* add pad func

* add pad

* test=develop, add pad op and apis

* restore pad2d

* test=develop, fix paddl declare

* fix pad interface

* test=develop, fix pad

* test=develop, add all pad api and cos_sim

* test=develop, remove padding default value

* test=develop, rename var to tensor

* test=develop, add more tests

* test=develop, rename tovar to totensor

* test=develop, fix init

* test=develop, add more test

* test=develop, add more tests

* reduce the execution time of affine_channel op unittests

* support ps-gpu

* add zero norm, inf norm support for p_norm op

* fix the invalid argument check, fix the dtype problem in test case.

* 1.Fix api sum:(1) input->sum; (2)dim->axis; (3)keep_dim->keepdim.

* 2. fix bug when len(axis) == len(x.shape). 

* add Sampler/SequenceSampler/RandomSampler. test=develop

* add SyncBatchNorm,test=develop

* logsigmoid and LogSigmoid, test=develop

* add elu gelu relu, test=develop

* update to_variable to to_tensor, test=develop

* address review comments, test=develop

* address review comments, test=develop

* change to_variable to to_tensor in test, test=develop

* add check if fluid.data() variable no feed data, test=develop

* Add testcase for feed check, test=develop

* Conv2d_transpose and mul support channnelwise quantization, test=develop
* Skip collecting out threshold for output tensor of which the type is not fp32 or fp64, test=develop
* Fix error in test_user_defined_quantization, test=develop
* Add depthwise_conv_bn_fuse, test=develop
* Add conv_transpose_bn_fuse_pass for post_training_quant, test=develop

align expand_as op to expand.

* add set/get default dtype

* enable mkldnn layout conversion

* review fix: remove tmp_place

* fix test mkldnn swish

* add UT for PrepareData CPU->MKLDNN

* add #ifdef PADDLE_WITH_MKLDNN

* Force-push commit
Co-authored-by: Ngrygielski <adam.grygielski@gmail.com>

add nn.functional.normalize API for API2.0, fix L1Loss API typo, test=develop

* add feature to fleet2.0 role_maker, distribute_strategy, test=develop

* set default value to strategy in distributed_optimizer test=develop

set default python3,fix incompatible,cache dir for third party,unify error code,for windows (#26178)

* set default python3 for paddle windows,test=win

* set default python3,cache dir for third party,error code,test=win

* fix some incompatible

* fix some error

* set virtual environment,test=win

* add nll loss API, update demo code of the comment

Update the sigmoid, Sigmoid layer for the api2.0

* Update `paddle.add`

remove `alpha`
update doc sample code
remove `:alias` etc

* Remove `imperative` reference from docstring

* add children and named_children

* adaptive expand op to 2.0 (align to torch.expand) , test=develop

* expose and unify the Tensor concepts to the user

* expose tensor to user

* add copy place for Tensor

* add copy place for Tensor

* add note

* add macro PADDLE_WITH_CUDA

* remove RUN_TYPE=DIST

* fix some error

* add tile_op, test=develop

* add Class KLDivLoss and function kl_div

* test=develop, bug fix for test_dist_fleet_ctr

* add paddle.nn.functional.mse_loss

* add name

* fix conflict

* Upgrade `no_grad` decorator

test=develop

- match torch decorator usage (i.e., with parenthesis)
- handle generator functions
- add `paddle.no_grad` alias

* Switch from `functools` to `decorator`

preserves signature

* Reword decorator usage note

* add unittest for paddlerolemaker with gloo