* unit tests faster

* speed up binary zero size tests

* fix style error

* remove some of the redundant code

* Fix warnings in zero-sized tests; fix zero-sized tests return vector type for comparison ops

* Revert changes to cpu_emitter

1. AvgPoolBackprop type checking
2. Padding for max-pool op and kernel
3. Max pool backprop kernel, and integrating everything with CPU backend/serializer
3a. Add optional backpointer from the max pool backprop op to the forward prop op

* wip

* wip

* remove get_vector from runtime::TensorView class as it was for unit test only

* cleanup

* move writting vector to runtime::TensorView to the unit test dir

* merge fix

* PR review change

* update from PR comment

* update changes file

* Average pool type checking and kernel; type checking tests

* Fix and enable average-pool tests

* Docstring fix

* Extend AvgPool op type checking to support padding

* Untested code for padded avg-pool

* Unit tests for padded avg-pool

* Add CPU implementation

* Temp delete

* Docstring fix

* Docstring fix

* Add tests mixing padding and stride

* Temporary cut to ease merge

* Restore temporary cut for merge

* Empty commit to try to force CI to wake up

* Numerically stable sum so we can pass mxnet unit tests

* Add a small initial residual

* add interpreter nan check option

* add unit test

* speed up reduceslice with kernel emitter

* const-ify and fix a clang warning

* add elementwise ops, slice to for loops

* add broadcast codegen

* add Exp

* fix bugs introduced in eigen kernels

* fix another introduced bug in Eigen

* Fix an Atomic Bug with Sum, do some cleanup

* unit tests pass

* Add Reshape Op, passes Tests

* rewrite sum to correctly handle muti-threading

* Code Cleanup

* add some extra unary ops

* Address review comments

* fix an error in the review comment refactor

* Add Power op

* Add (most) of the Logic Ops

* Make Concat default to OpenMP kernel

* fix n-D reshape issue

* much faster compile time
* Remove all variables and just directly access inputs, output, and temps.
* compare layouts when checking if two ops are equal
* make performance counters available to all backends

* Definition and type checking for max pool

* Implement kernel, integrate into INTERPRETER, add a few unit tests, make function result type mismatch error message more informative (still need to update tests to reflect that)

* Temporarily delete unit tests to ease merge

* Temporarily delete unit tests to ease merge

* Restore deleted unit tests

* Fix a broken error message check in the unit tests

* Update to handle various TensorViewType-related things going away; add NGVM support

* Add codegen case

* Change various get_blah_shape methods to return const refs, and while we're here, make a similar change where it should have been done in convolution

* Use NDArray for max-pool tests

* Function can have multiple results
Remove external use of ValueType, TupleType, Tuple
Remove many external uses of Output and Input

* corresponding CPU backend changes

* Update master changes.

* Remove type arg from Function, add changes.md

* Merge changes.

* Move bodies to .cpp, add brief doc

* Merge CPU changes.

* Remove xla includes from non-xla files

* Remove xla from tests

* First part of xla tuple support

* change fprop_cache to assume multi-output bprop functions

* New wrappers for handling tuples with XLA

* Review comments

* remove old xla files

* fix merge errors

* hand edit models to use multi output instead of tuples

* CMake: TBB integration placeholder

* CMake: Integrate TBB

* CMake: Indent

* CMake: Rewrite TBB integration

* CMake: More TBB integration changes

* CMake: Install TBB headers and DSOs

* CMake: Don't install the TBB debug DSO

* CMake: Propagate ngraph's configured compiler setting over to MKL-DNN

* CMake: Restore TBB debug DSO installation

* CMake: Add installed headers to search path.
This needs to be cleaned up along with other header search cleanup

* CPU: Build and execute TBB flowgraphs

* CPU: TBB fixes

* CPU: More TBB fixes

* CPU: Allow both TBB and serial codegen for now

* TBB: get_arguments -> get_input_ops

* CPU: Use node methods

* CPU: Add TBB headers in the build directory to the search path

* TBB: Incorporate various changes from master

* CMake: Indentation fix

* CMake: Indentation fix

* CMake: TBB is mandatory so remove additional predicates

* TBB: Add a test

* CMake: Fix linker flags with GCC

* nan unit test

* fix NAN issue

* add INFINITY support

* remove ngvm

* remove NGVM from cmake

* remove ParameterizedConstant
* use simpler element Type definition
* Move TraitedType to NGVM directory

* aliased output unit test
* add support for aliased outputs to INTERPRETER and CPU

* Implement 'not' in interpreter and CPU; add unit tests for same

* Fix compile failure on CI

* remove use of ParameterizedConstant from unit test. Now using Constant instead. Constant is not a templated class.

* move ParameterizedTensorView to NGVM directory

* element_type cleanup

* add unit test

* fix Parameter as output in INTERPRETER

* cpu working

* parameter_as_output passing for INTERPRETER

* still works

* cleanup

* Replace function->get_result()->get_outputs() call with function->get_outputs()

* Test for parameter straight to output

* Roll back cmake modification for INTERPRETER autodiff

* tuple -> xla_tuple (+headers)

xla_get_tuple_element

get_tuple_element headers

adding get_tuple_element + fixes

get_tuple_element

XLAFunction; breaks tests

compiling libngraph.so w/ XLAFunction and Function; fixing tests is next

build & test fixes

formatting

m_arguments -> get_arguments (first step to hide m_arguments)

removing some get_arguments

get_arguments -> get_arguments_DEPRECATED

more DEPRECATED

formatting

m_arguments no more! at least technically

halfway

tests pass

xla_function.cpp/hpp

move defs into cpp files

get_input_argument -> get_input_op

formatting

* add xla_function.* and assert -> throw

* assert -> throw, remove cassert

* removing emitter.cpp; poor guy got left behind

* addressing feedback p1

* rename GetTupleElement -> GetOutputElement

* adding a comment to get_arguments_FOR_GRAPH_REWRITE_ONLY

* removing virtual off set_name

* WIP generalized dot

* Add some multi-axis 3D, 4D, and 5D tests

* Add test on some 'pretty big' tensors

* Reworked dot to have less flexible axis-pairing behavior

* Backprop for dot... and a fix for a dumb bug in CoordinateTransform

* Forgot to commit some stuff in merge

* Disable tests that currently don't work on CPU

* Fix temporarily disabled test that should pass on NGVM and INTERPRETER but wasn't due to new axis-selection convention for dot

* Remove obsolete ScalarTensorProduct kernel/instruction

* Review comment

* s/n_dot_axes/dot_axis_count/

* s/dot_axis_count/reduction_axes_count/

* Adapt CPU emitter dot fallback to new kernel

* add support for n-dimensional operations by calling runtime/kernels

* rename cpu backend classes and files

* add tuple support to INTERPRETER

* move ndarray out of runtime to test where it belonged

* fix Tuple op per review comment

* Remove unnecessary dependency on codegen in interpreter

* Check for incorrect return type and undeclared parameters in Function's constructor

* Address review comments

* Add scarier error message when the result node has null return type (should never happen)
* Add new constructor for Function that doesn't require the return type, and unit test for same

* New Interpreter backend

* PR review comments

* More RP fixes

* oops

* make autodiff tests backend aware

* wip

* wip

* more ops

* wip

* fix merge error

* merge fixes

* Simpler kernel for broadcast

* Fixed behavior for integer divide-by-zero, added unit tests

* Strided and higher-dimensional slice (just tested to 3D)

* Higher-dimensional sum

* Replace-slice de-Eigenized; NOT TESTED AT HIGHER DIMENSIONS YET

* Correct sum behavior when eliminating zero-length axes; add unit tests; also, add higher-dim unit tests for replace-slice

* Higher-dimensional reduce, 'cause hey, why not?

* Remove BroadcastScalarInstruction

* Adding test for an observed failure at trivial sum on 5-tensors

* De-Eigenized and higher-dimmified concat

* Replace 'auto' in the kernels

* temporary delete to ease merge

* Re-insert tests that were deleted to ease merge

* Refactor view-iteration

* De-Eigenize reshape

* Rework divide kernel to use std::enable_if to distinguish between floating and non-floating types

* Update docs to reflect newly implemented cases in several ops

* Rename parameters to View for more clarity; remove axis_walk_order (it's redundant)

* Formatting

* More terminological rejiggering

* De-Eigenize scalar-tensor product

* De-Eigenize dot

* Update docstrings

* Remove 'implementation status' tables from docstrings

* Change step -> strides everywhere for consistent terminology

* Formatting

* Replace asserts in view.cpp with exceptions

* Fix typo

* Fix incorrect result type in dot1d test (ouch...)

* Add missing support for Float64 to ngvm/external_function

* Add int16 and uint16 (how was this missing?)

* A few more additions relative to the missing element types

* Disable tests that will not pass on CPU; they can still be run with test/unit-test --gtest_also_run_disabled_tests --gtest_filter='DISABLED_NGVM.*'

* Move project_ and inject_ functions to common.[ch]pp, not view.[ch]pp

* Rename View to CoordinateTransform

* Add prefix ++ and += to CoordinateIterator

* add json lib 2.1.1

* add json serialization of graph

* cleanup

* enhance unit test

* remove FunctionProvider class and replace with virtual get_function() in Node

* remove json code from element_type

* move serialize to be directly in the ngraph namespace. cleanup header file.

* add cname check to element::operator==

* add using json = nlohmann::json