* Support Mixed Precision training in @to_static

* fix block.vars logic

* fix GPU training loss diff

* remove unused code

* replace boost::algorithm::ends_with with self define ends_with function

* remove BOOST macro in certain operators

* remove boost::lexical_cast

* add test for string_helper

* add more test case for string_helper

* modify join_string func and test case

* fix build_strategy_test failed bug

* remove string_helper_test from parallel_UT_rule.py

* add NPU support for zero_copy_tensor.

* revert unnesessary codes.

* revert unnesessary codes.

* add not_equal NPU op

* add not_equal NPU op

* add not_equal NPU op

* add not_equal NPU op

* fix bug of p2p for partial

* fix error

* revert commit id 34212

* fix paddle.optimizer test=document_fix

* fix paddle.optimizer test=document_fix

* fix bugs in paddle.nn.functional document test=document_fix

* fix bugs in paddle.nn.functional document test=document_fix

* fix bugs in paddle.nn.functional document test=document_fix

* fix bugs in paddle.nn.functional document test=document_fix

* add gradients_with_optimizer api

* modify gradients_with_optimizer

* add gradients_with_optimizer api into paddle.auto.backward_mode

* add gradients_with_optimizer test case

* add doc for gradients_with_optimizer

* add doc for gradients_with_optimizer

* notest;benchmark-cpu

* notest;benchmark-cpu

* notest;benchmark-cpu

* notest;benchmark-cpu

* notest;benchmark-cpu

* notest;benchmark-cpu

* notest;benchmark-cpu

* notest;benchmark-cpu

* notest;benchmark-cpu

* notest;benchmark-cpu

* notest;benchmark-cpu

* notest;benchmark-cpu

* fix

* [NPU] Support npu kernel for assign_value op

* move test_assign_value_op_npu.py into unittests/npu folder

* correce copyright year; add TestAssignApi class using NPUplace in test files

* add crop npu op

* code style check

* update code to remove usless headers

* add testcase and upsate code

* fix backward bug

* format code style

* add test case for grad tensor accumulator

* supplement the function of slice

* edit unittest

* strided_slice_op support .

* polish error message.

* polish error message.

* polish code.

* polish unittest.

* polish code.

* polish code

* polish error message.

* add layer segment

* add segement for transformer

* add utest

* delete install_trt in release docker

* fix whl package path

* support Kunlun2

* support KL2

* support KL2

* support more dim for mul op npu

* update unit test according to reviewer's comment.

* support 1f1b for pipeline

* add utest

* add send_partial/recv_partial

* support amp for pp

* fix logger

* fix paddle.optimizer test=document_fix

* fix paddle.optimizer test=document_fix

* add basic APIs

* add attr_types

* follow comments

* change pass attr types

* add set pass attribute codes

* refine PADDLE_THROW

* test=develop

* update identity

* add unittest

* notest,test=mac_py3

* modify comment & testname

* test=document_fix

* update comment

* test=document_fix

* activate all of the CI

* notest;test=cpu-benchmark

* benchmark-cpu

* notest;test=cpu-benchmark

* notest;benchmark-cpu

* notest;benchmark-cpu

* notest;benchmark-cpu

* notest;benchmark-cpu

* notest;benchmark-cpu

* notest;benchmark-cpu

* fix

* fix

* add test_ci_model_benchmark.sh