* new general transformer inference support

* support CUDAGraph for partial graph

* add ut

* fix ci

* fix ut again because of eager mode

* fix kunlun ci

* fix win ci

* polish append op using

* fix var error

* fix group norm impl

* add fused_bias_dropout_residual_ln op and layer.

* support is empty

* fix error

* fix code error

* change to fake empty

* using fake empty first

* using fake empty first

* refactor the optional tensor

* remove optiona<MetaTensor> in InferMeta

* fix bug

* fix optional<vector<Tensor>>

* fix bug

* fix rmsprop

* fix amp of eager_gen

* polish code

* fix deleted code

* fix merge conflict

* polish code

* remove is_nullopt_

* fix merge conflict

* fix merge conflict

* Support memory stats for CPU

* Add UTs

* Fix typos

* Fix typos

* remove storage

* add glog include

* add glog include

* add glog include

* remove is_init_py of RandomGenerator, and use Global Generator if not OP seed

* fix comment

* fix Wtype-limits

* fix

* remove -Wno-error=type-limits

* remove shared_storage

* fix bug

* fix rnn bug

* refine enforce code

* refine enforce code

* fix compile failed

* fix infrt failed

* enable graph-engine to return all id

* change vector's dimension

* change vector's dimension

* enlarge returned ids dimensions

* add actual_val

* change vlog

* fix bug

* bug fix

* bug fix

* fix display test

* singleton of gpu_graph_wrapper

* change sample result's structure to fit training

* recover sample code

* fix

* secondary sample

* add graph partition

* fix pybind

* optimize buffer allocation

* fix node transfer problem

* remove log

* support 32G+ graph on single gpu

* remove logs

* fix

* fix

* fix cpu query

* display info

* remove log

* remove empyt file

* distribute labeled data evenly in graph engine
Co-authored-by: NDesmonDay <908660116@qq.com>

* adapt faster tokenizer op

* add eager test

* add unittest

* [Eager] Fix test_model.py under eager in windows-openblas ci

* Use || in this case

* recover _in_eager_mode_

* optimize cinn find graph by graph address

* graph_key use int64_t instead of program string

* fix framework _to_readable_code python code

* rename get_readable_comile_key to get_serialize_comile_key

* enable graph-engine to return all id

* change vector's dimension

* change vector's dimension

* enlarge returned ids dimensions

* add actual_val

* change vlog

* fix bug

* bug fix

* bug fix

* fix display test

* singleton of gpu_graph_wrapper

* change sample result's structure to fit training

* recover sample code

* fix

* secondary sample

* add graph partition

* fix pybind

* optimize buffer allocation

* fix node transfer problem

* remove log

* support 32G+ graph on single gpu

* remove logs

* fix

* fix

* fix cpu query

* display info

* remove log

* remove empyt file
Co-authored-by: NDesmonDay <908660116@qq.com>

* update

* [IPU] fix ipu and add python infer api, test=develop

* [IPU] add paddlepaddle-ipu package name, test=develop

* update to popart v2.5.0

* use a specific version of sdk2.5.0

* [Eager]Fix tensor.name is empty behavior

* fix unittest

* polish code details

* remove needless prefix

* revert needless change

* polish grad func generated format

* fix pylayer_memleak

* inc ref before return Py_None

* refine

* refine

* refine

* refine

* fix @ stray error in dygraph

* fix @ stray error in dygraph

* merge recent changes

* fix setting pipline

* enable graph-engine to return all id

* change vector's dimension

* change vector's dimension

* enlarge returned ids dimensions

* add actual_val

* change vlog

* fix bug

* bug fix

* bug fix

* fix display test

* singleton of gpu_graph_wrapper

* change sample result's structure to fit training

* recover sample code

* fix

* secondary sample

* add graph partition

* fix pybind
Co-authored-by: NDesmonDay <908660116@qq.com>

* enable graph-engine to return all id

* change vector's dimension

* change vector's dimension

* enlarge returned ids dimensions

* [DO NOT MERGE] test op_test

* update with more related modifications

* split op_test.py to use test=allcases for testing

* split op_test.py to use test=allcases for testing