* Adapt tensor axis for argmin/max

* Add UT

* Polish UT

* move layer_norm xpu kernel to phi, test=kunlun

* fix, test=kunlun

* [OpAttr]Adapt tensor axis for reduce_min/max/mean/sum/prod

* [OpAtttr]Add attribute var interface for Operator class

* fix unittest

* fix unittest

* add runtime config in phi

* add runtime attr for op desc and op

* fix no proto error

* adjust opdesc set_attr impl

* try to remove conv_op extra attrs

* add init runtime attr map

* change extra header path

* fix runtime_attr

* fix trace_op

* fix bug of pass

* fix merge conflict

* fix dygraph attrs

* fix bug of pass

* fix dygraph bug

* fix unittest module

* delete extra attr default

* fix dropout kernel

* polish code

* fix extra output of instance_norm

* fix merge confilct

* fix op_desc bug

* add extra attr in yaml for conv3d_transpose

* don't remove extra input and output

* fix save_inference_model

* fix bug of batch_norm

* revert some change

* polish log

* polish code

* add code comment
Co-authored-by: NChen Weihang <chenweihang@baidu.com>

fix missing keep_dim variable

    fix missing grad check in unittest

    add new test case

* [OpAttr]padding_value of Pad support Tensor type

* fix unittest

* fix unittest

* fix coverage

add constant folding pass， for some model，it will get less latency；

* fix memcpy_h2d bug related to cuda stream setting when allocate memory

* add header file

* fix compile error for cpu only

* move xpu kernel to phi

* delete fluid file

* fix compile

* add guard, test=kunlun

* xpu set constant

* fix xpu error, test=kunlun

* add interpretercore

* refine backward program id

* add code

* refine program

* refine code

* create forward/backward_program by prog2graph2prog method

* test, do not care

* refine code

* refine code

* refine code

* test, do not care

* add interpretorcore

* add scope

* refine scope create method

* add jit for new_exe

* solve conflict

* delete unused code

* polish code

* polish code

* refine scope in inplace

* refine for datatransfer

* refine _rebuild_from_desc

* refine control eager deletion attr

* refine used_for_jit

* refine jit for infer

* op size0 use ori program

* polish code

* refine jit

* refine run_program_op ut

* refine inplace

* refine control

* refine graph helper

* refine control

* refine inplace

* refine buffer_share_inplace_pass

* polish code

* polish code

* refine usage for compilerProgram

* refine control

* test

* test core cache

* refine code

* refine io.py

* increase test_seq2seq timeout

* refine convert program

* refine interpretercore_cache release

* delete buildinplace

* refine partial_program && io

* refine code for io

* test

* test

* test

* [MLU] fix compile error, test=develop

* fix more compile error, test=develop

* mv elementwise add to xpu , test=kunlun

* fix ci bugs, test=kunlun

* fix ci bugs , test=kunlun

* abs relu6 fwd

* abs bwd

* gaussian_random_kernel and mkldnn-onednn renaming

* scale kernel

* whitespace

* whitespace

* revert scale migration

* whitespaces

* revert changes to gaussian kernel

* whitespaces

* [XPU] migrate mul to phi;test=kunlun

* rm fluid mul xpu op;test=kunlun

* migrate assign xpu kernel, test=kunlun

* remove assign_value xpu, test=kunlun

* update PixelShuffle and PRelu api in vision.py and activation.py

* update activation.py, using paddle replace numpy

* update format issue

* update foramt issue

* support refuse forward dygraph

* modify backward api exponential__grad yaml

* remove print code

* 当反向复用前向时进行需不需要更高阶的反向判断，如果不需要调用c++ api，需要的话则调用前向动态图生成反向节点

* fix some backward bugs

* modify the generated dygraph function name

* fix bug in PostTrainingQuantizationProgram for not full quantization case

* fix coverage ci

* gather gather_grad gather_nd gaussian_random xpu to phi

* fix hard_swish inf

* skip_check_grad for mkldnn op

* 'fix code style'

* fix unittest

* pylayer set has grad with create_graph

* [OpAttr]num_rows/num_colums of eye support Tensor type

* fix attr cast with long type

* move incubate to geometric

* add paddle.geometric

* fix unittest bug

* add float16 support for segment op

* change reindex and sample neighbors flag name

* add heter graph reindex

* move sample_neighbors.py to neighbors.py

* delete khop_sampler in geometric

* delete unused code

* change sample_neighbors api input order

* fix en doc

* fix unittest

* fix unittest

* change reindex

* fix division by 0

* delete unnecessary input argument

* delete final_state

* unify the size and size() by jst

* fix bugs

* bug fix.

* fix bugs

* change all_close -> np.testing.assert_allclose

* support depthwise_conv2d ops
Co-authored-by: NZhixin Yao <zhixiny@graphcore.ai>
Co-authored-by: NZhaorui Chen <zhaoruic@graphcore.ai>

* fix duplicate name
Co-authored-by: NZhixin Yao <zhixiny@graphcore.ai>
Co-authored-by: NZhaorui Chen <zhaoruic@graphcore.ai>

* add legacy_api.yaml

* set merged_momentum inplace only

* support inplace optional<vector<tensor>>

* add dygraph_mode api

* add merged_adam yaml

* add merged_adam python api

* change testcase of merged_adam and adam

* fix import of test_merged_adam_op