* fix record event for operator type in new dygraph (#44582)

* fix new dygraph record event for op

* update unit test

* fix file mode

修复poisson op单测随机挂

原因：由于随机OP的无法直接验证数值正确性，该单测随机采样100万个样本，统计落到直方图各区间的数量，计算出粗略的概率密度函数，与标准概率密度函数对比，这种测试方式会有一定误差。
当采样数量越小，误差越大，因此该PR增大采样样本数量（100万->200万），误差进一步减小在rtol范围内。

* [Cherry-pick] Apply IOU to test_parallel_executor_seresnext_base_gpu (#43812)
1. Fix the conflict between #43812 and current release/2.3 branch
2. test_parallel_executor_seresnext_base_gpu failed on 2 P100 GPUs with `470.82` driver.

* fix emb pass for ernie3.0

* fix emb pass for ernie3.0

* fix emb pass for ernie3.0

Cherry-pick of #43536

Backgroud in #43262

In elementwise_div UT, the numeric gradient (validation) has large relative error in comparison to analytic gradient (Paddle OP).

    The default rtol for UTs is 0.005
    The rtol for float32 and float64 elementwise_div OP is set to be 0.05
    The rtol for float16 and bfloat16 elementwise_div OP is set to be 1.0

The relative error is too large, so this PR provides user defined gradients to test elementwise_div followed by the analytic method.

* update quantization clip and round

* fix quantization clip and round Attribute

* fix typo

* Create Tensor by paddle::empty  in custom operator (#41840)

* create tensor by empty in custom op

* fix some bug

* update relu custom op demo (#43173)

* Fix incompatible error for custom op Placetype (#43749)

* fix incompatible error

* rmeove default constructor

* add macro

* fix cpu make error

* add DefaultGPUPlace api
Co-authored-by: Nzyfncg <zhangyunfei07@baidu.com>

* Use all sitepackages path as the library/include path (#42940)

* Fix several unit tests and increase the unit tests stability (#43670)

* Reduce gather op unit tests size and increase the timeout

* Add NVIDIA_TF32_OVERRIDE for multi-processes environment

* Remove record test for device event ut

* Fix 3 unittest errors (#43532)

* Fix test_fuse_resnet_unit failure

* Fix test_imperative_auto_mixed_precision failure

* Fix sparse_attention_op error

* Fix sparse_attention_op error

* Use fixed random seed (#43659)

* for CI test_collective_sendrecv_api
Co-authored-by: Nzlsh80826 <rewang@nvidia.com>
Co-authored-by: NShijie <505749828@qq.com>

* fix structure infos conflict in static return_list mode. test=develop

* fix format. test=develop

* fix format. test=develop

* add bilinear_interp_v2 converter

* update op_teller.cc

* add unittest for bilinear_interp_v2 converter

* code format

* bug fix

* code format and add unittest

* remove merged modify in op_teller.cc

* code format

* code format

* fix scale init error

* merge_release_and_dev

* merge_release_dev

* update

* Use tempfile to place the temporary files (#43237)

* tempfile_fix

* update

* fix_CI

* update_word2vec.inference.model

* remove_change_in_word2vec_book

* fix_word2vec_book

* rm_affine

* update

 fix the bug that _DataLoaderIterMultiProcess use time to generate the seed

cherry-pick #43318

在 amp-o2功能开发过程中，为了支持指定网络存储数据类型的功能，添加state_dict hook功能，但是在Layer的set_state_dict是通过state_dict获取网络参数并加载的，hook接口的存在导致 set_state_dict无法加载到原本网络参数。
本pr通过增加hook控制开关，在set_state_dict中禁用hook解决该问题。

详见pr43407

bug：
当class Layer的_buffers中有参数为None的时候，调用to()方法将会报layer to 'NoneType' object has no attribute 'place'的错误。
修复方法：
to()方法增加对_buffers中None类型参数的判断，如果为None，跳过该参数的处理。

* cherry pick #43088 #40664

* fix clang format

* modify xpu.cmake,*test=kunlun (#41832)

* modify xpu.cmake,*test=kunlun

* modify xpu.cmake,*test=kunlun

* modify xpu.cmake,*test=kunlun

* modify xpu.cmake,*test=kunlun

* support bilstm,*test=kunlun

* [cherry-pick]support multi_layer of bilstm,*test=kunlun

* [cherry-pick]refactor sum unit test,*test=kunlun (#43561)

* cherry pick from #43397

* fix code

* place all save/load path into temporary directory

* rm no need unittest

* Rename dropout is test (#43098)

* replace dropout_is_test with is_test.
* improve atol on a100.

* fused_attention fused_feedforward api support Model Tensor Parallel (#42985)

* fix is_test bug in fused_feedforward. (#43508)
Co-authored-by: NLi Min <11663212+limin2021@users.noreply.github.com>

    Use tempfile for unit test & custom op test to replace temporary files to ensure that all temporary files will be deleted normally after a single measurement, avoiding the usage of disk files.
    The PR only involves single-test and op test modifications and does not affect existing functionality.
    Release/2.3 branch modified in PR43521;

* fix unit test temp file, test=develop (#43155)

* add cleanup code, test=develop (#43305)

* Fix numpy 1.20+ deprecation warnings (#42929)

* Replace np.bool/np.bool8 with np.bool_

* Replace np.object with np.object_

* Replace np.complex with np.complex128

* Replace np.float with np.float64

* Replace np.int with np.int_

* Rerun pre-commit for newer pre-commit configuration

* Use builtin bool instead of np.bool_ based on the context

* fix mode dtype
Co-authored-by: Nzlsh80826 <rewang@nvidia.com>

* [EinsumOp] Polish forward logic and backward logic for optimize (#42603)

* change logic for optimize

* modifty

* merge

* change einsum_v2 as default and add new flags: FLAG_einsum_opt=1|0 (#43010)

* [EinsumOp] Make EinsumOp support bfloat16. (#43085)

* change einsum_v2 as default and add new flags: FLAG_einsum_opt=1|0

* make EInsumOP support bf16

* add unittest for BF16

* add condition for test_BF16

* fix bugs

* fix

* change the backward api to fit einsum op

    使用 tempfile 替换临时文件，保证在单测结束后，所有临时文件都会被正常的删除，避免占用磁盘文件。
    此 PR 仅涉及单测修改，不影响现有功能。
    develop 分支修改在 PR 43376

* Fix cond_block_grad error when handle no need grad vras

* Add comment and UT

cherry-pick PR #42777

* [Dy2Stat]Modify all jit.save path into tempfile

* [Dy2Stat]Modify all jit.save path into tempfile

* [MLU] add mlu new profiler (#41138)

* [MLU] add mlu new profiler

* fix format

* [MLU] support add callback to stream (#41831)

* [MLU] add gather mlu kernel (#41969)

* [MLU] add mlu activation kernels (#41751)

    add class NameScopeHelper for adding namescope info
    添加更多 种类优化器状态的映射
    为 IpuStrategy 添加 compilation_progress_logger option 用于输出 编译进度
    部分代码清理和杂项优化

* Reduce time variation for cuda_managed_memory_test (#42458)

* Disable standalone executor for test_tensordot (#42476)

* fix wrong place

* skip bf16 test if not supported (#42503)