Merge branch 'ayzhuang/in-place-concat' of https://github.com/NervanaSystems/ngraph into ayzhuang/in-place-concat

* Update ArgReduce to handle i64 indices.

* Formatting.

* Add throw for output types other than int32/64.

* Add output type to hash.

* Add type to throw.

* Interpreter doesn't currently support 64bit output indices for argmin/max and so disabling this test [JIRA:NGRAPH-3183].

* Implement partial shape/type propagation for Convolution; fail for want of unit tests

* Implement unit tests for partial shapes/types for Convolution

* fix Klockwork warnings CPU part 1

* fix spelling error

* fix a typo

* Fix gemm `input_c` broadcasting.

* Comments.

* Add comment

* [ONNX] Non-linear ops

* Style check

* Temp bn update commit.

* Add CUDNNBatchNorm which adds two additional outputs to batchnorm, the batch mean and batch inv variance.
The batch mean is the same as the output mean if the cummulative average factor is 1.0. Add BatchNormCache pass which replaces all BatchNorm ops that are inputs to BatchNormBackprop
with CUDNNBatchNorm which outputs the saved batch statistics directly to the backprop step.

* Updated bn cache pass, removed extra tests, added test checking that provided stats are used in bprop instead of batch stats.
This test was disabled for interpreter as the reference kernel needs to be updated to use provided statistics.

* Formatting.

* Update to new batch norm API.

* CUDNNBatchNorm -> BatchNormTrainingWithStats

* new line

* Preprocess input variance into BN denominator for cudnn (#1885)

* Add explicit cuda kernel to calculate what cuDNN describes as the inverse
variance. In reality, the backward cudnn kernel for BN requires 1.0f / sqrt(variance + eps),
which is the batchnorm denominator for each channel (a numerically stable inverse stddev).

This introduces op annotations for batch norm backprop and updates the cudnn_emitter to support the insertion of this cuda kernel when required.

* Disable second test on INTERPRETER.

* [CI fix, c.f. @chris lishka] Roll back #906 and #1853.

* disable some quantize tests on gpu.

* skeleton backend

* Code owner from if conditioning

* add simple placement for interpreter and register pass in hybrid

* placement policy applied

* clone the function if needed

* split the function

* Compile subfunctions in corresponding backends

* hybrid backed works as is for abc test

* cleanup

* add placement policy for CPU

* cleanup a little

* add simple op cost method to backend

* enable CPU pass via flag

* address clang-format PR issue

* reslove build

* clean-up

* update manifest

* disable HYBRID as default build

* style

* addressing offline discussion

* more offline discussion

* onnx: add information about a domain to operators set
Signed-off-by: Artur Wojcik <artur.wojcik@intel.com>

* onnx: updates after review
Signed-off-by: Artur Wojcik <artur.wojcik@intel.com>

* onnx: update comments in the code
Signed-off-by: Artur Wojcik <artur.wojcik@intel.com>

* onnx: fix bug in node's description method
Signed-off-by: Artur Wojcik <artur.wojcik@intel.com>

* onnx: fix CentOS compilation
Signed-off-by: Artur Wojcik <artur.wojcik@intel.com>

* onnx: more after review changes
Signed-off-by: Artur Wojcik <artur.wojcik@intel.com>

* Support for parameterized testing combined with backend driven manifests.

* clang-format fix

* Improved nGraph variants of gtest fixture and parameterized testing macros

* Reverted mistaken change to src/ngraph/runtime/gpu/CMakeLists.txt

* added half_toward_zero; all previous tests passing

* all rounding modes added with unit tests

* fix cpu emitter

* round mode doc

* round out round modes

* doc typo

* using  names for round modes

* use ceil/floor for rounding functions instead of round/nearbyint

* clean up doc

* equidistant

* Collapse dimensions for arithmetic reduction ops to support faster kernels

* Propagate in-place constants and allow in-place reshapes for more cases

* style fix

* Additional checks for parameter and constant to help backends that dont propagate in-place parameter and constant inputs

* Allow non-destructive pass through onlyu if memory sharing is disabled

* Address PR feedback

* Bug fix for collapse dimensions in case of null reduction

* split bn into bn_inference bn_training

* fix warnings

* Add GPU support for the new BN ops (#1569)

* Add GPU support and change batchnorm_globalstats test to use BNInference.

* Changed test back to using BNTraining for global stats and updated cudnn backend to account for it.

* Fix issues in merge with master.

* Formatting.

* CPU fixes

* remove 5-arg training BN for now

* more fixes

* python batchnorm changes

* fix onnx_import

* fix a call BatchNormInference c-tor

* yet another fix to BatchNormInference c-tor

* AND yet another fix to batchnorm_inference c-tor

* ops.py

* address adam's feedback

* Remove unnecessary parameter/argument.

* remove batch_norm_training_relu_with_global_stats

* remove bn_relu (training)

* Update runCI.sh

Accommodate manual reproduction script.

* Update README.md

* Update runCI.sh

* fixes
* added rebuild parameter

* Fix rebuilding nGraph

* Added execution permission to runCI.sh

* Add bfloat16 data type in ngraph

* Update on bfloat16 files
- remove uint16_t related functions
- adding/removing const modifier in functions
- default assignment operator
- style conformance

* Add bf16 element type handling in Constant OP

* Update bfloat16 data type implementation
- support nan and infinity cases
- add rounding option when creating bfloat16 from float

* Update the comment for copyright info

* Adapt Tensor class to have partial shapes

* Add PartialShapes to Input, Output, Function, Node classes

* Terminological cleanup

* Add PartialShape propagation for Parameter and Result

* Implement partial-shape propagation for elementwise ops

* More comments

* One more comment tweak

* Add tests for the merge functions

* Add merging of undetermined element types

* Fix a goophup in deserializer implementation

* Implement fallback for ops that do not support partial shape/type validation

* Updates for some older unit tests, now that operator[] exists

* Add missing validate_punt_if_incomplete to AllReduce

* Implement partial shape/type propagation for AllReduce

* Implement partial shape/type propagation for Reshape

* Remove unneeded validate_punt from Result

* Implement partial shape/type propagation for Reverse

* Implement partial shape/type validation for ReverseSequence

* Implement partial shape/type validation for ArithmeticReduction

* Better docstrings for the stuff introduced in #1692; remove prototype for unimplemented, unused PartialShape::append()

* One more docstring thing I forgot to save

* Switch terminology from 'determined/undetermined' to 'static/dynamic'

* Switch terminology from 'complete/incomplete' to 'static/dynamic' for shapes; fix up some mushily worded comments

* Fix overzealous edits from the last commit

* Rename one test that escaped the Great Renaming

* Remove unnecessary validate_punt_if_dynamic from Reshape

* Fix comment typo

* Rewrite operator+ and operator* for Dimension as members, not friends

* Formatting tweak

* Show argument types/shapes in long NodeDescription; tank unit tests to block merge

* Fix dynamic element type propagation for elementwise ops, add some unit tests for same

* Fix error message

* Roll 'Not' back to existing behavior (non-boolean input types allowed)

* Add a TODO tag to a todo item

* Add unit tests for partial shape/type propagation with ReverseSequence

* Add unit tests for partial-shape/type propagation for ArithmeticReduction (via Sum)

* Implement partial type/shape propagation for GetOutputElement

* Implement partial type/shape validation for IndexReduction, and unit tests

* Function signatures

* Add implementations, unit tests for relaxes/refines functions

* Generalize project/reduce/inject functions to cover PartialShape, move to shape_util.[ch]pp

* Deal with std::find_if #include issues

* Fix more include madness

* Review comments

* Adapt Tensor class to have partial shapes

* Add PartialShapes to Input, Output, Function, Node classes

* Terminological cleanup

* Add PartialShape propagation for Parameter and Result

* Implement partial-shape propagation for elementwise ops

* More comments

* One more comment tweak

* Add tests for the merge functions

* Add merging of undetermined element types

* Fix a goophup in deserializer implementation

* Implement fallback for ops that do not support partial shape/type validation

* Updates for some older unit tests, now that operator[] exists

* Add missing validate_punt_if_incomplete to AllReduce

* Implement partial shape/type propagation for AllReduce

* Implement partial shape/type propagation for Reshape

* Remove unneeded validate_punt from Result

* Implement partial shape/type propagation for Reverse

* Implement partial shape/type validation for ReverseSequence

* Implement partial shape/type validation for ArithmeticReduction

* Better docstrings for the stuff introduced in #1692; remove prototype for unimplemented, unused PartialShape::append()

* One more docstring thing I forgot to save

* Switch terminology from 'determined/undetermined' to 'static/dynamic'

* Switch terminology from 'complete/incomplete' to 'static/dynamic' for shapes; fix up some mushily worded comments

* Fix overzealous edits from the last commit

* Rename one test that escaped the Great Renaming

* Remove unnecessary validate_punt_if_dynamic from Reshape

* Fix comment typo

* Rewrite operator+ and operator* for Dimension as members, not friends

* Formatting tweak

* Show argument types/shapes in long NodeDescription; tank unit tests to block merge

* Fix dynamic element type propagation for elementwise ops, add some unit tests for same

* Fix error message

* Roll 'Not' back to existing behavior (non-boolean input types allowed)

* Add a TODO tag to a todo item

* Add unit tests for partial shape/type propagation with ReverseSequence

* Add unit tests for partial-shape/type propagation for ArithmeticReduction (via Sum)

* Implement partial type/shape propagation for GetOutputElement

* Basic support for partial shape/type propagation for OneHot

* Function signatures

* Add implementations, unit tests for relaxes/refines functions

* Update OneHot to take PartialShape for result, with dynamic dimension allowed at non-one-hot axes

* Generalize project/reduce/inject functions to cover PartialShape, move to shape_util.[ch]pp

* Deal with std::find_if #include issues

* Fix more include madness

* (->{ ; )->}

* size_t{...} -> static_cast<size_t>(...)

* IntelGPU backend: Profiling with various statistics

* PR1863. Comments addressed. Types changed to allow scientific output

Fixes an error when building on macOS (with CPU backend disabled, but not sure if that makes a difference):

```
/Users/amprocte/Work/ngraph/src/tools/nbench/benchmark.cpp:36:29: error: use of undeclared identifier '_MM_FLUSH_ZERO_ON'
    _MM_SET_FLUSH_ZERO_MODE(_MM_FLUSH_ZERO_ON);
                            ^
/Users/amprocte/Work/ngraph/src/tools/nbench/benchmark.cpp:37:33: error: use of undeclared identifier '_MM_DENORMALS_ZERO_ON'
    _MM_SET_DENORMALS_ZERO_MODE(_MM_DENORMALS_ZERO_ON);
```

* Add merge_rank function

* Update infer_windowed_reduction_output_shape to use PartialShape

* Minor simplification

* Some unit tests and (whaddaya know) fixes for infer_windowed_reduction_output_shape

* Update infer_batched_pooling_forward to use PartialShape

* Update pooling fprop ops for partial shapes

* Update pooling bprop ops for partial shapes

* Add test-failing reminders to implement unit tests for partial shape/type prop for pooling ops

* Add unit tests for partial shape propagation for poolign ops

* Nuke C-style casts for Dimensions/Ranks in validation_util.cpp

* Adapt Tensor class to have partial shapes

* Add PartialShapes to Input, Output, Function, Node classes

* Terminological cleanup

* Add PartialShape propagation for Parameter and Result

* Implement partial-shape propagation for elementwise ops

* More comments

* One more comment tweak

* Add tests for the merge functions

* Add merging of undetermined element types

* Fix a goophup in deserializer implementation

* Implement fallback for ops that do not support partial shape/type validation

* Updates for some older unit tests, now that operator[] exists

* Add missing validate_punt_if_incomplete to AllReduce

* Implement partial shape/type propagation for AllReduce

* Implement partial shape/type propagation for Reshape

* Remove unneeded validate_punt from Result

* Implement partial shape/type propagation for Reverse

* Implement partial shape/type validation for ReverseSequence

* Implement partial shape/type validation for ArithmeticReduction

* Better docstrings for the stuff introduced in #1692; remove prototype for unimplemented, unused PartialShape::append()

* One more docstring thing I forgot to save

* Switch terminology from 'determined/undetermined' to 'static/dynamic'

* Switch terminology from 'complete/incomplete' to 'static/dynamic' for shapes; fix up some mushily worded comments

* Fix overzealous edits from the last commit

* Rename one test that escaped the Great Renaming

* Remove unnecessary validate_punt_if_dynamic from Reshape

* Fix comment typo

* Rewrite operator+ and operator* for Dimension as members, not friends

* Formatting tweak

* Show argument types/shapes in long NodeDescription; tank unit tests to block merge

* Fix dynamic element type propagation for elementwise ops, add some unit tests for same

* Fix error message

* Roll 'Not' back to existing behavior (non-boolean input types allowed)

* Add a TODO tag to a todo item

* Add unit tests for partial shape/type propagation with ReverseSequence

* Add unit tests for partial-shape/type propagation for ArithmeticReduction (via Sum)

* Implement partial type/shape propagation for GetOutputElement

* Function signatures

* Add implementations, unit tests for relaxes/refines functions

* Generalize project/reduce/inject functions to cover PartialShape, move to shape_util.[ch]pp

* Dynamic shpae/type prop for Quantize

* Add unit tests for partial shape/type validation for Quantize

* Implement partial shape/type validation for Dequantize, with unit tests

* Remove #if 0'd code

* Deal with std::find_if #include issues

* Fix more include madness

* comparisons

* move more unit test out of backend_test.in.cpp

* move more tests

* move more tests

* Add support for perf counter collection in DEX

* Addressed PR feedback