* remove tensor_call from backends

* remove obsolete methods

* CMake: Allow build target arch to be overridden

* Add DNGRAPH_TARGET_ARCH option to install docs

* get_input_op -> get_argument

* more replacing

* more replacing2

* Update default paths in setup.py

* Update defaults arguments in tox

* remove obsolete

* change to use new Backend API

* rename parameter

* remove deprecated

* remove all legacy Backend API usage

remove deprecated files

* pull in changes from master

* fix GPU calls

* disable tests in convolution generator

* update per PR comments. Enable performance counter feature.

* update per PR comments

* fix build error

* fix conditionally compiled test :(

* BatchNorm documentation

* Fix typo, install URL

* Switch to desired BatchNorm

fixed variable settings in contrib/docker/make-dimage.sh script

* Add python wrapper for nGraph Reduce operation.

- Add UT.

* Refactoring.

- Add UT case with default reduction on all axes.

* Extend `reduce` operation signature to also accept `Function` object.

- Add UT case.

* Fix formatting errors.

* Begin prototype of cudnn_emitter.

* Added GPURuntimeContext to gpu_external_function for passing through to JIT functions.

* gpu_emitters now utilize gpu runtime context.

* Moved cublas and cudnn handles into GPURuntimeContext pointer and out of callframe EntryPoint.

* Added CUDNNEmitter, comparable to MKLDNNEmitter,
which allows for cudnn kernels to be defined via
lambda primitives that are emitted and
subsequently called during graph execution.
An example implementation is provided for op::Sum.

* Added GPURuntimeContext to gpu_external_function for passing through to JIT functions.

* gpu_emitters now utilize gpu runtime context.

* Moved cublas and cudnn handles into GPURuntimeContext pointer and out of callframe EntryPoint.

* GPURuntimeContext should be stored as unique_ptr in external function.

* GPURuntimeContext should be stored as unique_ptr in external function.

* Extract raw pointer from unique for cudnn_emitter.

* Removing unrelated code from PR.

* GPURuntimeContext needs to be a strict C interface in case
the native compiler and clang are utilizing different glibc ABIs.
Updated to reflect this.

* Added cudnn::primitive typedef for better readability.

* Moved allocation of CudaFunctionPool to external function
so that it is available during gpu emission.

* Fixed too-late initialization of cudart.

* Fixed too-late initialization of cudart.

* CUDNNEmitter moved into superset class GPUPrimitiveEmitter.
The GPUPrimitiveEmitter handles the emission of all gpu primitives,
including cudnn, cuda, and cublas. CUBLASEmitter support not yet included.

* Added unordered_map for cacheing primitives in the gpu_emitter.

* Added dtor to GPUPrimitiveEmitter to cleanup compiled functions.

* Adding back a serialized model graph that was accidentally rem* Added a few additional helpers to use ngraph::row_major_strides.

* added whitespace per @fengleitian's comment

* added whitespace per @fengleitian's comment

* Remove implicit type conversions from size_t to int.

* Add op::MaxPool, op::MaxPoolBackprop and op::Pad to GPU transformer (#817)

* Added pooling for 1 and 2dimensions. 1d uses a cuda kernel and 2d utilizes cudnn.
Padding is not yet supported.

* Normalized call signature on gpu emission for 1d max pool. Added a few comments.

* Max pool backprop impl. inprogress. Amend this commit.

* Max pool backprop implemented. Note that cuDNN
requests the output tensor for the maxpool operation but it is not required for computation.

* Formatting and invokation for maxpool changed.

* Fixed too-late initialization of cudart.

* Added padding kernel that is used with maxpool. Need to investigate remaining tests.

* Changed dimensionality check to correctly
determine if data is 1d or not.

* Added 3d MaxPooling (forward), verified by forcing 2d case to use Nd pooling routines.

* Added 3d MaxPooling (backward), verified by forcing 2d case to use Nd pooling routines.

* Moved cudnn prologues for maxpool into ngraph runtime and out of primitive so
that the only execution occuring on the JIT runtime is the evaluation of the op kernel.

* Refactored forward and backward pooling into single CUDNNEmitter::build_pooling interface
with a runtime switch to determine if the op is forward or backward propagation.

* Cache preconstructed cudnn kernel for maxpool if it has already been constructed.

* Forgot to add padding arrays back into cudnn kernel for MaxPool in the 2d case.

* Fixed namespace issues and use join(...,'_')

* Refactored 4d/Nd tensor descriptor builder into single function.

* Changed conditionals and comments. Now throws if MaxPool on more than 3 spatial dimensions is requested.

* Fixed forward declare for GPURuntimeContext (class -> struct).

* Clang complains about missing braces on brace-initializer. Fixed implicit conversions.

* Fixed implicit conversions (clang).

* Reverting changes on autodiff test for maxpool. @Krovatkin will update later.

* add slice op, first version

* change size to output size

* fix bugs

* working version

* using exist function for join and strides

* clang format

* revert accidental change

* add a getter for root node

* recurrent graph rewrite

* fix perms, rename match_root -> get_match_root

* fix comp errors

* make match_root return the topmost match; fix tests

* add convolution in progress

* enable 1 test

* convolution in progress

* use filter descripter

* filter discreptor bug fix

* tensor format

* add missed dimension calculator

* forward convolution 4d without dilation and padding working

* data dilation(deconvolution) and enable some test

* add backprop convolution data and filter

* backprop can compile

* pass unit test, but still have problem on padding

* 2d, symmtric padding, no data dilation works now

* clean up code

* extend gpu convolution to nd

* fix some bugs

* working version for upto 3d convolution, code format.

* remove nunecessary changes

* add restriction for data dilation and asymmetric padding

* clang format

* support upto 3D convolution for now

* change comments to not implemented

* change comments to not implemented

* add quary for additional GPU workspace for convolution

* clang format

* code format

* using row_major_strides

* using join

* fix bug for join

* refactor dimension calculation

* Enable BatchNorm op

* Enable function call op

* Enable get output element op

* new backend API in graph partition

* update API

*  zero dimension tensor elimination init

* more ops + refactor + tests

* revert pattern.cpp

* add internal zero-length test

* address Scott's feedback

* fix comp errors

* proper static init

* get rid of unique-ptr

* refactor hashmap into virtual get_default_values on op classes

* fix formatting

* back out api change

* Remove the no longer supported alternative installation method for
python binding.

* Put back CMakeLists.txt as it is used by travis ci dockerfile.

* Remove python/CMakeLists.txt and update Travis CI

* CPU: Optimize 4D "nGraph" Reshapes (shuffle+reshape)

* CPU: Add kernel sources

* CPU: Replace 2D with 3D reshape

* CPU: Fixes

* CPU: Simplify

* remove parameter check from Function::get_ops()

* create validate pass to hold parameter validation

* unskipping passing gpu tests

* skipping failing gpu tests

* import pytest as needed

* fix style issues

* unskip passing test

* add additional skip reason, unable to compile

* repacking recurrent matching as a standalone class

* RecurrentMatcher

* add a getter for root node

* address Scott's feedback

* WIP editing so far for review and feedback

* Add missing env var export for neon install new process

* Add modified venv setup for TF

* More edits for FW integration and landpage

* Revise from PR feedback

* More PR feedback and editing for clarity

* Minor rewording, clearer explanation

* Final pass edit

* more editing

* adding support for GPU backend to contrib/docker

added gpu dockerfiles

renamed Dockerfile for centos74

fixed NGRAPH_GPU_ENABLE cmake flag name

* Check for GPU support on the host system and fall back to CPU if not present

* removed double option for PREBUILT_LLVM

* updated README.md with additional references for GPU support

* added clarifying comments

cleaned up duplicate settings

* removed deprecated targets from the contrib/docker/Makefile

* resolved absolute vs. conditional assignment for variables based on reference OS

* removed example using a custom DOCKERFILE from README file

* force backend compile() to make a copy of the graph

fix copy_with_new_args on ops that have function pointers internal

update unit test for new backend API

add unit test for multiple simulataneous backends

* move get_subdevices virtual method to Manager class

* update GPU to latest

* update call methods

* add remove_compiled_function()

[Py] Change Python version for tox

* CPU: Fuse zero-padded convolution backprop filters

* CPU: Add a testcase for zero-padded convolution backprop filters fusion

* initial support for recurring matching

* fix a bug where patterns weren't populated w/ matched nodes; add recurrent tests

* add a missing newline

* address feedback

* fix function comment

* Add/update Python wrappers for nGraph operations.

- NotEqual, OneHot, Power, Sqrt, Relu, Sign, Sin, Sinh, Tan, Subtract, Select, Tanh, Sum, Reduce,
Softmax, ReplaceSlice, Reverse
- Add UT for Relu, Sign, Sin, Sinh, Sqrt, Tan, Tanh,

* Add UT for cases when Cos and Sin are giving incorrect results.

* Alphabetically sorted imports.

* Small refactoring.

- Update docstrings
- Remove unnecesary auxiliary local variable.