* fix fleet api doc

* test=develop, rm pathlib

* add reducer

* refine envent for memorycopy

* add concat&split for allreduce

* apply concat & split for fuse tensor

* fix nccl dep

* fix the untest, compile problem and ddp initialize problem

* fix untest for mac & add some comments & solve the repeated param in sublayers

* fix untest for windows & fix document

* add Inmemorydataset

* add lars to fleet meta optimizer

* add lamb to proto

* add lamb to fleet meta optimizer

* fixed syntax bug

* fixed syntax bug

* fixed syntax error in lamb, add config setter of lamb in distributed_strategy

* trigger unitest to rerun

* add new unitest func for lamb

* revise unitest for lars and lamb

* revise dgc meta unitest

* revise lars document in distribute_strategy

* revise lars lamb document in distributed_strategy.py

* revise lars lamb document in distributed_strategy.py

* add weight decay exclude logic to lars

* restore optimzier.py

* restore optimizer.py as develop except lars

* add epsilon and exclude fn to distributed_sttrategy

* add lars epsilon

* revise unitest for fleet lars and lamb

* revise lars lamb unitest for CI coverage

* revise lars argument api

* revise lars argument api

* revise lars argument api

* revise api doc of lars

* fix op role

* add sharding save and add_sync_comm_for_test function

* add comm_analyse to utlis

* revise sharding_utils

* add sharding saving unittest

* revise sharding utils for unittest

* revise sharding en doc

* update sharding utils api

* add doc for sharding

* fixed bug in sharding var size count

* update varsize count in sharding

* fix sharding num_nccl_comm

* Revert "fix sharding num_nccl_comm"

This reverts commit d51587c15e9323acf226ddd36154275f0d1daf76.

* update, test=develop

* upgrade comment string to raw string

* fix string in

* fix string with ' '

* revert update on comments

* upgrade only necessary

* fix sample code checker

* fix comments with '''

* test=develop, optimize global_step

* update, test=develop

* add lars to fleet meta optimizer

* add lamb to proto

* add lamb to fleet meta optimizer

* fixed syntax bug

* fixed syntax bug

* fixed syntax error in lamb, add config setter of lamb in distributed_strategy

* trigger unitest to rerun

* add new unitest func for lamb

* revise unitest for lars and lamb

* revise dgc meta unitest

* revise lars document in distribute_strategy

* revise lars lamb document in distributed_strategy.py

* revise lars lamb document in distributed_strategy.py

* add weight decay exclude logic to lars

* restore optimzier.py

* restore optimizer.py as develop except lars

* add epsilon and exclude fn to distributed_sttrategy

* add lars epsilon

* revise unitest for fleet lars and lamb

* revise lars lamb unitest for CI coverage

* revise lars argument api

* revise lars argument api

* revise lars argument api

* revise api doc of lars

* fix op role

* add sharding save and add_sync_comm_for_test function

* add comm_analyse to utlis

* revise sharding_utils

* add sharding saving unittest

* revise sharding utils for unittest

* fix fleetrun heter ps on paddlecloud

* add sharding

fleet support paddle.optimzier

* bug fix

* fix fleet_base

* bug fix

* fix coverage

* update, test=develop

* add size method for large scale

* add large scale UT

* add ut for checkpoint

* test=develop, fix geo sgd communicator

* test=develop, gloo_init_method

* test=develop, bug fix for gloo http_init

* raise error if use multi-cards in fleet non_distributed mode; test=develop

* add sparse tensor load method

* test=develop, bug fix for parameter_recv
* test=develop, for unittest, test_fleet_rolemaker_new

* support gradient merge with recompute, test=develop
test=develop

* refine fleetrun.ps_launch

* update fleet run for multi device support

* ps_graph support ps-gpu

* fix heter save

* add heter save unittest

* fix unittest & simple code

* update fleetrun

* fix fleetrun

* fix launch barrier

* fix role maker

* add paddlecloud rolemaker unittest

* rename heter_worker_device_guard

* fleet support non_distributed training in dygraph mode; test=develop

* add double grad for expand, test=develop

* add gloo initializer, test=develop

* set default value to strategy in distributed_optimizer test=develop

* test=develop, rm netifaces