* elementwise_add_grad Op optimization  (#29575)

* optimize for long width for elementwise (#29602)

* refine (#29622)

* delete the code for fp16 optimization because it is not faster than common template code (#29715)

* fix the shape choose of vectorize for cuda

* optimization for fp16 elementwise add (#29744)

* Fix the compiler error for half type (#29799)

* refine the compiler error for half2 operation (#29816)

* fix the compiler error when gcc4 cuda9.0 (#29997)

* Support pure fp16 training for AMP API. (#29544)

* add cast ops before and after unsupported fp16 ops.

* Keep partial net in FP32 pattern.

* Support check_finite_and_unscale and update_loss_scaling for FP16 calculation mode.

* Add fp16 support for adam op.

* add multi precision attr for adam.

* Fix the bug of test_multi_precision_fp16_train UT.

* Code format for CI.

* Fix the redefine error about MPTypeTrait on windows.

* fix bugs of the _create_accumulators func in Momentum.

* fix bug when inserting post cast op.

* Add the update_loss_scaling op in allow_set of UnusedVarCheck.

* Update for ci coverage.

* Add some doc for OptimizerWithMixedPrecision.

* Fix the code style.

* Imporve the doc of `amp_init`.

* Change for fp16 testing if users have the infer program defined in separate way.

* Remove tensor copy in the update_loss_scaling op. (#29426)

* remove tensor copy in the update_loss_scaling op

* not use thrust.

* fix some cuda memory access error.

* improve dropout (#29465)

* improve drop out

* add VectorizedRandomGeneratorWithGenerator

* fix bug

* modify according to comments

* improve dropout grad (#29605)

* improve grad perf

* fix the bug of dropout_grad (#29813)

* Softmax vectorization (#29404)

* vec softmax fw

* vec softmax bw

* add a message argument for compiler compatibility

* optimize softmax forward (#30217)

* optimize softmax forward
Co-authored-by: Nzlsh80826 <zlsh80826@gmail.com>

* Optimization grad merge performance (#29784)

* [fleet] combine amp and gradient merge, test=develop (#30086)

* fix assign_op_xpu concat_op_xpu warining (#30120)
Co-authored-by: Nliuyuhui <liuyuhui@baidu.com>

* add aarch64 and sunway kunlun lib

* minor

* optimize elementwise_add for kunlun

* update kunlun dependence

* minor

* minor

[cherry-pick 2.0] Fix bug: In dynamic mode, if start or end is negetive, __getitem__  return wrong result(#30003) (#30146)

1. when slice_item is a slice:
 1) the start of __getitem__ should be std::max(start, 0) if slice
 2) the start of __getitem__ should be std::min(end, dim)
2. when slice_item is an integer, it should be in [-dim_len, dim_len)
3. Fix error message to use accurate data

1. Type of index: int, slice(step must be 1).

2. Type of value:
 (1) int32, int64, float32, bool;
 (2) numpy.array(int32, int64, float32, bool);<Note: float64 is not supported>
 (3) paddle.Tensor(int32, int64, float32, float64, bool);

* Add Lookahead and ModelAverage Optimizer (#30004)

* test=develop, add model_average and lookahead

* Improve Index select cuda kernel (#30139)

* test=develop, add index_select_cuda kernel

* Improve performance of elementwise_add grad op (#29187)

* pass stop_gradient for cast op

* improve performance of elementwise_add grad

* use tensor copy async

* dygraph branch

* fix dygraph branch

* add ut

* make gelu fp16 computing more robust (#29484)

* Add fast path for dropout when p == 0  (#29553)

* add fast path for p == 0 in dropout

* add ut

* Layer norm fp16 (#29169)

* add fp16 for layer_norm op

* revert layernorm api

* fix forward

* fix forward

* fix backward for layernorm with fp16

* fix unit test for layernorm with fp16

* fix with_mkldnn compile error for layernorm with fp16

* 1. revert to PADDLE_ENFORCE_NOT_NULL, 2. change static_cast<float> to static_cast<U>

* fix with_mkldnn compile error for layernorm with fp16

* fix with_mkldnn compile error for layernorm with fp16
Co-authored-by: Nzhiqiu <chenqiuliang@baidu.com>

* fix layer_norm accuracy (#29434)

* Layernorm opt (#29522)

* layernorm fw opt

* layernorm bw opt

* fix typo, test=develop

* remove const dim3 for windows CI compatibility

* merge develop
Co-authored-by: Nzlsh80826 <zlsh80826@gmail.com>

* Fix compile problem when cuda_arch < 6000 (#29576)

* fix compile problem when cuda_arch < 6000

* refine code

* refine code
Co-authored-by: Nzhiqiu <chenqiuliang@baidu.com>
Co-authored-by: Nzlsh80826 <zlsh80826@gmail.com>

* [Complex] Add support for complex grad accumulated (#29889)

* add support for complex grad accumulated

* add unittest for coverage

* update test dtype

* remove useless blank line

* [Complex] Handle complex to real after type promotion (#29855)

* try to add fwd op input dtypes

* refactor base impl

* return tmp_ins after dygraph prepare data

* fix typo found in debug

* polish comment & add complex net test

* revert detail change

* fix unittest failed

* add complex kernel condition control

* fix xpu test failed & polish comment

* polish details by review comments

* Complex op test (#29753)

* delete no need to calculate inputs in dygraph op_test

* delete no need to calculate inputs in dygraph op_test

* change grad elementwise_mul for complex types (#29757)

* add conj op for complex types

* add conj for complex types

* add more test case

* add conj_op test

* modify conj api and impl

* add complex type for fill_constant_op xpu

* add setConstant for complex type

* remove complex conj test file

* user define grad for test_conj_op

* add test case for static mode of conj api

* modify conj doc

* change input args name to x

* remove useless codes

* conj support real types

* add conj test case for real number

* delete no need to calculate inputs in dygraph op_test

* delete no need to calculate inputs in dygraph op_test

* modify grad of mul for complex types

* fix the grads of inputs args order not match bug

* change the grad of div when complex types (#29804)

* change the grad of div when complex types

* fix the grads of inputs args order not match bug
Co-authored-by: Nchentianyu03 <chentianyu03@baidu.com>

* cherry pick heter ps

* 　CMakeList

* support some shape in matmul and cast

* modify matmul

* feat: support check_nan_inf for kunlun device

* support kunlun stack

* minor

* add ps table (#29463)

* add ps table

Change-Id: I468a04bd071d21ff52654926fcf4d5f3da19e178

* add service (#29560)

* add service, remove ut on mac

* fix heter_profiler & add heter stop method

* fix code style

* merge pscore

Change-Id: Ie7f60d1cdde6755a0c29db26863c6283e9843d57

* fix cmake

Change-Id: I6773509a7b4ca79139ecc40b7bf3eb318ceff8bb

* fix conflit

Change-Id: I35575be0c96a8520f9d756ea7f1ff0b904a165ba

* fix conflit

Change-Id: Ic926ea0b0d67803226d51241397ba3b510226bfa

* add nearest_interp_v2 on kunlun

* add nearest_interp_v2 on kunlun
Co-authored-by: NTTerror <tangzhiyi11@users.noreply.github.com>

* gen nccl id use socket (#29431)

* fix gen_nccl_id_op_helper compile failed, test=develop (#29614)

* Add complex dtype op (add) test example (#29603)


* add op test case for complex

* polish code details

* add xpu set constant support

* fix argument rror

* remove useless pyc file

* [Complex] Add real & imag op and api for complex tensor (#29672)

* add complex real op & api & unittest

* add imag op & api & unittest

* refactor op impl

* revert simplify writing due to complile failed

* polish details

* polish grad op code

* add conj op for complex types (#29527)

* add conj op for complex types

* add conj for complex types

* add more test case

* add conj_op test

* modify conj api and impl

* add complex type for fill_constant_op xpu

* add setConstant for complex type

* remove complex conj test file

* user define grad for test_conj_op

* add test case for static mode of conj api

* modify conj doc

* change input args name to x

* remove useless codes

* conj support real types

* add conj test case for real number
Co-authored-by: Nchentianyu03 <chentianyu03@baidu.com>

* Fix the dowanload bug in the case of multiple machines (#29551)

* fix the dowanload bug
* add sort for ips

* Fix bug of matmul_v2 for broadcast case (#29599)

* fix bug of matmul_v2 for broadcast

* Rebuild group automatically in dynamic graph distributed (#29255)

* add tensor_indices in AssignGroupBySize

* add rebuild group in reducer

* fix error message of gather nd (#29521)

 fix #27935 (comment) by QA @OliverLPH (Could you add some MKLDNN-related print log when use FLAGS_use_mkldnn?)

* fix expand && concat/transpose to new api

* update xpu_header

* update activation op on kunlun

* update activation op on kunlun

* update activation op on kunlun

* update activation op on kunlun

* update activation op on kunlun

* add nearest_interp on kunlun

* update error message

* support roi_align & affine_channel for kunlun

* minor

* support mobilenet for kunlun (#29458)

* add xpu ops for training transformer in kunlun (#29539)

* 1.fix matmul bug 2. add one hot

* add xpu error msg
Co-authored-by: Nprocr <procrboo@gmail.com>
Co-authored-by: Ntaixiurong <taixiurong@126.com>

* add deformable_conv op on xpu (#29234)

* rebase develop

* update deformable_conv op on xpu

* update deformable_conv op on xpu

* update kunlun conv2d/softmax/elementwise implemetation (#29229)

* update conv2d & softmax to new xpu api
* test=kunlun

* remove useless comments
* test=kunlun

* remote softmax xpu op
* test=kunlun

* update kunlun softmax
* test=kunlun

* update xpu unitest
* test=kunlun

* fix elementwise_grad bug for kunlun
*test=kunlun

* support global pooling for kunlun (#29293)

* test=kunlun

* update reduce_sum op on xpu (#29367)

* update reduce_sum op on xpu

* update reduce_sum op on xpu

* support running on xpu

* fix expand/uniform_random && concat/transpose to new api on xpu (#29280)

* fix expand && concat/transpose to new api

* update uniform_random_op

* update xpu_header

* 1. fix elementwise ops'bug 2. fix softmax_with_cross_entropy_op 3. add biliner_interp_op (#29448)
Co-authored-by: Nroot <root@bjhw-sys-rpm0223.bjhw.baidu.com>
Co-authored-by: N卖鱼的哲学 <tangzhiyi11@users.noreply.github.com>
Co-authored-by: NQingshuChen <qingshu.chen714@gmail.com>
Co-authored-by: Ntaixiurong <taixiurong@126.com>
Co-authored-by: Nroot <root@bjhw-sys-rpm0223.bjhw.baidu.com>

* fix gpu emb out of range

Change-Id: I5794ac73bd634d5ea069a6fbbd914274b6d6b7bf

* fix doc

Change-Id: I5a3350b2930a9ab2f52116c192b087307faf8fdf

* fix random failed of complex matmul

* Make transpose, trace, kron, reshape, sum op support complex type (#29321)

* 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

* kron, reshape, transpose support complex types

* sum and trace op support complex types

* add test case of sum and trace op

* fix the bug of imag part of complex not initialized

* format file

* format code style

* kron support type promotion; modify test cases

* fix tensorrt output shape error

* fix unittest tensorrt_engine_op_test

* fix code style for unitest

* 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

* 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.

* 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