* Robust debugging docs

* Add section on nbench and address comments from review

* Collaborate with Gauri to revise profiling section

* Revise and PR feedback

* Move note

* Fix wording

* Order sections more logically and fix a comma

* Phrasing fix on nbench_tf summary

* More prominent notice of experimental debug flags

* Better description for diagnostic tools

* Remove miscellaneous framework support

* clean up section

* Remove deprecated links

* Update sitemap to not use a page title

* Useful descriptions

* PR feedback

* Not a flag

* Prebuilt MLIR compile flag available

* Remove duplicate flag

* update pass manager example

* Meta documentation note in release notes

* Ensure docs build with lastest upstream ops

* Transpose op doc fixes

* Better intra-doc links

* Commas in csv format are important

* Final review with Gauri

* Remove dupes, CPU-specific envvars

* changes re: Comments from Gauri
Co-authored-by: Robert Kimball <robert.kimball@intel.com>
Co-authored-by: Sang Ik Lee <sang.ik.lee@intel.com>

* WIP

* WIP

* Refactored existing convolution

* Add Channel and num of filters bounds parameters to helper

* Works on unit-tests. v1 op gets converted and breaks

* Fixed group conv with groups in filters shape. Tests pass

* style

* add LIT tests

* Switch outer loop to affine loop

* re-org code

* PR fixes

* Revert ops.td

* PR fixes

* Remove halide.

* Remove stale code.

* Put back incorrectly removed code.

* Use optimal processor count for parallel build.

* Add missing dependency.

* Build LLVM at CMake config time and use CMake config file for dependencies.

* Temp save.

* Codegen functioning.

* Support prebuilt LLVM.

* Remove zlib.

* Relax CMake minimum version.

* Fix bug.

* Lower CMake version requirement.

* Pass more args.

* Add proper quote.

* Remove quotes.

* Try to separate flags for external projects.

* Don't pass flags to LLVM.

* Pass CXX abi.

* Remove stale print.

* Update gtest flags.

* Try to fix build issue.

* Patch LLVM to prevent building clang-tools-extra.
Co-authored-by: aslepko <44713115+aslepko@users.noreply.github.com>

* [MLIR] Use call back for MatMul.

* Use callback for Gemm.

* Use mkldnn callback for Softmax.

* Address PR feedback.

* Fix merge errors.

* Change to tail allocation struct.

* Use mkldnn callback for AvgPool.

* Add callbacks for AvgPoolBackprop, MaxPool, and MaxPoolBackprop.

* Fix merge errors.

* Use UnrankedMemRefType for callbacks.

* Address PR feedback.

* Cleanup.

* Address PR feedback.

* Fix a bug.

* Use global variable to hold attributes.

* Convert layout if needed for pooling.

* Address PR feedback.

* Add header.

* Address PR feedback.

* Update Copyright to 2017-2020.

* Address PR feedback.
Co-authored-by: Scott Cyphers <diyessi@users.noreply.github.com>

* Changes version from v0 to default for Min and Max ops.

* Added broadcast option in variadic function.

* Added tests.

* Added test for opset8.

* Undo changes for Mean op.
Co-authored-by: Tomasz Dołbniak <tomasz.dolbniak@intel.com>
Co-authored-by: Scott Cyphers <diyessi@users.noreply.github.com>

* add lookup table v2 fused op placeholder

* implement frop

* fix fprop and add ut

* turn off fluid test build

* add missed ut

* remove get_output_shared_ptr

* implement bprop

Co-authored-by: Scott Cyphers <diyessi@users.noreply.github.com>

This reverts commit 1e58565a.

* [ONNX] Remove unnecesary Mean in opset 8

* Use opset1 instead of v1

* Fix merge artifact

* Use opset1 Divide instead of opset0
Co-authored-by: Robert Kimball <robert.kimball@intel.com>

* Checking if provenance_tags key exists

* Add provenance tag prototype

* Format provenance tag

* Display provenance tag

* Clean debug printing

* Add const to variables

* Separate method for add provenance tags

* Return NodeVector reference

* Return const NodeVector

* Moved add_provenance_tags function to commons

* Style apply

* Simple model for tests

* Provenance tag test

* Expect substring instead of  equal

* Add provenance tags to intermediate nodes recursively

* One tag per node

* Add traverse node args instead of recursion

* Return NodeVector instead of set of pointers

* Use treverse_nodes and lambda function

* Remove unused helper functions

* Remove is_constant() condition

* Update test model prototxt

* Update test substring

* Use node name and output names to build provenance tags in onnx importer

* Unit tests for onnx_importer provenance tags

* Missing <numeric> include

* Add provenance tags to constants buit from ONNX initializers

* Add provenance tags to Constants and Parameters created out of ONNX inputs and initializers

* More strict assertions in onnx provenance tests

* Unit test for onnx importer Parameter nodes tagging

* Helper function for the onnx provenance tests

* Some docs

* Obsolete comment removal

* Separate file for onnx provenance tags unit tests

* Code formatting

* Move the inputs tagging to the Graph class

* Tagging moved to the Graph class entirely

* Missing include and extra helper variable

* Unit tests helper documentation

* Change the UT helper to lowercase
Co-authored-by: Katarzyna Mitrus <katarzyna.mitrus@intel.com>

* [PY] Add flag to compile python API in debug mode

* Move debug and release flags to helper function

* Add missing Docstring
Co-authored-by: Sang Ik Lee <sang.ik.lee@intel.com>

* AliasMap WIP

* Added liveness info

* WIP

* WIP: Tests

* WIP: LIT tests

* Added knobs for mem optimization pass. More LIT tests

* Revert affine_lowerer change

* More elaborate comment

* Minor fixes

* style-apply

* Rename liveness

* Switch to Analysis framework

* Fix optimization conditions

* Remove LIT test

* style

* Switch to equivalence relationship impl of non-alias relationship

* refined comment

* Switch non-alias to equivalence relationship

* Fix bad merge

* Adding tests. WIP

* Added buffer size tracking and unit-tests

* Added LIT and unit-tests

* Turn optimization ON

* style

* fix unit-tests

* Fix useCount

* Fix copyright and typo

* Refine few comments, remove new lines

* style fix
Co-authored-by: Scott Cyphers <diyessi@users.noreply.github.com>
Co-authored-by: Sang Ik Lee <sang.ik.lee@intel.com>

Co-authored-by: Scott Cyphers <diyessi@users.noreply.github.com>

* Update sitemap to not use a page title

* Document Release Notes for 0.28.0; update section on Distributed Training
Co-authored-by: Scott Cyphers <diyessi@users.noreply.github.com>

* Restore constant folding for DynReshape until users are converted to v1 Reshape

* Disbale test when no serialization
Co-authored-by: baojun <32073718+baojun-nervana@users.noreply.github.com>

* Move GCPU to runtime/gcpu so that unit test manifest works

* style

* Add tests to manifest

* Fix CPU only check

* Update GCPU dependency
Co-authored-by: Scott Cyphers <diyessi@users.noreply.github.com>

* Removed LogSoftmax FusedOp, changed onnx to produce v1

* Code review remakrs introduced

* fix after merge from master
Co-authored-by: Scott Cyphers <diyessi@users.noreply.github.com>

* Added round op

* Add CPU support, unit tests

* Disable UT for PlaidML

* Update year
Co-authored-by: Scott Cyphers <diyessi@users.noreply.github.com>

* Return v1::Reshape in builder::reshape function.

* Use default_opset in ONNX Convolution operators.

* Revert builder::reshape to return v0 op.

* Remove unused code.

* Update year in license header.

* Remove AutoBroadcastSpec arg and use default value.
Co-authored-by: Scott Cyphers <diyessi@users.noreply.github.com>

* Fixed creation NonMaxSuppression op with dynamic shapes

* Fixed code style
Co-authored-by: Robert Kimball <robert.kimball@intel.com>
Co-authored-by: Scott Cyphers <diyessi@users.noreply.github.com>
Co-authored-by: Michał Karzyński <postrational@users.noreply.github.com>
Co-authored-by: Sang Ik Lee <sang.ik.lee@intel.com>

Co-authored-by: Diego Caballero <diego.caballero@intel.com>
Co-authored-by: Scott Cyphers <diyessi@users.noreply.github.com>

* Add serializer/deserializer check to interpreter

* Fix TopK

* Fix GRUCell

* Fix RNNCell. Does anybody test their own code? Apparently not.

* Fix LSTMCell

* Fix MVN

* Fix Select v1

* Fix GroupConvolution

* Fix ScalarConstantLike

* General cleanup

* Revert "General cleanup"

This reverts commit d765d2c2451cf5d3c9a41c4d7d672c278783b0a2.

* Fix op_version_tbl.hpp

* More cleanup

* Fix LSTMSequence

* revert

* Disable INTERPRETER serialize test by default
Co-authored-by: Scott Cyphers <diyessi@users.noreply.github.com>

* Add using  Reshape:v1

* Use Reshape:v0 in group_conv

* Use Reshape_V1 in builder:flatten

* builder:v1 introduced

* Revert old builders to use Reshape:v0

* removed unused Transpose test

* Update test/opset_pass/transpose_opset_pass.cpp

* Changed builders to opset1

* Use opset1 instead of v1
Co-authored-by: Robert Kimball <robert.kimball@intel.com>
Co-authored-by: Sang Ik Lee <sang.ik.lee@intel.com>

* Added GatherElements op to ONNX importer.

* Changed attributes to inputs.

* Undo unnecessary changes.

* Added Pad version 11.

* Added 11 version for Pad op.

* Changed axis element type.

* Added optional input.

* Code formatting.

* Added test.

* Update onnx_import.in.cpp

* Removed debug code.

* Added function for setting mode.

* Added support for pads argument provided as Constant.

* Code refactoring.

* Code refactoring.
Co-authored-by: Scott Cyphers <diyessi@users.noreply.github.com>
Co-authored-by: Sang Ik Lee <sang.ik.lee@intel.com>

* Removed Recoprocal FusedOp and moved definition to ONNX importer.

* Removed unnecessary changes.

* Added test.

* Undo unnecessary changes.
Co-authored-by: Sang Ik Lee <sang.ik.lee@intel.com>

Co-authored-by: Sang Ik Lee <sang.ik.lee@intel.com>
Co-authored-by: Scott Cyphers <diyessi@users.noreply.github.com>

Co-authored-by: Sang Ik Lee <sang.ik.lee@intel.com>

Co-authored-by: aslepko <44713115+aslepko@users.noreply.github.com>
Co-authored-by: Scott Cyphers <diyessi@users.noreply.github.com>

* Changed onnx slice to produce v1

* onnx where produces v1 ops
Co-authored-by: Scott Cyphers <diyessi@users.noreply.github.com>

* Move some ops into v0

* namespace

* Make comments pretty

* Make comments pretty

* Merge fix
Co-authored-by: Robert Kimball <robert.kimball@intel.com>

* Checking if provenance_tags key exists

* Add provenance tag prototype

* Format provenance tag

* Display provenance tag

* Clean debug printing

* Add const to variables

* Separate method for add provenance tags

* Return NodeVector reference

* Return const NodeVector

* Moved add_provenance_tags function to commons

* Style apply

* Simple model for tests

* Provenance tag test

* Expect substring instead of  equal

* Add provenance tags to intermediate nodes recursively

* One tag per node

* Add traverse node args instead of recursion

* Return NodeVector instead of set of pointers

* Use treverse_nodes and lambda function

* Remove unused helper functions

* Remove is_constant() condition

* Update test model prototxt

* Update test substring

* Make code slightly more readable
Co-authored-by: Tomasz Dołbniak <tomasz.dolbniak@intel.com>
Co-authored-by: Michał Karzyński <postrational@users.noreply.github.com>
Co-authored-by: Scott Cyphers <diyessi@users.noreply.github.com>
Co-authored-by: Sang Ik Lee <sang.ik.lee@intel.com>

Co-authored-by: Michał Karzyński <postrational@users.noreply.github.com>
Co-authored-by: Sang Ik Lee <sang.ik.lee@intel.com>

* Use normalize in onnx importer

* Introduced changes after code review
Co-authored-by: Sang Ik Lee <sang.ik.lee@intel.com>

Co-authored-by: Sang Ik Lee <sang.ik.lee@intel.com>
Co-authored-by: Scott Cyphers <diyessi@users.noreply.github.com>
Co-authored-by: Robert Kimball <robert.kimball@intel.com>

* Call generic CPU kernels for Broadcast and Reshape

* Move generic kernels to opt_kernel

* Use strides in broadcast
Co-authored-by: Scott Cyphers <diyessi@users.noreply.github.com>