* Fix hAPI bug of not compatible with LayerHook

https://github.com/PaddlePaddle/Paddle/issues/47000

* Fix hAPI bug of not compatible with LayerHook

* Allow to specify train_bs and eval_bs separately in hapi.fit()

* Update model.py

* Update Model.py

* Update test_model.py

* update model.py

* clip extra and intermediate output of op

* fix bug

* fix bug

* polich code

* polich log

* remove unstack in nn.py under fluid

* remove unstack under fluid

* clean fluid elementwise_pow, remove API

* clean elem_pow doc

* clean elementwise_mod

* clean elementwise min, floordiv, mod

* [NPU] add _npu_identity op and api, test=develop

* fix doc

* address comments

* remove swish in nn.py under fluid

* fix tswish test case

* int scale

* round

* revert commit

* standard api

* fix xpu bugs

* add_cast_bool

* cast

Support bfloat16 for adamw and adam optimizer. Fit the lr for pure bf16 training with tensor fusion. (#48041)

* add bfloat16 for adamw

* set lr not to bfloat16 for pure bf16 training

* update the logic

* update the adamw optimizer

* support bfloat for adam

* add vectorized bfloat16 atomicAdd

* fix compile error

* fix compile error again

* fix V100 compile error

* fix V100 compile again

* clean fluid elementwise_min

* fix elementwise_min op testcase

* add_fill_any_like

* add_fill_any_like

* elementwise_op

* add teller

* modify ut

* comments

* modify ut

* return

* modify

* [remove fluid] under fleet meta_optimizers

* [remove fluid] under fleet meta_optimizers

* [remove fluid] under fleet meta_optimizers

* [remove fluid] under fleet meta_optimizers

* [remove fluid] under fleet meta_optimizers

* [remove fluid] under fleet meta_optimizers

* [remove fluid] under fleet meta_optimizers

* [remove fluid] under fleet meta_optimizers

* [remove fluid] under fleet meta_optimizers

* [remove fluid] under fleet meta_optimizers

* [remove fluid] under fleet meta_optimizers

* [remove fluid] under fleet meta_optimizers

* remove chunk_eval

* simplify depthwise_conv2d phi kernel selection

* fix depthwise_conv2d

* Enable bf16 in oneDNN bilinear_interp kernel

* Fix bilinear_interp_v2 not enabled in models

* Remove unnecessary checks

* add stat tool

* add roll and roll_grad kernels and strided_slice and strided_slice_grad kernels, test=kunlun

* embedding and embedding_grad add int32 input, test=kunlun

* refactor: update pg custom

* fix: use new api in ut

* fix: typo

* revert: recover legacy apis

* fix: add GetDeviceContext

* fix some docs bugs; test=document_fix

* Update batch_sampler.py

* Update dataset.py

* Update dataset.py

* Update sampler.py

* for codestyle; test=document_fix

* fix copy-from issue; test=document_fix
Co-authored-by: NLigoml <39876205+Ligoml@users.noreply.github.com>
Co-authored-by: Ligoml <limengliu@tiaozhan.com>