* # This is a combination of 10 commits.
# The first commit's message is:
add expert count op

add ut for expert_count

# This is the 2nd commit message:

update UT only for cuda

# This is the 3rd commit message:

fix for rocm

# This is the 4th commit message:

update ut

# This is the 5th commit message:

add moe module

# This is the 6th commit message:

add expert count op

add ut for expert_count

# This is the 7th commit message:

update UT only for cuda

# This is the 8th commit message:

update ut

# This is the 9th commit message:

add moe module

# This is the 10th commit message:

make expert count private

* add assign pos op

* fix upper num name

* add api _assign pos

* add ut for assign pos op

* update date

* fix for win

* update for test (timeout)

* fix ut

* update

* fix ut for number count
Co-authored-by: Nhlygit66666 <2570058140@qq.com>

* test=document_fix , fix launch doc

* test=document_fix , fix typo

* [Auto Parallel] gradient merge pass support dist attribute

* refactor cost model

* fix benchmark and communicator config

* fix bugs of the_one_ps

* multi program and fix bug in optimizer

* multi program in the_one_ps

* public commcontext

* ps optimizer multi programs

* cvm & datanorm backend

* fix dim

* fix unittest

* fix

* the one ps merge

* remove comm

* add DownpourLiteWorker

* all

* fix

* fix

* device worker downpour lite

* fix

* fix bug in global shuffle

* save inference model

* fix & add log

* fix

* remove log

* fix

* fix save summary

* fix

* fix pscore

* fix

* fix

* fix

* fix

* fix

* remove logs

* fix

* fix

* fix

* fix

* fix

* add some comments

* fix
Co-authored-by: Nesythan <esythan@126.com>

* add dist_saver and update engine

* add dist_saver and update engine

* [Auto Parallel] Add distributed mul for the old version

* [Auto Parallel] Support the auto completion of while_op

* [Auto Parallel] Improve the completion algorithms

* [Auto Parallel] Fix bugs for ernie inference

* [Auto Parallel] Remove attrs which cannot be pickled

* [Auto Parallel] make the dims_mappings of LodTensorArray vars empty

* [Auto Parallel] Fix bugs for the ernie inference in the pipeline parallel

* [Auto Parallel] Remove unncessary comments

* [Auto Parallel] Fix a bug of the CMakeLists

* [Auto Parallel] Use the newest APIs to write the unit test

* [Auto Parallel] Remove unnecessary statements

* add expert count op

add ut for expert_count

* update UT only for cuda

* fix for rocm

* update ut

* add moe module

* add expert count op

add ut for expert_count

* update UT only for cuda

* update ut

* add moe module

* make expert count private

* rename expert count op
Co-authored-by: Nhlygit66666 <2570058140@qq.com>

Add the recorder

* [Auto Parallel] Redesign the tunner for Auto Parallel

* [AutoParallel] Converter
Converter API

* update reshard for while sub block

* fix code format error

* [MLU] add fleet init api and collective api pytest for mlu

* fix no value for argument 'data_type' in method call

* engine support pp

* fix format

* avoid multi print

* fix convert

* bug fix

* add pp unittest

* fix benchmark and communicator config

* fix bugs of the_one_ps

* multi program and fix bug in optimizer

* multi program in the_one_ps

* public commcontext

* ps optimizer multi programs

* the one ps merge

* fix bug in test

* delete gloo connect retry

* the_one_ps dirs reconstruct

* .

* .

* create the_one_ps dirs

* create the_one_ps dirs

* create the_one_ps dirs

* create the_one_ps dirs

* create the_one_ps dirs

* create the_one_ps dirs

* the one ps dirs modify

* the one ps dirs modify

* the one ps dirs modify

* the one ps dirs modify

* refactor ps optimize

* refactor ps optimize

* refactor ps optimize

* .

* .

* .

* .

* .

* .

* refactor theoneps

* the_one_ps

* add ps pass unittest

* add ps pass unittest

* ps unitest frame

* ps unittest frame

* ps unittest frame

* ps unittest frame

* ps unittest frame

* ps unittest frame

* ps unittest frame

* ps unittest frame

* ps unittest frame

* ps unittest frame

* ps unittest frame

* ps unittest frame

* ps unittest frame

* ps unittest frame

* ps unittest frame

* ps unittest frame

* ps unittest frame

* add cpu_async_ps_mode test

* add cpu_async_ps_mode test

* add cpu_async_ps_mode test

* ps unittest ready

* ps unittest ready

* solve dist_pass init conflict

* solve import CommContext error

* unittest ok

* implement AllocateFrom

* solve setup.py.in conflict

* solve conflict

* solve conflict

* solve conflict

* .

* .

* cpu-async-ps minimize test ok & gpu minimize test ok

* add heter 2stage unittest

* add heter 2stage unittest

* add heter 2stage unittest

* sync/geo test ok & fix heter_worker program ok

* .

* new fleet desc generator

* new fleet_desc builder

* new fleet_desc builder

* .

* .

* correct ps.proto compile

* .
Co-authored-by: Nzkh2016 <zhangkaihuo@baidu.com>

* adapot dist op

* add dist_fill_constant_batch_size_like

* remvoe print

* update compitable

* add unitest

* fix 'invalid escape sequence'

* fix assert error

* add subblock logic for context and partitioner

* partitioner support sub blocks

* revise typos

* fixed param init bug for while

* chmod 644

* add unitest

* mv forward parser

* update unitest

* update dist op ctx

* update dist op ctx

* fixed bug in dist op ctx

* fixed bug for recompute subblock

* fix benchmark and communicator config

* fix bugs of the_one_ps

* multi program and fix bug in optimizer

* multi program in the_one_ps

* public commcontext

* [Auto Parallel] Add the high-level Engine API

* Update the test cmakefile

* support dtype param for auto_cast

* add amp_dtype for tracer

* add unsupported bf16 list

* support bf16 amp for O2

* refine python interface for bfloat16

* refine code

* refine code

* refine unittest

* refine code

* refine code

* add bf16 o1

* refine code by comment

* add gradient accumulator

* add recompute