* Debug

* Fixed issue with reset_grad_inplace_version when used with clear_gradient & cross-batch accumulation

* Rearranged interfaces

* Fixed ci issues

* refine structure for cuda and rocm

* update

* update

* update

* update

* optimizer __call__ to make dygraph faster

* fix return type

* remove cpp Layer

reset_inplace_version removes all inplace related records to VarBase/VariableWrapper, the essential purpose of which is to let you use inplace operations as if using its non-inplaced version, which of course will cause unexpected consequences if not used with care.

This is essentially a hack interface to satisfy one specific request

* Added HCCL backend support in dynamic graph mode

* fix segmentation fault

* add ut

* Removed one ENFORCE statement

* Changed func name to _share_buffer_to

* Improve error reporting information

* Updated the logic of _is_share_buffer_to func

* Add default arg to enhance varbase ClearGradient func

* Removed default arg, use a Flag to enhance varbase ClearGradient func

* Renamed Flags to FLAGS_real_release

* Use default arg to enhance varbase ClearGradient func and expose two func to set/get gradient isEmpty

* Removed DECLARE_bool statement

* Polished Code

* Expose func for varbase

* Expose func for varbase and enhance varbase init func

* Change func name and add test case for _CopyGradientWith

* Rename func

* Add test cases to increase coverage

* Refine the logic of _to func

* Replace numel() with _numel(), Add test code

Add Tokenizer related functionalities for Transformer model in order that the process of training and predicting is consistent.

* support the text string as an input Tensor
* support the "VOCAB"unordered_map<wstring, int> as an input Tensor to lookup tokens
* Tokenizer used for BERT. This tokenizer applies an end-to-end, text string to wordpiece tokenization.
* It first applies basic tokenization, followed by wordpiece tokenization.

* fix async_read bug

* change index place to cpu

* add tensor size judge

* add async_read & async_write test

* fix bug in async_write

* fix mac py3 ci

* fix bug for cpu version paddle

* fix windows ci bug

* change input argument error type

* change const_cast to mutable_data

* add async_write out-of-bound check and consumate error hint

* fix a small bug for dst_tensor

* add docs and refine codes

* refine docs

* notest,test=windows_ci

* fix windows ci

* fix require

* fix code-block

* add core.is_compiled_with_cuda()

* refine amp level

* fix typo

* update tracer._amp_level

* add pure fp16 major function in auto_cast & tracer

* support master weight in dygraph for pure fp16

* check mix dtype of fp16&fp32 for check_finite_and_unscale op

* change pure fp16 funtion name

* refine some bug in auto_cast

* refine auto_cast interface logic

* add param _casted_by_pure_fp16 for class Layer

* support state_dict hook for save model by user appointed dtype in pure_fp16_decorator

* refine pure_fp16_decorator as decorator

* add unittest

* add comment

* add comment

* support recompute

* add comment for auto_cast and decorator

* support to_static_state_dict for paddle.jit.save

* unlimite models num and optimizers num

* add lookup_table in black_list

* fix momentum and layer state_dict

* fix bug in layer state_dict

* fix bug in layer state_dict_helper

* refine unittest

* refine test_momentun_op

* refine interface and some code

* refine amp_decorator interface

* refine pure fp16 interface

* refine master weight interface

* Change the invoking method of settiem from numpy to set_value op when value is not tensor

* fix the check logic for inplace in setitem

* fix the unittest problem caused by setitem doesn't support fp16

* modify some code format in setitem

* change metaclass of Layer from pybind11_builtins.pybind11_type to type

* fix cast

* add ut

* can pass the fake test

* add files

* modify cmake to pass windows-ci

* for ci pass

* WITH_GLOO=ON

* for pass coverage test

* add cpuonly testcase

* add

* disable nccl when compile with cuda

* change python version in cpuonly

* add backend argument

* add required gpu

* add required:gpu

* support numpy dtype and polish code of list index.

* polish code.

* Support getitem by Bool index

* delete some debug info of bool index

* support the case that the shape of bool index is different from indexed tensor

* support setitem by bool index

* add the unittest for throwing exception

* merge conflict

* add check for int tensor when index is bool

* polish code

* polish code.

* polish code.

* polish code.

* polish code.

* fix index tensor leak in __setitem__

* fix another usage of PyTuple_Pack

* refine code

* refine code

* handle None index

* add Py_DecRef

* revert ut

* refine code

* merge develop

* use RAII

* follow comments

* Support getitem by Bool index

* delete some debug info of bool index

* support the case that the shape of bool index is different from indexed tensor

Change the invoking method of settiem by Ellipsis and None index from numpy to set_value op (#34911)

* Change invoking mathod of the settiem by Ellipsis and None index from numpy to set_value op

* add none_axes into attr of set_value_op in dygraph mode

* Support getitem by ellipsis index in dynamic mode

* change some code style

* Support getitem by none index in dynamic mode

* modify a comments style and remove useless code

* Support getitem by ellipsis index in dynamic mode

* change some code style

* remove check on kwargs

* refine code, reuse commom function

* in NPU environment, use CPUPlace for missing operators.

* in NPU environment, use CPUPlace for missing operators.

* fix TensorCopy bug and add unit test.

* fix code style.

* add more unit tests.

* use PYTHON_C_API in dygraph, test=develop

* add varbasecopy func to fix the paraBase type bug in layers.to API

* overload _copy_to func for ParamBase

* add xpuplace

* add waiting varbsecopy completion when not blocking

* fix dst_device bug

* modify varbase to shared_ptr