* set output et

* set output et

* overwrote validate and infer

* Force AutoBroadcast to be specified at the op level since no default is correct for all ops.

* exports

* Added support mode for SpaceToDepth

* Added unit tests

* Fixed styles

* Revert changes in prototxt files

* [SPEC] Add auto_broadcast parameter to SquaredDifference

* Rename set_autobroadcast->set_autob

* [ONNX] Importer should use fused op for MatMul

* Fix a bug in fused matmul op

* Dont reshape matmul inputs to at least 2D any more

* LogSoftmax introduced

* Added LogSoftmax to serializer

* Fixed style

* Fixed CmakeLists style

* code review remarks introduced

* Code review remarks introduced

* Sketch of attribute walker

* Review comments

* merge error?

* Remove unused method

* simplify, make some ser tests work

* Don't look for keys that aren't there

* Factory registry, more ops visited, generic ser/dser start

* More merge

* cleanup

* Adapter for enums

* Compiler error

* Test of user-defined op

* Simplify enum name pairing

* Update distributed.hpp

* Review comments

* compiler error

* Direct access to non-primitive types from adapters

* Define and export type info

* attr enums, AvgPool*, vectors

* Cleanup

* some comments

* Allow type info to be used as a key.

* Don't leave output serialization shapes set.

* Auto adapter

* More ops, adapters

* Missing symbol

* Remove PartialShape and element::Type methods from visitor

* Fix type info

* Remove unused variable

* Simplify

* namespace error

* exports

* Uniform names

* Some better names

* More name cleanup, simplify visitor implementation

* Fix template, add test

* Revert serializer

* Add instantiations

* Work-around gcc issue

* VS exports

* VS exports

* windows export

* vs

* vs

* vs

* vs

* Simplify

* vs

* vs

* Add some missing attributes

* Missing factories

* Merge error

* Fix Add factories

* Missed type

* Round the right bit with denorms

* Rounding to inf

* Added reproducer for issue with broadcast v1

* Make reference broadcast work with V1 broadcast

Comparing size_t >= 0
|= for bools

* Add anchors attribute to RegionYolo operator.

* Make op doc more consistent.

* Make anchors optional attribute.

* Added support for ONNX's clip op in set_11.

* Registered clip in set_11.

* Code formatting.

* Added reshape.

* Changed to auto_braodcast.

* Code refactoring.

* Removed unnecessary import.

* Added constant sharing support

* Added unit tests for specialize_function

* Added parameter description

* Code style fix

*  WIP graph pass to lower ngraph to ngraph dialect

* resolved compiler errors

* - refactor ngraph-dialect to graph pass

* - fix compilation issue
- unit test passes

*  - style fix

* Addressed PR comments

* - move NgDialectConversionPass to anonymous namespace

* - use getModule() to access module inside the graph pass
- address PR comments

*  - fix failing unit test case for negative padding
- make builder as an object isntead of pointer to an object

* Address PR Comments

* Add folder to hold PDPD Fluid helper codes.
Planning to move code from PDPD bridge to this folder as part of code
cleanup restructuring.

* Move folder to src/ngraph/frontend/fluid

* implemented TopK:v1 downgrade

* Using v1 in onnx_importer; styles applied

* Fixed onnx_importer

* code review remarks introduced; fix clang errors

* Fixed problem with incorrect order of TopK outputs

* Code review remarks introduced

* Fixed output_order style

* Autodiff for atan2

* atan2 not supported on PlaidML

* Atan2 op

* TensorIterator

* ssize_t is not on windows

* RNN building test

* simplify

* Simplify output

* typo

* typos

* remove arg

* Sequence version

* style

* Serialization for all but TensorIterator

* Add ops for igpu

* style

* typo, ngpu

* missing headers, output vector

* Fix const json issues

* TensorIterator serialization

* Serialization for TensorIterator
Switch Outout<T> to use shared_ptr do nodes don't vanish
Switch Result to new node style
Add serialization/deserialization to test

* Switch Output to use a shared_ptr to prevent nodes from disappearing early.

* Eliminate wrapped enum
Switch allreduce to new op form

* Convert to new op form

* Disambiguate concat

* Add autobroadcast for SequencePush
Add validation for SequencePush

* compute shapes for SequenceRepeat

* Add explicit conversion from PartialShape to dimension vector
validate and infer types for SliceInput

* validate and infer types for SequenceOutput

* Add sequence attributes

* Move test to serializer so it doesn't fail when there is no serializer?

* const arg

* Beginning of TensorIterator validation

* Validation up to parameters

* Fix shape in test

* Remove mis-typed AxisSet

* Simplify, add doc

* Review comments

* Tweaks

* free/bound

* Try fused op

* Discussion

* more

* comments

* Start of LSTMCell test

* Add LSTMCell example

* Reorg

* Reorg

* Fused ops don't need handlers

* Serialization

* Use `as_type` and `is_type` for up-conversions of descriptions
Allocate output space for each output

* Clean up type checking

* Fix ser/deser issues

* Refactor, cleanup type info to make it safer to use for non-ops

* Implement validate_and_infer_types and modify unit tests.

* For ops in the loop body: revalidate and infer types.

Nested loop is not supported.

* Put body ops in a set and call revalidate and infer types on the set.

* Set slice[axis] to part_size.

Call set_partial_shape to set shape for body parameters.

Add more unit tests.

* Give tensor iterator body a lambda

* Update validate_and_infer_types and unit tests.

* Serialization of body

* Change static function to TensorIterator function.

* review comments

* Fix for openvino ci

* Fix

* Resolve comments

* Revert_changes

* Add upgrade/downgrade pass for Add op.

* Add upgrade/downgrade pass for Divide op.

- Change default value of autobradcasting in v1 into NUMPY.

* Add v1 version for Equal operator.

* Rename helper functions to fix compiler errros.

Fix Divide op version.

* Add downgrade and upgrade passes for Equal op.

* Reformat test cases. Add helper functions.

Add UT for Equal op.

* Add upgrade/downgrade pass and UT for Greater op.

* Add upgrade/downgrade pass and UT for GreaterEq op.

* Add upgrade/downgrade pass and UT for Less op.

* Add upgrade/downgrade pass and UT for LessEq op.

* Add upgrade/downgrade pass and UT for Maximum op.

* Add upgrade/downgrade pass and UT for Minimum op.

* Add upgrade/downgrade pass and UT for Multiply op.

* Add upgrade/downgrade pass and UT for NotEqual op.

* Add upgrade/downgrade pass and UT for Power op.

* Force ops version 1.

* Don't inline templates.

* Fix namespaces and some formatting.

* Update ONNX Importer to produce v1 nGraph nodes.

* Fix function return type.

* Fix uninitialized local variable warning.

* Fix confilicting declarations.

* Apply clang-format.

* Fix errors for distributed nGraph with unavailable classes.

* Fix downgrade pass for LessEqual op.

* [FUSED] Add new Selu fused op.

* Add Selu to serializer

* Add Selu to ngraph.hpp

* Fix float16 denorm conversion, add rounding

* review comments

* Shape as a node in ConvBpropData (v1)

* Fix constant op creation

* Convbpropfilters dynamic shape

* Avgpool bprop dynamic shape

* Add downgrade pass

* Downgrade pass Avgpoolbackprop

* Shape checks and shape relevancy

* Dyn avgpoolbprop test case

* Revert "Dyn avgpoolbprop test case"

This reverts commit b094f2f99d00023593c667ff16cf060a586ad16a.

* Revert "Avgpool bprop dynamic shape"

This reverts commit ef8a4d197577c0d277d634baa9d6d082adcddae0.

* Revert "Downgrade pass Avgpoolbackprop"

This reverts commit 3f31f0fea8fe6e79bd958cbaa7cb8495008a795f.

* Call shape relevancy

* Revert Avgpoolbprop changes

* mark nodes as dynamic

* Add opset transform tests

* Change API

* Adjust strides

* Use size

* Merge PR 3776

* add shape relevance

* Add convbprop_data test case

* Add convbprop_filter test case

* Use is_constant helper

* Remove opset pdpd test case

* resolve conflict

* add checks for validate and infer

* Type prop tests

* Return pass created with register_pass call

* Add test

This matches the behavior in FindProtobuf.cmake, https://github.com/Kitware/CMake/blob/master/Modules/FindProtobuf.cmake#L624-L626 (https://cmake.org/cmake/help/latest/module/FindPackageHandleStandardArgs.html#command:find_package_handle_standard_args)

* Allow fused ops to create outputs and set types/shapes in pre_validation

* style fix

* Init commit to implement interface

*  Add two op interfaces for v0 and v1. Add a unit-test

* Add missing files

* Move test to separate file

* Add Fused Op interface

* Missing files

* style

* fused ops

* Remove V1 ops for now

* Added enum attributes. WIP

* Completed non-experiemntal non-fused-ops

* Add ops_attributes

* Minor fixes

* Minor fixes

* Added enum setting/reading test

* style-apply

* Added attributes tests

* Fix dialect init

* style

* fix typo

* Fix merge errors

* Include file with MLIR on

* Added tool for running Onnx models via nGraph C++ API.

* Code formatting.

* Added option to load inputs from binary files.

* Removed unused variables.

* Changed description.

* Moved read_binary_file function from test_tools to ngraph utils.

* Added option to provide arguments with equals sign.

* Bugfix.

* Changed reading file.

* Code refactoring.

* Added output printing.

* Added comments.

* Undo moved function.

* Removed unused imports.

* Update docstring
Co-Authored-By: Michał Karzyński <postrational@users.noreply.github.com>

* Update docstring
Co-Authored-By: Michał Karzyński <postrational@users.noreply.github.com>

* Added catching exception.

* Code refactoring.

* [SPEC] Rename operator Xor->LogicalXor

* Fix clang issue

* Fix bug in opset transformations. Add support for LogicalXor in backends

* Style fix

* Fix a bug in CPU emmiter

* [SPEC] Rename operator Or->LogicalOr

* [SPEC] Rename operator And->LogicalAnd

* [SPEC] Rename operator Not->LogicalNot

* [SPEC] Rename operator LessEq->LessEqual