#27712)

* add the weight decay func for the momentum op

* Add the multi_precision function in Momentum Optimizer.

* Make sure that the initial value of master weights are same with the fp16 weights.

* add static loss scaling.

* add the rescale_grad function in the pure fp16 training.

* use the original momentum updating method.

* Polish some codes, such as variable names.

* add docstring for apis.

* update the var creation details of _create_master_weight.

* not modify codes about imperative momentum updating.

* Fix the error of test_dist_sparse_tensor_load_momentum UT.

* add unit test for multi precision fp16 training.

* add more unit tests for CI.

* Use lower threshold values for allclose comparing in test_multi_precision_fp16_train UT.

* fix DATA_HOME path in win

Cherry-pick of PR #29226

fix nll_loss doc;test=document_fix

* add compile option WITH_TENSORRT

* add WITH_TENSORRT to ci paddle_buils.sh

* add WITH_TENSORRT to paddle_build.sh

* change FATAL to WARNING when TensorRT is not found and WITN_TENSORRT=ON, just to pass ci-py3 temporarily

* hot fix complle failed in gcc4.8

* fix failed unittest

* add lamb optimizer and unittest

* fix momentum resume training

* fix momentum acc

* add complex64 and complex128 type; add +-*/@ and slice opreator for complex types

* add test cases for complex elementwise, matmul and getitem unittest

* add test cases for complex types

* add test cases for complex matmul unittest

* fix fleet api doc

* The leaf tensor concept is exposed and the gradient accumulation of leaf tensor

* The leaf tensor concept is exposed and the gradient accumulation of leaf tensor

* fix coverage

* fix api doc

* fix CI unittest

* fix CI unittest

* fix unitest

* empty tensor does’t need inner_var_

* fix some error message

* fix en doc, test=document_fix

* add blank after code declare, test=document_fix

* refine doc of dropout, test=document_fix

* refine npair_loss and dropout, test=document_fix

- Make sure that oneDNN memory descriptors are created only once at
first iteration.

* lazy import for scipy

* rm unused check